Makes it consistent with the "timeline create" and "timeline import"
commands, which allowed you to pass the timeline id as argument. This
also makes it unnecessary to parse the timeline ID from the output in
the python function that calls it.
The tenant ID was not actually generated here but in NeonEnvBuilder.
And the "neon_local init" command hasn't been able to generate the
initial tenant since 8712e1899e anyway.
## Problem
Having run in production for a while, we see that nodes are generally
safely oversubscribed by about a factor of 2.
## Summary of changes
Tweak the is_overloaded method to check for utililzation over 200%
rather than over 100%
## Problem
In https://github.com/neondatabase/neon/pull/8621, validation of keys
during ingest was removed because the places where we actually store
keys are now past the point where we have already converted them to
CompactKey (i128) representation.
## Summary of changes
Reinstate validation at an earlier stage in ingest. This doesn't cover
literally every place we write a key, but it covers most cases where
we're trusting postgres to give us a valid key (i.e. one that doesn't
try and use a custom spacenode).
The current code assumes that most of this functionality is
version-independent, which is only true up to v16 - PostgreSQL 17 has a
new field in CheckPoint that we need to keep track of.
This basically removes the file-level dependency on v14, and replaces it
with switches that load the correct version dependencies where required.
PR #7782 set the dependency in Cargo.toml to 'master', and locked the
version to commit that contained a specific fix, because we needed the
fix before it was included in a versioned release. The fix was later
included in parquet crate version 52.0.0, so we can now switch back to
using a released version. The latest release is 53.0.0, switch straight
to that.
---------
Co-authored-by: Conrad Ludgate <conradludgate@gmail.com>
close https://github.com/neondatabase/neon/issues/8838
## Summary of changes
This patch modifies the split delta layer writer to avoid taking
start_key and end_key when creating/finishing the layer writer. The
start_key for the delta layers will be the first key provided to the
layer writer, and the end_key would be the `last_key.next()`. This
simplifies the delta layer writer API.
On that, the layer key hack is removed. Image layers now use the full
key range, and delta layers use the first/last key provided by the user.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
I wish it worked, but it's been broken for a long time, so let's admit
defeat and remove it.
The idea of sharing the same pageserver and safekeeper environment
between tests is still sound, and it could save a lot of time in our
CI. We should perhaps put some time into doing that, but we're better
off starting from scratch than trying to make TEST_SHARED_FIXTURES
work in its current form.
I wanted to use some features from the newer version. The PR that needed
the new version is not ready yet (and might never be), but seems nice to
stay up in any case.
We modified the crate in an incompatible way and upgraded to the new
version in PR #8076. However, it was reverted in #8654. The revert
reverted the Cargo.lock reference to it, but since Cargo.toml still
points to the (tip of the) 'neon' branch, every time you make any other
unrelated changes to Cargo.toml, it also tries to update the
rust-postgres crates to the tip of the 'neon' branch again, which
doesn't work.
To fix, lock the crates to the exact commit SHA that works.
Currently using gc blocking and unblocking with storage controller
managed pageservers is painful. Implement the API on storage controller.
Fixes: #8893
For control-plane managed tenants, we have the page in the admin console
that lists all tenants on a specific pageserver. But for
storage-controller managed ones, we don't have that functionality for
now.
## Summary of changes
Adds an API that lists all shards on a given node (intention + observed)
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
One of the PRs opened by a `neondatabase` org member got labelled as
`external` because the `gh api` call failed in the wrong way:
```
Get "https://api.github.com/orgs/neondatabase/members/<username>": dial tcp 140.82.114.5:443: i/o timeout
is-member=false
```
## Summary of changes
- Check that the error message is expected before labelling PRs
- Retry `gh api` call for 10 times in case of unexpected error messages
- Add `workflow_dispatch` trigger
When walproposer observes now higher term it restarts instead of
crashing whole compute with PANIC; this avoids compute crash after
term_bump call. After successfull election we're still checking
last_log_term of the highest given vote to ensure basebackup is good,
and PANIC otherwise.
It will be used for migration per
035-safekeeper-dynamic-membership-change.md
and
https://github.com/neondatabase/docs/pull/21
ref https://github.com/neondatabase/neon/issues/8700
Check that truncation point is not from the future by comparing it with
write_record_lsn, not write_lsn, and explain that xlog switch changes
their normal order.
ref https://github.com/neondatabase/neon/issues/8911
Addresses the 1.82 beta clippy lint `too_long_first_doc_paragraph` by
adding newlines to the first sentence if it is short enough, and making
a short first sentence if there is the need.
## Problem
We want to do AZ aware scheduling, but don't have enough metadata.
## Summary of changes
Introduce a `preferred_az_id` concept for each managed tenant shard.
In a future PR, the scheduler will use this as a soft preference.
The idea is to try and keep the shard attachments within the same AZ.
Under the assumption that the compute was placed in the correct AZ,
this reduces the chances of cross AZ trafic from between compute and PS.
In terms of code changes we:
1. Add a new nullable `preferred_az_id` column to the `tenant_shards`
table. Also include an in-memory counterpart.
2. Populate the preferred az on tenant creation and shard splits.
3. Add an endpoint which allows to bulk-set preferred AZs.
(3) gives us the migration path. I'll write a script which queries the
cplane db in the region and sets the preferred az of all shards with an
active compute to the AZ of said compute. For shards without an active compute,
I'll use the AZ of the currently attached pageserver
since this is what cplane uses now to schedule computes.
## Problem
https://github.com/neondatabase/neon/pull/8852 introduced a new nullable
column for the `nodes` table: `availability_zone_id`
## Summary of changes
* Make neon local and the test suite always provide an az id
* Make the az id field in the ps registration request mandatory
* Migrate the column to non-nullable and adjust in memory state
accordingly
* Remove the code that was used to populate the az id for pre-existing
nodes
## Problem
Building on MacOS failed due to missing m4. Although a window was
popping up claiming to install m4, this was not helping.
## Summary of changes
Add instructions to install m4 using brew and link it (thanks to Folke
for helping).
Sometimes, the benchmarks fail to start up pageserver in 10s without any
obvious reason. Benchmarks run sequentially on otherwise idle runners.
Try running `sync(2)` after each bench to force a cleaner slate.
Implement this via:
- SYNC_AFTER_EACH_TEST environment variable enabled autouse fixture
- autouse fixture seems to be outermost fixture, so it works as expected
- set SYNC_AFTER_EACH_TEST=true for benchmarks in build_and_test
workflow
Evidence:
https://neon-github-public-dev.s3.amazonaws.com/reports/main/10678984691/index.html#suites/5008d72a1ba3c0d618a030a938fc035c/1210266507534c0f/
---------
Co-authored-by: Alexander Bayandin <alexander@neon.tech>
This PR simplifies the pageserver configuration parsing as follows:
* introduce the `pageserver_api::config::ConfigToml` type
* implement `Default` for `ConfigToml`
* use serde derive to do the brain-dead leg-work of processing the toml
document
* use `serde(default)` to fill in default values
* in `pageserver` crate:
* use `toml_edit` to deserialize the pageserver.toml string into a
`ConfigToml`
* `PageServerConfig::parse_and_validate` then
* consumes the `ConfigToml`
* destructures it exhaustively into its constituent fields
* constructs the `PageServerConfig`
The rules are:
* in `ConfigToml`, use `deny_unknown_fields` everywhere
* static default values go in `pageserver_api`
* if there cannot be a static default value (e.g. which default IO
engine to use, because it depends on the runtime), make the field in
`ConfigToml` an `Option`
* if runtime-augmentation of a value is needed, do that in
`parse_and_validate`
* a good example is `virtual_file_io_engine` or `l0_flush`, both of
which need to execute code to determine the effective value in
`PageServerConf`
The benefits:
* massive amount of brain-dead repetitive code can be deleted
* "unused variable" compile-time errors when removing a config value,
due to the exhaustive destructuring in `parse_and_validate`
* compile-time errors guide you when adding a new config field
Drawbacks:
* serde derive is sometimes a bit too magical
* `deny_unknown_fields` is easy to miss
Future Work / Benefits:
* make `neon_local` use `pageserver_api` to construct `ConfigToml` and
write it to `pageserver.toml`
* This provides more type safety / coompile-time errors than the current
approach.
### Refs
Fixes#3682
### Future Work
* `remote_storage` deser doesn't reject unknown fields
https://github.com/neondatabase/neon/issues/8915
* clean up `libs/pageserver_api/src/config.rs` further
* break up into multiple files, at least for tenant config
* move `models` as appropriate / refine distinction between config and
API models / be explicit about when it's the same
* use `pub(crate)` visibility on `mod defaults` to detect stale values
## Problem
A tenant may ingest a lot of data between being drained for node restart
and being moved back
in the fill phase. This is expensive and causes the fill to stall.
## Summary of changes
We make a tactical change to reduce secondary warm-up time for
migrations in fills.
Part of https://github.com/neondatabase/neon/issues/8623
## Summary of changes
It seems that we have tenants with aux policy set to v1 but don't have
any aux files in the storage. It is still safe to force migrate them
without notifying the customers. This patch adds more details to the
warning to identify the cases where we have to reach out to the users
before retiring aux v1.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
There was a confusion on the REL_14_STABLE_neon branch. PR
https://github.com/neondatabase/postgres/pull/471 was merged ot the
branch, but the corresponding PRs on the other REL_15_STABLE_neon and
REL_16_STABLE_neon branches were not merged. Also, the submodule
reference in the neon repository was never updated, so even though the
REL_14_STABLE_neon branch contained the commit, it was never used.
That PR https://github.com/neondatabase/postgres/pull/471 was a few
bricks shy of a load (no tests, some differences between the different
branches), so to get us to a good state, revert that change from the
REL_14_STABLE_neon branch. This PR in the neon repository updates the
submodule reference past two commites on the REL_14_STABLE_neon branch:
first the commit from PR
https://github.com/neondatabase/postgres/pull/471, and immediately after
that the revert of the same commit. This brings us back to square one,
but now the submodule reference matches the tip of the
REL_14_STABLE_neon branch again.
## Problem
The initial implementation of the validate API treats the in-memory
generations as authoritative.
- This is true when only one storage controller is running, but if a
rogue controller was running that hadn't been shut down properly, and
some pageserver requests were routed to that bad controller, it could
incorrectly return valid=true for stale generations.
- The generation in the main in-memory map gets out of date while a live
migration is in flight, and if the origin location for the migration
tries to do some deletions even though it is in AttachedStale (for
example because it had already started compaction), these might be
wrongly validated + executed.
## Summary of changes
- Continue to do the in-memory check: if this returns valid=false it is
sufficient to reject requests.
- When valid=true, do an additional read from the database to confirm
the generation is fresh.
- Revise behavior for validation on missing shards: this used to always
return valid=true as a convenience for deletions and shard splits, so
that pageservers weren't prevented from completing any enqueued
deletions for these shards after they're gone. However, this becomes
unsafe when we consider split brain scenarios. We could reinstate this
in future if we wanted to store some tombstones for deleted shards.
- Update test_scrubber_physical_gc to cope with the behavioral change:
they must now explicitly flush the deletion queue before splits, to
avoid tripping up on deletions that are enqueued at the time of the
split (these tests assert "scrubber deletes nothing", which check fails
if the split leaves behind some remote objects that are legitimately
GC'able)
- Add `test_storage_controller_validate_during_migration`, which uses
failpoints to create a situation where incorrect generation validation
during a live migration could result in a corruption
The rate of validate calls for tenants is pretty low: it happens as a
consequence deletions from GC and compaction, which are both
concurrency-limited on the pageserver side.
Commit cfa45ff5ee (PR #8860) updated the vendor/postgres submodules, but
didn't use the same commit SHAs that were pushed as the corresponding
REL_*_STABLE_neon branches in the postgres repository. The contents were
the same, but the REL_*_STABLE_neon branches pointed to squashed
versions of the commits, whereas the SHAs used in the submodules
referred to the pre-squash revisions.
Note: The vendor/postgres-v14 submodule still doesn't match with the tip
of REL_14_STABLE_neon branch, because there has been one more commit on
that branch since then. That's another confusion which we should fix,
but let's do that separately. This commit doesn't change the code that
gets built in any way, only changes the submodule references to point to
the correct SHAs in the REL_*_STABLE_neon branch histories, rather than
some detached commits.
We currently do not record safekeepers in the storage controller
database. We want to migrate timelines across safekeepers eventually, so
start recording the safekeepers on deploy.
Cc: #8698
## Problem
Each test might wait for up to 5s in order to HB the pageserver.
## Summary of changes
Make the heartbeat interval configurable and use a really tight one for
neon local => startup quicker
ref https://github.com/neondatabase/neon/issues/8872
## Summary of changes
We saw stuck storage scrubber in staging caused by infinite retries. I
believe here we should use `min` instead of `max` to avoid getting
minutes or hours of retry backoff.
Signed-off-by: Alex Chi Z <chi@neon.tech>
Set the field to optional, otherwise there will be decode errors when
newer version of the storage controller receives the JSON from older
version of the pageservers.
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
Currently, DatadirModification keeps a key-indexed map of all pending
writes, even though we (almost) never need to read back dirty pages for
anything other than metadata pages (e.g. relation sizes).
Related: https://github.com/neondatabase/neon/issues/6345
## Summary of changes
- commit() modifications before ingesting database creation wal records,
so that they are guaranteed to be able to get() everything they need
directly from the underlying Timeline.
- Split dirty pages in DatadirModification into pending_metadata_pages
and pending_data_pages. The data ones don't need to be in a
key-addressable format, so they just go in a Vec instead.
- Special case handling of zero-page writes in DatadirModification,
putting them in a map which is flushed on the end of a WAL record. This
handles the case where during ingest, we might first write a zero page,
and then ingest a postgres write to that page. We used to do this via
the key-indexed map of writes, but in this PR we change the data page
write path to not bother indexing these by key.
My least favorite thing about this PR is that I needed to change the
DatadirModification interface to add the on_record_end call. This is not
very invasive because there's really only one place we use it, but it
changes the object's behaviour from being clearly an aggregation of many
records to having some per-record state. I could avoid this by
implicitly doing the work when someone calls set_lsn or commit -- I'm
open to opinions on whether that's cleaner or dirtier.
## Performance
There may be some efficiency improvement here, but the primary
motivation is to enable an earlier stage of ingest to operate without
access to a Timeline. The `pending_data_pages` part is the "fast path"
bulk write data that can in principle be generated without a Timeline,
in parallel with other ingest batches, and ultimately on the safekeeper.
`test_bulk_insert` on AX102 shows approximately the same results as in
the previous PR #8591:
```
------------------------------ Benchmark results -------------------------------
test_bulk_insert[neon-release-pg16].insert: 23.577 s
test_bulk_insert[neon-release-pg16].pageserver_writes: 5,428 MB
test_bulk_insert[neon-release-pg16].peak_mem: 637 MB
test_bulk_insert[neon-release-pg16].size: 0 MB
test_bulk_insert[neon-release-pg16].data_uploaded: 1,922 MB
test_bulk_insert[neon-release-pg16].num_files_uploaded: 8
test_bulk_insert[neon-release-pg16].wal_written: 1,382 MB
test_bulk_insert[neon-release-pg16].wal_recovery: 18.264 s
test_bulk_insert[neon-release-pg16].compaction: 0.052 s
```
wal_storage.rs already checks this, but since this is a quite legit scenario
check it at safekeeper.rs (consensus level) as well.
ref https://github.com/neondatabase/neon/issues/8212
This is a take 2; previous PR #8640 had been reverted because interplay
with another change broke test_last_log_term_switch.
Endpoint implementation sends msg to manager requesting to do the
reset. Manager stops current partial backup upload task if it exists and
performs the reset.
Also slightly tweak eviction condition: all full segments before
flush_lsn must be uploaded (and committed) and there must be only one
segment left on disk (partial). This allows to evict timelines which
started not on the first segment and didn't fill the whole
segment (previous condition wasn't good because last_removed_segno was
0).
ref https://github.com/neondatabase/neon/issues/8759
## Problem
Neon local set-up does not inject an az id in `metadata.json`. See real
change in https://github.com/neondatabase/neon/pull/8852.
## Summary of changes
We piggyback on the existing `availability_zone` pageserver
configuration in order to avoid making neon local even more complex.
## Problem
Metrics event idempotency keys differ across S3 and Vector. The events
should be identical.
Resolves#8605.
## Summary of changes
Pre-generate the idempotency keys and pass the same set into both
metrics sinks.
Co-authored-by: John Spray <john@neon.tech>
Implement the timeline specific `archival_config` endpoint also in the
storage controller.
It's mostly a copy-paste of the detach handler: the task is the same: do
the same operation on all shards.
Part of #8088.
## Problem
This is a followup to #8783
- The old blocking ensure_attached function had been retained to handle
the case where a shard had a None generation_pageserver, but this wasn't
really necessary.
- There was a subtle `.1` in the code where a struct would have been
clearer
Closes#8819
## Summary of changes
- Add ShardGenerationState to represent the results of peek_generation
- Instead of calling ensure_attached when a tenant has a non-attached
shard, check the shard's policy and return 409 if it isn't Attached,
else return 503 if the shard's policy is attached but it hasn't been
reconciled yet (i.e. has a None generation_pageserver)
The test is very rudimentary, it only checks that before and after
tenant deletion, we can run `scan_metadata` for the safekeeper node
kind. Also, we don't actually expect any uploaded data, for that we
don't have enough WAL (needs to create at least one S3-uploaded file,
the scrubber doesn't recognize partial files yet).
The `scan_metadata` scrubber subcommand is extended to support either
specifying a database connection string, which was previously the only
way, and required a database to be present, or specifying the timeline
information manually via json. This is ideal for testing scenarios
because in those, the number of timelines is usually limited,
but it is involved to spin up a database just to write the timeline
information.
The pull request https://github.com/neondatabase/neon/pull/8679
explicitly mentioned that it will evict layers earlier than before.
Given that the eviction metrics is solely based on eviction threshold
(which is 86400s now), we should consider the early eviction and do not
fire alert if it's a covered layer.
## Summary of changes
Record eviction timer only when the layer is visible + accessed.
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Summary of changes
- Setting default io_buffer_alignment to 512 bytes.
- Fix places that assumed `DEFAULT_IO_BUFFER_ALIGNMENT=0`
- Adapt unit tests to handle merge with `chunk size <= 4096`.
## Testing and Performance
We have done sufficient performance de-risking.
Enabling it by default completes our correctness de-risking before the
next release.
Context: https://neondb.slack.com/archives/C07BZ38E6SD/p1725026845455259
Signed-off-by: Yuchen Liang <yuchen@neon.tech>
Co-authored-by: Christian Schwarz <christian@neon.tech>
When implementing bottom-most gc-compaction, we analyzed the structure
of layer maps that the current compaction algorithm could produce, and
decided to only support structures without delta layer overlaps and LSN
intersections with the exception of single key layers.
## Summary of changes
This patch adds the layer map valid check in the storage scrubber.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
`promote-compatibility-data` job got broken and slightly outdated after
- https://github.com/neondatabase/neon/pull/8552 -- we don't upload
artifacts for ARM64
- https://github.com/neondatabase/neon/pull/8561 -- we don't prepare
`debug` artifacts in the release branch anymore
## Summary of changes
- Promote artifacts from release PRs to the latest version (but do it
from `release` branch)
- Upload artifacts for both X64 and ARM64
## Problem
Live migration retries when it fails to notify the compute of the new
location. It should sleep between attempts.
Closes: https://github.com/neondatabase/neon/issues/8820
## Summary of changes
- Do an `exponential_backoff` in the retry loop for compute
notifications
`safekeeper::random_test test_random_schedules` debug test takes over 2
minutes to run on our arm runners. Running it 6 times with pageserver
settings seems redundant.
In #8863 I replaced the threadpool with tokio tasks, but there was a
behaviour I missed regarding cancellation. Adding the JoinHandle wrapper
that triggers abort on drop should fix this.
Another change, any panics that occur in password hashing will be
propagated through the resume_unwind functionality.
Removes additional async_trait usages from safekeeper and neon_local.
Also removes now redundant dependencies of the `async_trait` crate.
cc earlier work: #6305, #6464, #7303, #7342, #7212, #8296
It's better to reject invalid keys on the write path than storing it and
panic-ing the pageserver.
https://github.com/neondatabase/neon/issues/8636
## Summary of changes
If a key cannot be represented using i128, we don't allow writing that
key into the pageserver.
There are two versions of the check valid function: the normal one that
simply rejects i128 keys, and the stronger one that rejects all keys
that we don't support.
The current behavior when a key gets rejected is that safekeeper will
keep retrying streaming that key to the pageserver. And once such key
gets written, no new computes can be started. Therefore, there could be
a large amount of pageserver warnings if a key cannot be ingested. To
validate this behavior by yourself, the reviewer can (1) use the
stronger version of the valid check (2) run the following SQL.
```
set neon.regress_test_mode = true;
CREATE TABLESPACE regress_tblspace LOCATION '/Users/skyzh/Work/neon-test/tablespace';
CREATE SCHEMA testschema;
CREATE TABLE testschema.foo (i int) TABLESPACE regress_tblspace;
insert into testschema.foo values (1), (2), (3);
```
For now, I'd like to merge the patch with only rejecting non-i128 keys.
It's still unknown whether the stronger version covers all the cases
that basebackup doesn't support. Furthermore, the behavior of rejecting
a key will produce large amounts of warnings due to safekeeper retry.
Therefore, I'd like to reject the minimum set of keys that we don't
support (i128 ones) for now. (well, erroring out is better than panic on
`to_compact_key`)
The next step is to fix the safekeeper behavior (i.e., on such key
rejections, stop streaming WAL), so that we can properly stop writing.
An alternative solution is to simply drop these keys on the write path.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
Some tests were very slow and some tests occasionally stalled. This PR
improves some test performance and replaces the custom threadpool in
order to fix the stalling of tests.
refs https://github.com/neondatabase/neon/issues/7524
Problem
-------
When browsing Pageserver logs, background loop iterations that take a
long time are hard to spot / easy to miss because they tend to not
produce any log messages unless:
- they overrun their period, but that's only one message when the
iteration completes late
- they do something that produces logs (e.g., create image layers)
Further, a slow iteration that is still running does will not
log nor bump the metrics of `warn_when_period_overrun`until _after_
it has finished. Again, that makes a still-running iteration hard to
spot.
Solution
--------
This PR adds a wrapper around the per-tenant background loops
that, while a slow iteration is ongoing, emit a log message
every $period.
If a timeline unarchival request comes in, give an error if the parent
timeline is archived. This prevents us from the situation of having an
archived timeline with children that are not archived.
Follow up of #8824
Part of #8088
---------
Co-authored-by: Joonas Koivunen <joonas@neon.tech>
# Motivation
In https://github.com/neondatabase/neon/pull/8832 I get tokio runtime
worker stack overflow errors in debug builds.
In a similar vein, I had tokio runtimer worker stack overflow when
trying to eliminate `async_trait`
(https://github.com/neondatabase/neon/pull/8296).
The 2MiB default is kind of arbitrary - so this PR bumps it to 4MiB.
It also adds an env var to control it.
# Risk Assessment
With our 4 runtimes, the worst case stack memory usage is `4 (runtimes)
* ($num_cpus (executor threads) + 512 (blocking pool threads)) * 4MiB`.
On i3en.3xlarge, that's `8384 MiB`.
On im4gn.2xlarge, that's `8320 MiB`.
Before this change, it was half that.
Looking at production metrics, we _do_ have the headroom to accomodate
this worst case case.
# Alternatives
The problems only occur with debug builds, so technically we could only
raise the stack size for debug builds.
However, it would be another configuration where `debug != release`.
# Future Work
If we ever enable single runtime mode in prod (=>
https://github.com/neondatabase/neon/issues/7312 ) then the worst case
will drop to 25% of its current value.
Eliminating the use of `tokio::spawn_blocking` / `tokio::fs` in favor of
`tokio-epoll-uring` (=> https://github.com/neondatabase/neon/issues/7370
) would reduce the worst case to `4 (runtimes) * $num_cpus (executor
threads) * 4 MiB`.
In proxy I switched to a leaky-bucket impl using the GCRA algorithm. I
figured I could share the code with pageserver and remove the
leaky_bucket crate dependency with some very basic tokio timers and
queues for fairness.
The underlying algorithm should be fairly clear how it works from the
comments I have left in the code.
---
In benchmarking pageserver, @problame found that the new implementation
fixes a getpage throughput discontinuity in pageserver under the
`pagebench get-page-latest-lsn` benchmark with the clickbench dataset
(`test_perf_olap.py`).
The discontinuity is that for any of `--num-clients={2,3,4}`, getpage
throughput remains 10k.
With `--num-clients=5` and greater, getpage throughput then jumps to the
configured 20k rate limit.
With the changes in this PR, the discontinuity is gone, and we scale
throughput linearly to `--num-clients` until the configured rate limit.
More context in
https://github.com/neondatabase/cloud/issues/16886#issuecomment-2315257641.
closes https://github.com/neondatabase/cloud/issues/16886
---------
Co-authored-by: Joonas Koivunen <joonas@neon.tech>
Co-authored-by: Christian Schwarz <christian@neon.tech>
refs https://github.com/neondatabase/cloud/issues/13750
The logging in this commit will make it easier to detect lagging ingest.
We're trusting compute timestamps --- ideally we'd use SK timestmaps
instead.
But trusting the compute timestamp is ok for now.
## Problem
See #8620
## Summary of changes
Remove walloping of replorigin file because it is reconstructed by PS
## Checklist before requesting a review
- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
---------
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
Part of #7497
## Problem
Static computes pinned at some fix LSN could be created initially within
PITR interval but eventually go out it. To make sure that Static
computes are not affected by GC, we need to start using the LSN lease
API (introduced in #8084) in compute_ctl.
## Summary of changes
**compute_ctl**
- Spawn a thread for when a static compute starts to periodically ping
pageserver(s) to make LSN lease requests.
- Add `test_readonly_node_gc` to test if static compute can read all
pages without error.
- (test will fail on main without the code change here)
**page_service**
- `wait_or_get_last_lsn` will now allow `request_lsn` less than
`latest_gc_cutoff_lsn` to proceed if there is a lease on `request_lsn`.
Signed-off-by: Yuchen Liang <yuchen@neon.tech>
Co-authored-by: Alexey Kondratov <kondratov.aleksey@gmail.com>
Part of [Epic: Bypass PageCache for user data
blocks](https://github.com/neondatabase/neon/issues/7386).
# Problem
`InMemoryLayer` still uses the `PageCache` for all data stored in the
`VirtualFile` that underlies the `EphemeralFile`.
# Background
Before this PR, `EphemeralFile` is a fancy and (code-bloated) buffered
writer around a `VirtualFile` that supports `blob_io`.
The `InMemoryLayerInner::index` stores offsets into the `EphemeralFile`.
At those offset, we find a varint length followed by the serialized
`Value`.
Vectored reads (`get_values_reconstruct_data`) are not in fact vectored
- each `Value` that needs to be read is read sequentially.
The `will_init` bit of information which we use to early-exit the
`get_values_reconstruct_data` for a given key is stored in the
serialized `Value`, meaning we have to read & deserialize the `Value`
from the `EphemeralFile`.
The L0 flushing **also** needs to re-determine the `will_init` bit of
information, by deserializing each value during L0 flush.
# Changes
1. Store the value length and `will_init` information in the
`InMemoryLayer::index`. The `EphemeralFile` thus only needs to store the
values.
2. For `get_values_reconstruct_data`:
- Use the in-memory `index` figures out which values need to be read.
Having the `will_init` stored in the index enables us to do that.
- View the EphemeralFile as a byte array of "DIO chunks", each 512 bytes
in size (adjustable constant). A "DIO chunk" is the minimal unit that we
can read under direct IO.
- Figure out which chunks need to be read to retrieve the serialized
bytes for thes values we need to read.
- Coalesce chunk reads such that each DIO chunk is only read once to
serve all value reads that need data from that chunk.
- Merge adjacent chunk reads into larger
`EphemeralFile::read_exact_at_eof_ok` of up to 128k (adjustable
constant).
3. The new `EphemeralFile::read_exact_at_eof_ok` fills the IO buffer
from the underlying VirtualFile and/or its in-memory buffer.
4. The L0 flush code is changed to use the `index` directly, `blob_io`
5. We can remove the `ephemeral_file::page_caching` construct now.
The `get_values_reconstruct_data` changes seem like a bit overkill but
they are necessary so we issue the equivalent amount of read system
calls compared to before this PR where it was highly likely that even if
the first PageCache access was a miss, remaining reads within the same
`get_values_reconstruct_data` call from the same `EphemeralFile` page
were a hit.
The "DIO chunk" stuff is truly unnecessary for page cache bypass, but,
since we're working on [direct
IO](https://github.com/neondatabase/neon/issues/8130) and
https://github.com/neondatabase/neon/issues/8719 specifically, we need
to do _something_ like this anyways in the near future.
# Alternative Design
The original plan was to use the `vectored_blob_io` code it relies on
the invariant of Delta&Image layers that `index order == values order`.
Further, `vectored_blob_io` code's strategy for merging IOs is limited
to adjacent reads. However, with direct IO, there is another level of
merging that should be done, specifically, if multiple reads map to the
same "DIO chunk" (=alignment-requirement-sized and -aligned region of
the file), then it's "free" to read the chunk into an IO buffer and
serve the two reads from that buffer.
=> https://github.com/neondatabase/neon/issues/8719
# Testing / Performance
Correctness of the IO merging code is ensured by unit tests.
Additionally, minimal tests are added for the `EphemeralFile`
implementation and the bit-packed `InMemoryLayerIndexValue`.
Performance testing results are presented below.
All pref testing done on my M2 MacBook Pro, running a Linux VM.
It's a release build without `--features testing`.
We see definitive improvement in ingest performance microbenchmark and
an ad-hoc microbenchmark for getpage against InMemoryLayer.
```
baseline: commit 7c74112b2a origin/main
HEAD: ef1c55c52e
```
<details>
```
cargo bench --bench bench_ingest -- 'ingest 128MB/100b seq, no delta'
baseline
ingest-small-values/ingest 128MB/100b seq, no delta
time: [483.50 ms 498.73 ms 522.53 ms]
thrpt: [244.96 MiB/s 256.65 MiB/s 264.73 MiB/s]
HEAD
ingest-small-values/ingest 128MB/100b seq, no delta
time: [479.22 ms 482.92 ms 487.35 ms]
thrpt: [262.64 MiB/s 265.06 MiB/s 267.10 MiB/s]
```
</details>
We don't have a micro-benchmark for InMemoryLayer and it's quite
cumbersome to add one. So, I did manual testing in `neon_local`.
<details>
```
./target/release/neon_local stop
rm -rf .neon
./target/release/neon_local init
./target/release/neon_local start
./target/release/neon_local tenant create --set-default
./target/release/neon_local endpoint create foo
./target/release/neon_local endpoint start foo
psql 'postgresql://cloud_admin@127.0.0.1:55432/postgres'
psql (13.16 (Debian 13.16-0+deb11u1), server 15.7)
CREATE TABLE wal_test (
id SERIAL PRIMARY KEY,
data TEXT
);
DO $$
DECLARE
i INTEGER := 1;
BEGIN
WHILE i <= 500000 LOOP
INSERT INTO wal_test (data) VALUES ('data');
i := i + 1;
END LOOP;
END $$;
-- => result is one L0 from initdb and one 137M-sized ephemeral-2
DO $$
DECLARE
i INTEGER := 1;
random_id INTEGER;
random_record wal_test%ROWTYPE;
start_time TIMESTAMP := clock_timestamp();
selects_completed INTEGER := 0;
min_id INTEGER := 1; -- Minimum ID value
max_id INTEGER := 100000; -- Maximum ID value, based on your insert range
iters INTEGER := 100000000; -- Number of iterations to run
BEGIN
WHILE i <= iters LOOP
-- Generate a random ID within the known range
random_id := min_id + floor(random() * (max_id - min_id + 1))::int;
-- Select the row with the generated random ID
SELECT * INTO random_record
FROM wal_test
WHERE id = random_id;
-- Increment the select counter
selects_completed := selects_completed + 1;
-- Check if a second has passed
IF EXTRACT(EPOCH FROM clock_timestamp() - start_time) >= 1 THEN
-- Print the number of selects completed in the last second
RAISE NOTICE 'Selects completed in last second: %', selects_completed;
-- Reset counters for the next second
selects_completed := 0;
start_time := clock_timestamp();
END IF;
-- Increment the loop counter
i := i + 1;
END LOOP;
END $$;
./target/release/neon_local stop
baseline: commit 7c74112b2a origin/main
NOTICE: Selects completed in last second: 1864
NOTICE: Selects completed in last second: 1850
NOTICE: Selects completed in last second: 1851
NOTICE: Selects completed in last second: 1918
NOTICE: Selects completed in last second: 1911
NOTICE: Selects completed in last second: 1879
NOTICE: Selects completed in last second: 1858
NOTICE: Selects completed in last second: 1827
NOTICE: Selects completed in last second: 1933
ours
NOTICE: Selects completed in last second: 1915
NOTICE: Selects completed in last second: 1928
NOTICE: Selects completed in last second: 1913
NOTICE: Selects completed in last second: 1932
NOTICE: Selects completed in last second: 1846
NOTICE: Selects completed in last second: 1955
NOTICE: Selects completed in last second: 1991
NOTICE: Selects completed in last second: 1973
```
NB: the ephemeral file sizes differ by ca 1MiB, ours being 1MiB smaller.
</details>
# Rollout
This PR changes the code in-place and is not gated by a feature flag.
We get many HTTP connect timeout errors from scrubber logs, and it
turned out that the scrubber is retrying, and this is not an actual
error. In the future, we should revisit all places where we log errors
in the storage scrubber, and only error when necessary (i.e., errors
that might need manual fixing)
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
In order to build AZ aware scheduling, the storage controller needs to
know what AZ pageservers are in.
Related https://github.com/neondatabase/neon/issues/8848
## Summary of changes
This patch set adds a new nullable column to the `nodes` table:
`availability_zone_id`. The node registration
request is extended to include the AZ id (pageservers already have this
in their `metadata.json` file).
If the node is already registered, then we update the persistent and
in-memory state with the provided AZ.
Otherwise, we add the node with the AZ to begin with.
A couple assumptions are made here:
1. Pageserver AZ ids are stable
2. AZ ids do not change over time
Once all pageservers have a configured AZ, we can remove the optionals
in the code and make the database column not nullable.
Part of #8130, closes#8719.
## Problem
Currently, vectored blob io only coalesce blocks if they are immediately
adjacent to each other. When we switch to Direct IO, we need a way to
coalesce blobs that are within the dio-aligned boundary but has gap
between them.
## Summary of changes
- Introduces a `VectoredReadCoalesceMode` for `VectoredReadPlanner` and
`StreamingVectoredReadPlanner` which has two modes:
- `AdjacentOnly` (current implementation)
- `Chunked(<alignment requirement>)`
- New `ChunkedVectorBuilder` that considers batching `dio-align`-sized
read, the start and end of the vectored read will respect
`stx_dio_offset_align` / `stx_dio_mem_align` (`vectored_read.start` and
`vectored_read.blobs_at.first().start_offset` will be two different
value).
- Since we break the assumption that blobs within single `VectoredRead`
are next to each other (implicit end offset), we start to store blob end
offsets in the `VectoredRead`.
- Adapted existing tests to run in both `VectoredReadCoalesceMode`.
- The io alignment can also be live configured at runtime.
Signed-off-by: Yuchen Liang <yuchen@neon.tech>
## Problem
Storage controller upgrades (restarts, more generally) can cause
multi-second availability gaps.
While the storage controller does not sit on the main data path, it's
generally not acceptable
to block management requests for extended periods of time (e.g.
https://github.com/neondatabase/neon/issues/8034).
## Summary of changes
This RFC describes the issues around the current storage controller
restart procedure
and describes an implementation which reduces downtime to a few
milliseconds on the happy path.
Related https://github.com/neondatabase/neon/issues/7797
## Problem
When folks open github issues for feature requests, they don't have a
clear recipient: engineers usually see them during bug triage, but that
doesn't necessarily get the work prioritized.
## Summary of changes
Give end users a clearer path to submitting feedback to Neon
Protocol version 2 has been the default for a while now, and we no
longer have any computes running in production that used protocol
version 1. This completes the migration by removing support for v1 in
both the pageserver and the compute.
See issue #6211.
## Problem
Currently, we compare `neon.safekeepers` values as is, so we
unnecessarily restart walproposer even if safekeepers set didn't change.
This leads to errors like:
```log
FATAL: [WP] restarting walproposer to change safekeeper list
from safekeeper-8.us-east-2.aws.neon.tech:6401,safekeeper-11.us-east-2.aws.neon.tech:6401,safekeeper-10.us-east-2.aws.neon.tech:6401
to safekeeper-11.us-east-2.aws.neon.tech:6401,safekeeper-8.us-east-2.aws.neon.tech:6401,safekeeper-10.us-east-2.aws.neon.tech:6401
```
## Summary of changes
Split the GUC into the list of individual safekeepers and properly
compare. We could've done that somewhere on the upper level, e.g.,
control plane, but I think it's still better when the actual config
consumer is smarter and doesn't rely on upper levels.
## Problem
pg_hintplan test seems to be flaky, sometimes it fails, while usually it
passes
## Summary of changes
The regression test is changed to filter out the Neon service queries. The
expected file is changed as well.
## Checklist before requesting a review
- [x] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
Routes and their handlers were in a bit different order in 1) routes
list 2) their implementation 3) python client 4) openapi spec, making
addition of new ones intimidating. Make it the same everywhere, roughly
lexicographically but preserving some of existing logic.
No functional changes.
Part of #8002, the final big PR in the batch.
## Summary of changes
This pull request uses the new split layer writer in the gc-compaction.
* It changes how layers are split. Previously, we split layers based on
the original split point, but this creates too many layers
(test_gc_feedback has one key per layer).
* Therefore, we first verify if the layer map can be processed by the
current algorithm (See https://github.com/neondatabase/neon/pull/8191,
it's basically the same check)
* On that, we proceed with the compaction. This way, it creates a large
enough layer close to the target layer size.
* Added a new set of functions `with_discard` in the split layer writer.
This helps us skip layers if we are going to produce the same persistent
key.
* The delta writer will keep the updates of the same key in a single
file. This might create a super large layer, but we can optimize it
later.
* The split layer writer is used in the gc-compaction algorithm, and it
will split layers based on size.
* Fix the image layer summary block encoded the wrong key range.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
Co-authored-by: Arpad Müller <arpad-m@users.noreply.github.com>
Co-authored-by: Christian Schwarz <christian@neon.tech>
refs https://github.com/neondatabase/neon/issues/6989
Problem
-------
After unclean shutdown, we get restarted, start reading the local
filesystem,
and make decisions based on those reads. However, some of the data might
have
not yet been fsynced when the unclean shutdown completed.
Durability matters even though Pageservers are conceptually just a cache
of state in S3. For example:
- the cloud control plane is no control loop => pageserver responses
to tenant attachmentm, etc, needs to be durable.
- the storage controller does not rely on this (as much?)
- we don't have layer file checksumming, so, downloaded+renamed but not
fsynced layer files are technically not to be trusted
- https://github.com/neondatabase/neon/issues/2683
Solution
--------
`syncfs` the tenants directory during startup, before we start reading
from it.
This is a bit overkill because we do remove some temp files
(InMemoryLayer!)
later during startup. Further, these temp files are particularly likely
to
be dirty in the kernel page cache. However, we don't want to refactor
that
cleanup code right now, and the dirty data on pageservers is generally
not that high. Last, with [direct
IO](https://github.com/neondatabase/neon/issues/8130) we're going to
have near-zero kernel page cache anyway quite soon.
This PR:
* Implements the rule that archived timelines require all of their
children to be archived as well, as specified in the RFC. There is no
fancy locking mechanism though, so the precondition can still be broken.
As a TODO for later, we still allow unarchiving timelines with archived
parents.
* Adds an `is_archived` flag to `TimelineInfo`
* Adds timeline_archival_config to `PageserverHttpClient`
* Adds a new `test_timeline_archive` test, loosely based on
`test_timeline_delete`
Part of #8088
## Problem
We need some metric to sneak peek into how many people use inbound
logical replication (Neon is a subscriber).
## Summary of changes
This commit adds a new metric `compute_subscriptions_count`, which is
number of subscriptions grouped by enabled/disabled state.
Resolves: neondatabase/cloud#16146
Add binary for local-proxy that uses the local auth backend. Runs only
the http serverless driver support and offers config reload based on a
config file and SIGHUP
## Problem
- If a reconciler was waiting to be able to notify computes about a
change, but the control plane was waiting for the controller to finish a
timeline creation/deletion, the overall system can deadlock.
- If a tenant shard was migrated concurrently with a timeline
creation/deletion, there was a risk that the timeline operation could be
applied to a non-latest-generation location, and thereby not really be
persistent. This has never happened in practice, but would eventually
happen at scale.
Closes: #8743
## Summary of changes
- Introduce `Service::tenant_remote_mutation` helper, which looks up
shards & generations and passes them into an inner function that may do
remote I/O to pageservers. Before returning success, this helper checks
that generations haven't incremented, to guarantee that changes are
persistent.
- Convert tenant_timeline_create, tenant_timeline_delete, and
tenant_timeline_detach_ancestor to use this helper.
- These functions no longer block on ensure_attached unless the tenant
was never attached at all, so they should make progress even if we can't
complete compute notifications.
This increases the database load from timeline/create operations, but
only with cheap read transactions.
## Problem
Previously, the controller only used the shard counts for scheduling.
This works well when hosting only many-sharded tenants, but works much
less well when hosting single-sharded tenants that have a greater
deviation in size-per-shard.
Closes: https://github.com/neondatabase/neon/issues/7798
## Summary of changes
- Instead of UtilizationScore, carry the full PageserverUtilization
through into the Scheduler.
- Use the PageserverUtilization::score() instead of shard count when
ordering nodes in scheduling.
Q: Why did test_sharding_split_smoke need updating in this PR?
A: There's an interesting side effect during shard splits: because we do
not decrement the shard count in the utilization when we de-schedule the
shards from before the split, the controller will now prefer to pick
_different_ nodes for shards compared with which ones held secondaries
before the split. We could use our knowledge of splitting to fix up the
utilizations more actively in this situation, but I'm leaning toward
leaving the code simpler, as in practical systems the impact of one
shard on the utilization of a node should be fairly low (single digit
%).
In case of corrupted delta layers, we can detect the corruption and bail
out the compaction.
## Summary of changes
* Detect wrong delta desc of key range
* Detect unordered deltas
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
We have been naughty and curl-ed storcon to fix-up drains and fills.
## Summary of changes
Add support for starting/cancelling drain/fill operations via
`storcon_cli`.
close https://github.com/neondatabase/neon/issues/8579
## Summary of changes
The `is_l0` check now takes both layer key range and the layer type.
This allows us to have image layers covering the full key range in
btm-most compaction (upcoming PR). However, we still don't allow delta
layers to cover the full key range, and I will make btm-most compaction
to generate delta layers with the key range of the keys existing in the
layer instead of `Key::MIN..Key::HACK_MAX` (upcoming PR).
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
Failed / flaky tests for different arches don't have any difference in
GitHub Autocomment
## Summary of changes
- Add arch to build type for GitHub autocomment
Our new policy is to use the "rebase" method and slice all the Neon
commits into a nice patch set when doing a new major version, and use
"merge" method on minor version upgrades on the release branches.
"git merge" preserves the git history of Neon commits on the Postgres
branches. While it's nice to rebase all the Neon changes to a logical
patch set against upstream, having to do it between every minor release
is a fair amount work, and it loses the history, and is more
error-prone.
part of https://github.com/neondatabase/neon/issues/8623
We want to discover potential aux v1 customers that we might have missed
from the migrations.
## Summary of changes
Log warnings on basebackup, load timeline, and the first put_file.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
See https://neondb.slack.com/archives/C07J14D8GTX/p1724347552023709
Manipulations with LRU list in relation size cache are performed under
shared lock
## Summary of changes
Take exclusive lock
## Checklist before requesting a review
- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
This reverts commit cbe8c77997.
This change was originally made to test a hypothesis, but after that,
the proper fix#8669 was merged, so now it's not needed. Moreover, the
test is still flaky, so probably this bug was not a reason of the
flakiness.
Related to #8097
This copies a piece of code from `test_scrubber_physical_gc_ancestors`
to fix a source of flakiness: later on we rely on stuff being older than
a second, but the test can run faster under optimal conditions (as
happened to me locally, but also obvservable in
[this](https://neon-github-public-dev.s3.amazonaws.com/reports/main/10470762360/index.html#testresult/f713b02657db4b4c/retries)
allure report):
```
test_runner/regress/test_storage_scrubber.py:169: in test_scrubber_physical_gc
assert gc_summary["remote_storage_errors"] == 0
E assert 1 == 0
```
## Problem/Solution
TimelineWriter::put_batch is simply a loop over individual puts. Each
put acquires and releases locks, and checks for potentially starting a
new layer. Batching these is more efficient, but more importantly
unlocks future changes where we can pre-build serialized buffers much
earlier in the ingest process, potentially even on the safekeeper
(imagine a future model where some variant of DatadirModification lives
on the safekeeper).
Ensuring that the values in put_batch are written to one layer also
enables a simplification upstream, where we no longer need to write
values in LSN-order. This saves us a sort, but also simplifies follow-on
refactors to DatadirModification: we can store metadata keys and data
keys separately at that level without needing to zip them together in
LSN order later.
## Why?
In this PR, these changes are simplify optimizations, but they are
motivated by evolving the ingest path in the direction of disentangling
extracting DatadirModification from Timeline. It may not obvious how
right now, but the general idea is that we'll end up with three phases
of ingest:
- A) Decode walrecords and build a datadirmodification with all the
simple data contents already in a big serialized buffer ready to write
to an ephemeral layer **<-- this part can be pipelined and parallelized,
and done on a safekeeper!**
- B) Let that datadirmodification see a Timeline, so that it can also
generate all the metadata updates that require a read-modify-write of
existing pages
- C) Dump the results of B into an ephemeral layer.
Related: https://github.com/neondatabase/neon/issues/8452
## Caveats
Doing a big monolithic buffer of values to write to disk is ordinarily
an anti-pattern: we prefer nice streaming I/O. However:
- In future, when we do this first decode stage on the safekeeper, it
would be inefficient to serialize a Vec of Value, and then later
deserialize it just to add blob size headers while writing into the
ephemeral layer format. The idea is that for bulk write data, we will
serialize exactly once.
- The monolithic buffer is a stepping stone to pipelining more of this:
by seriailizing earlier (rather than at the final put_value), we will be
able to parallelize the wal decoding and bulk serialization of data page
writes.
- The ephemeral layer's buffered writer already stalls writes while it
waits to flush: so while yes we'll stall for a couple milliseconds to
write a couple megabytes, we already have stalls like this, just
distributed across smaller writes.
## Benchmarks
This PR is primarily a stepping stone to safekeeper ingest filtering,
but also provides a modest efficiency improvement to the `wal_recovery`
part of `test_bulk_ingest`.
test_bulk_ingest:
```
test_bulk_insert[neon-release-pg16].insert: 23.659 s
test_bulk_insert[neon-release-pg16].pageserver_writes: 5,428 MB
test_bulk_insert[neon-release-pg16].peak_mem: 626 MB
test_bulk_insert[neon-release-pg16].size: 0 MB
test_bulk_insert[neon-release-pg16].data_uploaded: 1,922 MB
test_bulk_insert[neon-release-pg16].num_files_uploaded: 8
test_bulk_insert[neon-release-pg16].wal_written: 1,382 MB
test_bulk_insert[neon-release-pg16].wal_recovery: 18.981 s
test_bulk_insert[neon-release-pg16].compaction: 0.055 s
vs. tip of main:
test_bulk_insert[neon-release-pg16].insert: 24.001 s
test_bulk_insert[neon-release-pg16].pageserver_writes: 5,428 MB
test_bulk_insert[neon-release-pg16].peak_mem: 604 MB
test_bulk_insert[neon-release-pg16].size: 0 MB
test_bulk_insert[neon-release-pg16].data_uploaded: 1,922 MB
test_bulk_insert[neon-release-pg16].num_files_uploaded: 8
test_bulk_insert[neon-release-pg16].wal_written: 1,382 MB
test_bulk_insert[neon-release-pg16].wal_recovery: 23.586 s
test_bulk_insert[neon-release-pg16].compaction: 0.054 s
```
close https://github.com/neondatabase/neon/issues/8558
* Directly generate image layers for sparse keyspaces during initdb
optimization.
* Support force image layer generation for sparse keyspaces.
* Fix a bug of incorrect image layer key range in case of duplicated
keys. (The added line: `start = img_range.end;`) This can cause
overlapping image layers and keys to disappear.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
After #8655, we needed to mark some tests to shut down immediately. To
aid these tests, try the new pattern of `flush_ep_to_pageserver`
followed by a non-compacting checkpoint. This moves the general graceful
shutdown problem of having too much to flush at shutdown into the test.
Also, add logging for how long the graceful shutdown took, if we got to
complete it for faster log eyeballing.
Fixes: #8712
Cc: #8715, #8708
Going through the list of recent flaky tests, trying to fix those
related to graceful shutdown.
- test_forward_compatibility: flush and wait for uploads to avoid
graceful shutdown
- test_layer_bloating: in the end the endpoint and vanilla are still up
=> immediate shutdown
- test_lagging_sk: pageserver shutdown is not related to the test =>
immediate shutdown
- test_lsn_lease_size: pageserver flushing is not needed => immediate
shutdown
Additionally:
- remove `wait_for_upload` usage from workload fixture
Cc: #8708Fixes: #8710
This removes workspace hack from all libs, not from any binaries. This
does not change the behaviour of the hack.
Running
```
cargo clean
cargo build --release --bin proxy
```
Before this change took 5m16s. After this change took 3m3s. This is
because this allows the build to be parallelisable much more.
## Problem
We want to run our regression test suite on ARM.
## Summary of changes
- run regression tests on release ARM builds
- run `build-neon` (including rust tests) on debug ARM builds
- add `arch` parameter to test to distinguish them in the allure report
and in a database
## Problem
Compaction jobs and other background loops are concurrency-limited
through a global semaphore.
The current counters allow quantifying how _many_ tasks are waiting.
But there is no way to tell how _much_ delay is added by the semaphore.
So, add a counter that aggregates the wall clock time seconds spent
acquiring the semaphore.
The metrics can be used as follows:
* retroactively calculate average acquisition time in a given time range
* compare the degree of background loop backlog among pageservers
The metric is insufficient to calculate
* run-up of ongoing acquisitions that haven't finished acquiring yet
* Not easily feasible because ["Cancelling a call to acquire makes you
lose your place in the
queue"](https://docs.rs/tokio/latest/tokio/sync/struct.Semaphore.html#method.acquire)
## Summary of changes
* Refactor the metrics to follow the current best practice for typed
metrics in `metrics.rs`.
* Add the new counter.
## Problem
We don't have a convenient way for a human to ask "how far are secondary
downloads along for this tenant".
This is useful when driving migrations of tenants to the storage
controller, as we first create a secondary location and want to see it
warm up before we cut over. That can already be done via storcon_cli,
but we would like a way that doesn't require direct API access.
## Summary of changes
Add a metric that reports to total size of layers in the heatmap: this
may be used in conjunction with the existing
`pageserver_secondary_resident_physical_size` to estimate "warmth" of
the secondary location.
## Problem
Storage controllers did not have the right token to speak to their peers
for leadership transitions.
## Summary of changes
Accept a peer jwt token for the storage controller.
Epic: https://github.com/neondatabase/cloud/issues/14701
## Problem
#8736 is getting too big. splitting off some simple changes here
## Summary of changes
Local proxy wont always be using tls, so make it optional. Local proxy
wont be using ws for now, so make it optional. Remove a dead config var.
## Problem
Previously, we would run db migrations before doing the step-down
sequence. This meant that the current leader would have to deal with
the schema changes and that's generally not safe.
## Summary of changes
Push the step-down procedure earlier in start-up and
do db migrations right after it (but before we load-up the in-memory
state from the db).
Epic: https://github.com/neondatabase/cloud/issues/14701
## Problem
The default Postgres version is set to 15 in code, while we use 16 in
most of the other places (and Postgres 17 is coming)
## Summary of changes
- Run `benchmarks` job with Postgres 16 (instead of Postgres 14)
- Set `DEFAULT_PG_VERSION` to 16 in all places
- Remove deprecated `--pg-version` pytest argument
- Update `test_metadata_bincode_serde_ensure_roundtrip` for Postgres 16
Removes the `_generic` postfix from the `GenericRemoteStorage` using
APIs, as `remote_storage` is the "default" now, and add a `_s3` postfix
to the remaining APIs using the S3 SDK (only in tenant snapshot). Also,
remove two unused functions: `list_objects_with_retries` and
`stream_tenants functions`.
Part of https://github.com/neondatabase/neon/issues/7547
Migrates most of the remaining parts of the scrubber to remote_storage:
* `pageserver_physical_gc`
* `scan_metadata` for pageservers (safekeepers were done in #8595)
* `download()` in `tenant_snapshot`. The main `tenant_snapshot` is not
migrated as it uses version history to be able to work in the face of
ongoing changes.
Part of #7547
It's been rolled out everywhere, no configs are referencing it.
All code that's made dead by the removal of the config option is removed
as part of this PR.
The `page_caching::PreWarmingWriter` in `::No` mode is equivalent to a
`size_tracking_writer`, so, use that.
part of https://github.com/neondatabase/neon/issues/7418
After the rollout has succeeded, we now set the default image
compression to be enabled.
We also remove its explicit mention from `neon_fixtures.py` added in
#8368 as it is now the default (and we switch to `zstd(1)` which is a
bit nicer on CPU time).
Part of https://github.com/neondatabase/neon/issues/5431
Part of #8128.
## Description
This PR creates a unified command to run both physical gc and metadata
health check as a cron job. This also enables us to add additional tasks
to the cron job in the future.
Signed-off-by: Yuchen Liang <yuchen@neon.tech>
## Problem
Multiple increase/decrease LFC limit may cause unlimited growth of LFC
file because punched holes while LFC shrinking are not reused when LFC
is extended.
## Summary of changes
Keep track of holes and reused them when LFC size is increased.
## Checklist before requesting a review
- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
---------
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
## Problem
Current superuser check always passes because it returns a tuple like
`(False,)`, and then the `if not superuser` passes.
## Summary of changes
Fixes the issue by unwrapping the tuple. Verified that it works against
a project where I don't have superuser.
## Problem
When a secondary location is trying to catch up while a tenant is
receiving new writes, it can become quite wasteful:
- Downloading L0s which are soon destroyed by compaction to L1s
- Downloading older layer files which are soon made irrelevant when
covered by image layers.
## Summary of changes
Sort the layer files in the heatmap:
- L0 layers are the lowest priority
- Other layers are sorted to download the highest LSNs first.
Previously, we protected from multiple ProposerElected messages from the same
walproposer with the following condition:
msg.term == self.get_last_log_term() && self.flush_lsn() >
msg.start_streaming_at
It is not exhaustive, i.e. we could still proceed to truncating WAL even though
safekeeper inserted something since the divergence point has been
calculated. While it was most likely safe because walproposer can't use
safekeeper position to commit WAL until last_log_term reaches the current
walproposer term, let's be more careful and properly calculate the divergence
point like walproposer does.
## Problem
https://github.com/neondatabase/neon/pull/8588 implemented the mechanism
for storage controller
leadership transfers. However, there's no tests that exercise the
behaviour.
## Summary of changes
1. Teach `neon_local` how to handle multiple storage controller
instances. Each storage controller
instance gets its own subdirectory (`storage_controller_1, ...`).
`storage_controller start|stop` subcommands
have also been extended to optionally accept an instance id.
2. Add a storage controller proxy test fixture. It's a basic HTTP server
that forwards requests from pageserver
and test env to the currently configured storage controller.
3. Add a test which exercises storage controller leadership transfer.
4. Finally fix a couple bugs that the test surfaced
We've had physical replication support for a long time, but we never
created an RFC for the feature. This RFC does that after the fact. Even
though we've already implemented the feature, let's have a design
discussion as if it hadn't done that. It can still be quite insightful.
This is written from a pretty compute-centric viewpoint, not much
on how it works in the control plane.
## Problem
We want to store Nightly Replication test results in the database and
notify the relevant Slack channel about failures
## Summary of changes
- Store test results in the database
- Notify `on-call-compute-staging-stream` about failures
## Problem
`secrets.GITHUB_TOKEN` (with any permissions) is not enough to get
a user's membership info if they decide to hide it.
## Summary of changes
- Use `secrets.CI_ACCESS_TOKEN` for `gh api` call
- Use `pull_request_target` instead of `pull_request` event to get
access to secrets
## Problem
Logical replication BGW checking replication lag is not reloading config
## Summary of changes
Add handling of reload config request
## Checklist before requesting a review
- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
Per #8674, disallow node configuration while drain/fill are ongoing.
Implement it by adding a only-http wrapper
`Service::external_node_configure` which checks for operation existing
before configuring.
Additionally:
- allow cancelling drain/fill after a pageserver has restarted and
transitioned to WarmingUp
Fixes: #8674
## Problem
On macOS, clippy fails with the following error:
```
error: unused import: `crate::virtual_file::owned_buffers_io::io_buf_ext::IoBufExt`
--> pageserver/src/tenant/remote_timeline_client/download.rs:26:5
|
26 | use crate::virtual_file::owned_buffers_io::io_buf_ext::IoBufExt;
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
= note: `-D unused-imports` implied by `-D warnings`
= help: to override `-D warnings` add `#[allow(unused_imports)]`
```
Introduced in https://github.com/neondatabase/neon/pull/8717
## Summary of changes
- allow `unused_imports` for
`crate::virtual_file::owned_buffers_io::io_buf_ext::IoBufExt` on macOS
in download.rs
## Problem
The control file contains the id of the safekeeper that uploaded it.
Previously, when sending a snapshot of the control file to another sk,
it would eventually be gc-ed by the receiving sk. This is incorrect
because the original sk might still need it later.
## Summary of Changes
When sending a snapshot and the control file contains an uploaded
segment:
* Create a copy of the segment in s3 with the destination sk in the
object name
* Tweak the streamed control file to point to the object create in the
previous step
Note that the snapshot endpoint now has to know the id of the requestor,
so the api has been extended to include the node if of the destination
sk.
Closes https://github.com/neondatabase/neon/issues/8542
The `tokio_epoll_uring::Slice` / `tokio_uring::Slice` type is weird.
The new `FullSlice` newtype is better. See the doc comment for details.
The naming is not ideal, but we'll clean that up in a future refactoring
where we move the `FullSlice` into `tokio_epoll_uring`. Then, we'll do
the following:
* tokio_epoll_uring::Slice is removed
* `FullSlice` becomes `tokio_epoll_uring::IoBufView`
* new type `tokio_epoll_uring::IoBufMutView` for the current
`tokio_epoll_uring::Slice<IoBufMut>`
Context
-------
I did this work in preparation for
https://github.com/neondatabase/neon/pull/8537.
There, I'm changing the type that the `inmemory_layer.rs` passes to
`DeltaLayerWriter::put_value_bytes` and thus it seemed like a good
opportunity to make this cleanup first.
## Problem
A bunch of small fixes and improvements for CI, that are too small to
have a separate PR for them
## Summary of changes
- CI(build-and-test): fix parenthesis
- CI(actionlint): fix path to workflow file
- CI: remove default args from actions/checkout
- CI: remove `gen3` label, using a couple `self-hosted` +
`small{,-arm64}`/`large{,-arm64}` is enough
- CI: prettify Slack messages, hide links behind text messages
- C(build-and-test): add more dependencies to `conclusion` job
## Problem
It's recommended that a couple of additional RUSTFLAGS be set up to
improve the performance of Rust applications on AWS Graviton.
See
57dc813626/rust.md
Note: Apple Silicon is compatible with neoverse-n1:
```
$ clang --version
Apple clang version 15.0.0 (clang-1500.3.9.4)
Target: arm64-apple-darwin23.6.0
Thread model: posix
InstalledDir: /Applications/Xcode_15.4.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin
$
$ clang --print-supported-cpus 2>&1 | grep neoverse-
neoverse-512tvb
neoverse-e1
neoverse-n1
neoverse-n2
neoverse-v1
neoverse-v2
```
## Summary of changes
- Add `-Ctarget-feature=+lse -Ctarget-cpu=neoverse-n1` to RUSTFLAGS for
ARM images
Some benchmarks and tests might still fail because of #8655 (tracked in
#8708) because we are not fast enough to shut down ([one evidence]).
Partially this is explained by the current validation mode of streaming
k-merge, but otherwise because that is where we use a lot of time in
compaction. Outside of L0 => L1 compaction, the image layer generation
is already guarded by vectored reads doing cancellation checks.
32768 is a wild guess based on looking how many keys we put in each
layer in a bench (1-2 million), but I assume it will be good enough
divisor. Doing checks more often will start showing up as contention
which we cannot currently measure. Doing checks less often might be
reasonable.
[one evidence]:
https://neon-github-public-dev.s3.amazonaws.com/reports/main/10384136483/index.html#suites/9681106e61a1222669b9d22ab136d07b/96e6d53af234924/
Earlier PR: #8706.
## Problem
During `Run rust tests` step (for debug builds), we accidentally rebuild
neon twice (by `cargo test --doc` and by `cargo nextest run`).
It happens because we don't set `cov_prefix` for the `cargo test --doc`
command, which triggers rebuilding with different build flags, and one
more rebuild by `cargo nextest run`.
## Summary of changes
- Set `cov_prefix` for `cargo test --doc` to prevent unneeded rebuilds
## Problem
This command is kind of a hack, used when we're migrating large tenants
and want to get their resident size down. It sets the tenant config to a
fixed value, which omitted heatmap_period, so caused secondaries to get
out of date.
## Summary of changes
- Set heatmap period to the same 300s default that we use elsewhere when
updating eviction settings
This is not as elegant as some general purpose partial modification of
the config, but it practically makes the command safer to use.
basic JWT implementation that caches JWKs and verifies signatures.
this code is currently not reachable from proxy, I just wanted to get
something merged in.
We can get CompactionError::Other(Cancelled) via the error handling with
a few ways.
[evidence](https://neon-github-public-dev.s3.amazonaws.com/reports/pr-8655/10301613380/index.html#suites/cae012a1e6acdd9fdd8b81541972b6ce/653a33de17802bb1/).
Hopefully fix it by:
1. replace the `map_err` which hid the
`GetReadyAncestorError::Cancelled` with `From<GetReadyAncestorError> for
GetVectoredError` conversion
2. simplifying the code in pgdatadir_mapping to eliminate the token
anyhow wrapping for deserialization errors
3. stop wrapping GetVectoredError as anyhow errors
4. stop wrapping PageReconstructError as anyhow errors
Additionally, produce warnings if we treat any other error (as was legal
before this PR) as missing key.
Cc: #8708.
## Problem
`author_association` doesn't properly work if a GitHub user decides not
to show affiliation with the org in their profile (i.e. if it's private)
## Summary of changes
- Call
`/orgs/ORG/members/USERNAME` API to check whether
a PR/issue author is a member of the org
## Problem
When pageservers do compaction, they frequently create image layers that
make earlier layers un-needed for reads, but then keep those earlier
layers around for 24 hours waiting for time-based eviction to expire
them.
Now that we track layer visibility, we can use it as an input to
eviction, and avoid the 24 hour "disk bump" that happens around
pageserver restarts.
## Summary of changes
- During time-based eviction, if a layer is marked Covered, use the
eviction period as the threshold: i.e. these layers get to remain
resident for at least one iteration of the eviction loop, but then get
evicted. With current settings this means they get evicted after 1h
instead of 24h.
- During disk usage eviction, prioritized evicting covered layers above
all other layers.
Caveats:
- Using the period as the threshold for time based eviction in this case
is a bit of a hack, but it avoids adding yet another configuration
property, and in any case the value of a new property would be somewhat
arbitrary: there's no "right" length of time to keep covered layers
around just in case.
- We had previously planned on removing time-based eviction: this change
would motivate us to keep it around, but we can still simplify the code
later to just do the eviction of covered layers, rather than applying a
TTL policy to all layers.
With additional phases from #8430 the `detach_ancestor::Error` became
untenable. Split it up into phases, and introduce laundering for
remaining `anyhow::Error` to propagate them as most often
`Error::ShuttingDown`.
Additionally, complete FIXMEs.
Cc: #6994
## Problem
The storage scrubber was reporting warnings for lots of timelines like:
```
WARN Missed some shards at count ShardCount(0) tenant_id=25eb7a83d9a2f90ac0b765b6ca84cf4c
```
These were spurious: these tenants are fine. There was a bug in
accumulating the ShardIndex for each tenant, whereby multiple timelines
would lead us to add the same ShardIndex more than one.
Closes: #8646
## Summary of changes
- Accumulate ShardIndex in a BTreeSet instead of a Vec
- Extend the test to reproduce the issue
## Problem
See https://github.com/neondatabase/neon/issues/8499
## Summary of changes
Save HEAP_COMBOCID flag in WAL and do not clear it in redo handlers.
Related Postgres PRs:
https://github.com/neondatabase/postgres/pull/457https://github.com/neondatabase/postgres/pull/458https://github.com/neondatabase/postgres/pull/459
## Checklist before requesting a review
- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
---------
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
Co-authored-by: Heikki Linnakangas <heikki@neon.tech>
With the persistent gc blocking, we can now retry reparenting timelines
which had failed for whatever reason on the previous attempt(s).
Restructure the detach_ancestor into three phases:
- prepare (insert persistent gc blocking, copy lsn prefix, layers)
- detach and reparent
- reparenting can fail, so we might need to retry this portion
- complete (remove persistent gc blocking)
Cc: #6994
After #8655 we've had a few issues (mostly tracked on #8708) with the
graceful shutdown. In order to shutdown more of the processes and catch
more errors, for example, from all pageservers, do an immediate shutdown
for those nodes which fail the initial (possibly graceful) shutdown.
Cc: #6485
## Problem
We use a set of **Neon** reuse databases in benchmarking.yml which are
still using pg14.
Because we want to compare apples to apples and have migrated the AWS
reuse clusters to pg16 we should also use pg16 for Neon.
## Summary of changes
- Automatically restore the test databases for Neon project
A few of the benchmarks have started failing after #8655 where they are
waiting for compactor task. Reads done by image layer creation should
already be cancellation sensitive because vectored get does a check each
time, but try sprinkling additional cancellation points to:
- each partition
- after each vectored read batch
It seems that some benchmarks are failing because they are simply not
stopping to ingest wal on shutdown. It might mean that the tests were
never ran on a stable pageserver situation and WAL has always been left
to be ingested on safekeepers, but let's see if this silences the
failures and "stops the bleeding".
Cc: https://github.com/neondatabase/neon/issues/8712
## Problem
We want to mark new PRs and issues created by external users
## Summary of changes
- Add a new workflow which adds `external` label for issues and PRs
created by external users
## Problem
When the utilization API was added, it was just a stub with disk space
information.
Disk space information isn't a very good metric for assigning tenants to
pageservers, because pageservers making full use of their disks would
always just have 85% utilization, irrespective of how much pressure they
had for disk space.
## Summary of changes
- Use the new layer visibiilty metric to calculate a "wanted size" per
tenant, and sum these to get a total local disk space wanted per
pageserver. This acts as the primary signal for utilization.
- Also use the shard count to calculate a utilization score, and take
the max of this and the disk-driven utilization. The shard count limit
is currently set as a constant 20,000, which matches contemporary
operational practices when loading pageservers.
The shard count limit means that for tiny/empty tenants, on a machine
with 3.84TB disk, each tiny tenant influences the utilization score as
if it had size 160MB.
## Problem
When pooled connections are used, session semantic its not preserved,
including GUC settings.
Many customers have particular problem with setting search_path.
But pgbouncer 1.20 has `track_extra_parameters` settings which allows to
track parameters included in startup package which are reported by
Postgres. Postgres has [an official list of parameters that it reports
to the
client](https://www.postgresql.org/docs/15/protocol-flow.html#PROTOCOL-ASYNC).
This PR makes Postgres also report `search_path` and so allows to
include it in `track_extra_parameters`.
## Summary of changes
Set GUC_REPORT flag for `search_path`.
## Checklist before requesting a review
- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
---------
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
## Problem
Pageserver exposes some vectored get related configs which are not in
use.
## Summary of changes
Remove the following pageserver configs: get_impl, get_vectored_impl,
and `validate_get_vectored`.
They are not used in the pageserver since
https://github.com/neondatabase/neon/pull/8601.
Manual overrides have been removed from the aws repo in
https://github.com/neondatabase/aws/pull/1664.
## Problem
This follows a PR that insists all input keys are representable in 16
bytes:
- https://github.com/neondatabase/neon/pull/8648
& a PR that prevents postgres from sending us keys that use the high
bits of field2:
- https://github.com/neondatabase/neon/pull/8657
Motivation for this change:
1. Ingest is bottlenecked on CPU
2. InMemoryLayer can create huge (~1M value) BTreeMap<Key,_> for its
index.
3. Maps over i128 are much faster than maps over an arbitrary 18 byte
struct.
It may still be worthwhile to make the index two-tier to optimize for
the case where only the last 4 bytes (blkno) of the key vary frequently,
but simply using the i128 representation of keys has a big impact for
very little effort.
Related: #8452
## Summary of changes
- Introduce `CompactKey` type which contains an i128
- Use this instead of Key in InMemoryLayer's index, converting back and
forth as needed.
## Performance
All the small-value `bench_ingest` cases show improved throughput.
The one that exercises this index most directly shows a 35% throughput
increase:
```
ingest-small-values/ingest 128MB/100b seq, no delta
time: [374.29 ms 378.56 ms 383.38 ms]
thrpt: [333.88 MiB/s 338.13 MiB/s 341.98 MiB/s]
change:
time: [-26.993% -26.117% -25.111%] (p = 0.00 < 0.05)
thrpt: [+33.531% +35.349% +36.974%]
Performance has improved.
```
## Problem
We use infrastructure as code (TF) to deploy AWS Aurora and AWS RDS
Postgres database clusters.
Whenever we have a change in TF (e.g. **every year** to upgrade to a
higher Postgres version or when we change the cluster configuration) TF
will apply the change and create a new AWS database cluster.
However our benchmarking testcase also expects databases in these
clusters and tables loaded with data.
So we add auto-detection - if the AWS RDS instances are "empty" we
create the necessary databases and restore a pg_dump.
**Important Notes:**
- These steps are NOT run in each benchmarking run, but only after a new
RDS instance has been deployed.
- the benchmarking workflows use GitHub secrets to find the connection
string for the database. These secrets still need to be (manually or
programmatically using git cli) updated if some port of the connection
string (e.g. user, password or hostname) changes.
## Summary of changes
In each benchmarking run check if
- database has already been created - if not create it
- database has already been restored - if not restore it
Supported databases
- tpch
- clickbench
- user example
Supported platforms:
- AWS RDS Postgres
- AWS Aurora serverless Postgres
Sample workflow run - but this one uses Neon database to test the
restore step and not real AWS databases
https://github.com/neondatabase/neon/actions/runs/10321441086/job/28574350581
Sample workflow run - with real AWS database clusters
https://github.com/neondatabase/neon/actions/runs/10346816389/job/28635997653
Verification in second run - with real AWS database clusters - that
second time the restore is skipped
https://github.com/neondatabase/neon/actions/runs/10348469517/job/28640778223
## Problem
Storage controller restarts cause temporary unavailability from the
control plane POV. See RFC for more details.
## Summary of changes
* A couple of small refactors of the storage controller start-up
sequence to make extending it easier.
* A leader table is added to track the storage controller instance
that's currently the leader (if any)
* A peer client is added such that storage controllers can send
`step_down` requests to each other (implemented in
https://github.com/neondatabase/neon/pull/8512).
* Implement the leader cut-over as described in the RFC
* Add `start-as-candidate` flag to the storage controller to gate the
rolling restart behaviour. When the flag is `false` (the default), the
only change from the current start-up sequence is persisting the leader
entry to the database.
It should give us all possible allowed_errors more consistently.
While getting the workflows to pass on
https://github.com/neondatabase/neon/pull/8632 it was noticed that
allowed_errors are rarely hit (1/4). This made me realize that we always
do an immediate stop by default. Doing a graceful shutdown would had
made the draining more apparent and likely we would not have needed the
#8632 hotfix.
Downside of doing this is that we will see more timeouts if tests are
randomly leaving pause failpoints which fail the shutdown.
The net outcome should however be positive, we could even detect too
slow shutdowns caused by a bug or deadlock.
## Problem
This test was disabled.
## Summary of changes
- Remove the skip marker.
- Explicitly avoid doing compaction & gc during checkpoints (the default
scale doesn't do anything here, but when experimeting with larger scales
it messes things up)
- Set a data size that gives a ~20s runtime on a Hetzner dev machine,
previous one gave very noisy results because it was so small
For reference on a Hetzner AX102:
```
------------------------------ Benchmark results -------------------------------
test_bulk_insert[neon-release-pg16].insert: 25.664 s
test_bulk_insert[neon-release-pg16].pageserver_writes: 5,428 MB
test_bulk_insert[neon-release-pg16].peak_mem: 577 MB
test_bulk_insert[neon-release-pg16].size: 0 MB
test_bulk_insert[neon-release-pg16].data_uploaded: 1,922 MB
test_bulk_insert[neon-release-pg16].num_files_uploaded: 8
test_bulk_insert[neon-release-pg16].wal_written: 1,382 MB
test_bulk_insert[neon-release-pg16].wal_recovery: 25.373 s
test_bulk_insert[neon-release-pg16].compaction: 0.035 s
```
It should take syncrep flush_lsn into account because WAL before it on endpoint
restart is lost, which makes replication miss some data if slot had already been
advanced too far. This commit adds test reproducing the issue and bumps
vendor/postgres to commit with the actual fix.
## Problem
In several workflows, we have repeating code which is separated into
two steps:
```bash
mkdir -p $(pwd)/.docker-custom
echo DOCKER_CONFIG=/tmp/.docker-custom >> $GITHUB_ENV
...
rm -rf $(pwd)/.docker-custom
```
Such copy-paste is prone to errors; for example, in one case, instead of
`$(pwd)/.docker-custom`, we use `/tmp/.docker-custom`, which is shared
between workflows.
## Summary of changes
- Create a new action `actions/set-docker-config-dir`, which sets
`DOCKER_CONFIG` and deletes it in a Post action part
Noticed this while debugging a test failure in #8673 which only occurs
with real S3 instead of mock S3: if you authenticate to S3 via
`AWS_PROFILE`, then it requires the `HOME` env var to be set so that it
can read inside the `~/.aws` directory.
The scrubber abstraction `StorageScrubber::scrubber_cli` in
`neon_fixtures.py` would otherwise not work. My earlier PR #6556 has
done similar things for the `neon_local` wrapper.
You can try:
```
aws sso login --profile dev
export ENABLE_REAL_S3_REMOTE_STORAGE=y REMOTE_STORAGE_S3_BUCKET=neon-github-ci-tests REMOTE_STORAGE_S3_REGION=eu-central-1 AWS_PROFILE=dev
RUST_BACKTRACE=1 BUILD_TYPE=debug DEFAULT_PG_VERSION=16 ./scripts/pytest -vv --tb=short -k test_scrubber_tenant_snapshot
```
before and after this patch: this patch fixes it.
## Problem
This page had many dead links, and was confusing for folks looking for
documentation about our product.
Closes: https://github.com/neondatabase/neon/issues/8535
## Summary of changes
- Add a link to the product docs up top
- Remove dead/placeholder links
## Problem
We install and try to use `cachepot`. But it is not configured correctly
and doesn't work (after https://github.com/neondatabase/neon/pull/2290)
## Summary of changes
- Remove `cachepot`
## Problem
Migrations of tenant shards with cold secondaries are holding up drains
in during production deployments.
## Summary of changes
If a secondary locations is lagging by more than 256MiB (configurable,
but that's the default), then skip cutting it over to the secondary as part of the node drain.
## Problem
This type of error can happen during shutdown & was triggering a circuit
breaker alert.
## Summary of changes
- Map NotIntialized::Stopped to CompactionError::ShuttingDown, so that
we may handle it cleanly
## Problem
Azure login fails in `pin-build-tools-image` workflow because the job
doesn't have the required permissions.
```
Error: Please make sure to give write permissions to id-token in the workflow.
Error: Login failed with Error: Error message: Unable to get ACTIONS_ID_TOKEN_REQUEST_URL env variable. Double check if the 'auth-type' is correct. Refer to https://github.com/Azure/login#readme for more information.
```
## Summary of changes
- Add `id-token: write` permission to `pin-build-tools-image`
- Add an input to force image tagging
- Unify pushing to Docker Hub with other registries
- Split the job into two to have less if's
part of https://github.com/neondatabase/neon/issues/8653
Disable create tablespace stmt. It turns out it requires much less
effort to do the regress test mode flag than patching the test cases,
and given that we might need to support tablespaces in the future, I
decided to add a new flag `regress_test_mode` to change the behavior of
create tablespace.
Tested manually that without setting regress_test_mode, create
tablespace will be rejected.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
Co-authored-by: Heikki Linnakangas <heikki@neon.tech>
This reverts #8076 - which was already reverted from the release branch
since forever (it would have been a breaking change to release for all
users who currently set TimeZone options). It's causing conflicts now so
we should revert it here as well.
## Problem
Latency from one cloud provider to another one is higher than within the
same cloud provider.
Some of our benchmarks are latency sensitive - we run a pgbench or psql
in the github action runner and the system under test is running in Neon
(database project).
For realistic perf tps and latency results we need to compare apples to
apples and run the database client in the same "latency distance" for
all tests.
## Summary of changes
Move job steps that test Neon databases deployed on Azure into Azure
action runners.
- bench strategy variant using azure database
- pgvector strategy variant using azure database
- pgbench-compare strategy variants using azure database
## Test run
https://github.com/neondatabase/neon/actions/runs/10314848502
## Problem
We're adding more third party dependencies to support more diverse +
realistic test cases in `test_runner/logical_repl`. I ❤️ these
tests, they are a good thing.
The slight glitch is that python packaging is hard, and some third party
python packages have issues. For example the current kafka dependency
doesn't work on latest python. We can mitigate that by only importing
these more specialized dependencies in the tests that use them.
## Summary of changes
- Move the `kafka` import into a test body, so that folks running the
regular `test_runner/regress` tests don't have to have a working kafka
client package.
## Problem
This code was to mitigate risk in
https://github.com/neondatabase/neon/pull/8427
As expected, we did not hit this code path - the new continuous updates
of gc_info are working fine, we can remove this code now.
## Summary of changes
- Remove block that double-checks retain_lsns
avoid "leaking" the completions of BackgroundPurges by:
1. switching it to TaskTracker for provided close+wait
2. stop using tokio::fs::remove_dir_all which will consume two units of
memory instead of one blocking task
Additionally, use more graceful shutdown in tests which do actually some
background cleanup.
## Problem
See
https://neondb.slack.com/archives/C03QLRH7PPD/p1723038557449239?thread_ts=1722868375.476789&cid=C03QLRH7PPD
Logical replication subscription by default use `synchronous_commit=off`
which cause problems with safekeeper
## Summary of changes
Set `synchronous_commit=on` for logical replication subscription in
test_subscriber_restart.py
## Checklist before requesting a review
- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
---------
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
Co-authored-by: Heikki Linnakangas <heikki@neon.tech>
## Problem
Some developers build on MacOS, which doesn't have io_uring.
## Summary of changes
- Add `io_engine_for_bench`, which on linux will give io_uring or panic
if it's unavailable, and on MacOS will always panic.
We do not want to run such benchmarks with StdFs: the results aren't
interesting, and will actively waste the time of any developers who
start investigating performance before they realize they're using a
known-slow I/O backend.
Why not just conditionally compile this benchmark on linux only? Because
even on linux, I still want it to refuse to run if it can't get
io_uring.
Part of #8130, [RFC: Direct IO For Pageserver](https://github.com/neondatabase/neon/blob/problame/direct-io-rfc/docs/rfcs/034-direct-io-for-pageserver.md)
## Description
Add pageserver config for evaluating/enabling direct I/O.
- Disabled: current default, uses buffered io as is.
- Evaluate: still uses buffered io, but could do alignment checking and
perf simulation (pad latency by direct io RW to a fake file).
- Enabled: uses direct io, behavior on alignment error is configurable.
Signed-off-by: Yuchen Liang <yuchen@neon.tech>
We've noticed increased memory usage with the latest release. Drain the
joinset of `page_service` connection handlers to avoid leaking them
until shutdown. An alternative would be to use a TaskTracker.
TaskTracker was not discussed in original PR #8339 review, so not hot
fixing it in here either.
Earlier I was thinking we'd need a (ancestor_lsn, timeline_id) ordered
list of reparented. Turns out we did not need it at all. Replace it with
an unordered hashset. Additionally refactor the reparented direct
children query out, it will later be used from more places.
Split off from #8430.
Cc: #6994
Ephemeral files cleanup on drop but did not delay shutdown, leading to
problems with restarting the tenant. The solution is as proposed:
- make ephemeral files carry the gate guard to delay `Timeline::gate`
closing
- flush in-memory layers and strong references to those on
`Timeline::shutdown`
The above are realized by making LayerManager an `enum` with `Open` and
`Closed` variants, and fail requests to modify `LayerMap`.
Additionally:
- fix too eager anyhow conversions in compaction
- unify how we freeze layers and handle errors
- optimize likely_resident_layers to read LayerFileManager hashmap
values instead of bouncing through LayerMap
Fixes: #7830
## Problem
1. Hard to correlate startup parameters with the endpoint that provided
them.
2. Some configurations are not needed in the `ProxyConfig` struct.
## Summary of changes
Because of some borrow checker fun, I needed to switch to an
interior-mutability implementation of our `RequestMonitoring` context
system. Using https://docs.rs/try-lock/latest/try_lock/ as a cheap lock
for such a use-case (needed to be thread safe).
Removed the lock of each startup message, instead just logging only the
startup params in a successful handshake.
Also removed from values from `ProxyConfig` and kept as arguments.
(needed for local-proxy config)
Timeline cancellation running in parallel with gc yields error log lines
like:
```
Gc failed 1 times, retrying in 2s: TimelineCancelled
```
They are completely harmless though and normal to occur. Therefore, only
print those messages at an info level. Still print them at all so that
we know what is going on if we focus on a single timeline.
Part of #8128.
## Problem
Currently, scrubber `scan_metadata` command will return with an error
code if the metadata on remote storage is corrupted with fatal errors.
To safely deploy this command in a cronjob, we want to differentiate
between failures while running scrubber command and the erroneous
metadata. At the same time, we also want our regression tests to catch
corrupted metadata using the scrubber command.
## Summary of changes
- Return with error code only when the scrubber command fails
- Uses explicit checks on errors and warnings to determine metadata
health in regression tests.
**Resolve conflict with `tenant-snapshot` command (after shard split):**
[`test_scrubber_tenant_snapshot`](https://github.com/neondatabase/neon/blob/yuchen/scrubber-scan-cleanup-before-prod/test_runner/regress/test_storage_scrubber.py#L23)
failed before applying 422a8443dd
- When taking a snapshot, the old `index_part.json` in the unsharded
tenant directory is not kept.
- The current `list_timeline_blobs` implementation consider no
`index_part.json` as a parse error.
- During the scan, we are only analyzing shards with highest shard
count, so we will not get a parse error. but we do need to add the
layers to tenant object listing, otherwise we will get index is
referencing a layer that is not in remote storage error.
- **Action:** Add s3_layers from `list_timeline_blobs` regardless of
parsing error
Signed-off-by: Yuchen Liang <yuchen@neon.tech>
## Problem
We lack a rust bench for the inmemory layer and delta layer write paths:
it is useful to benchmark these components independent of postgres & WAL
decoding.
Related: https://github.com/neondatabase/neon/issues/8452
## Summary of changes
- Refactor DeltaLayerWriter to avoid carrying a Timeline, so that it can
be cleanly tested + benched without a Tenant/Timeline test harness. It
only needed the Timeline for building `Layer`, so this can be done in a
separate step.
- Add `bench_ingest`, which exercises a variety of workload "shapes"
(big values, small values, sequential keys, random keys)
- Include a small uncontroversial optimization: in `freeze`, only
exhaustively walk values to assert ordering relative to end_lsn in debug
mode.
These benches are limited by drive performance on a lot of machines, but
still useful as a local tool for iterating on CPU/memory improvements
around this code path.
Anecdotal measurements on Hetzner AX102 (Ryzen 7950xd):
```
ingest-small-values/ingest 128MB/100b seq
time: [1.1160 s 1.1230 s 1.1289 s]
thrpt: [113.38 MiB/s 113.98 MiB/s 114.70 MiB/s]
Found 1 outliers among 10 measurements (10.00%)
1 (10.00%) low mild
Benchmarking ingest-small-values/ingest 128MB/100b rand: Warming up for 3.0000 s
Warning: Unable to complete 10 samples in 10.0s. You may wish to increase target time to 18.9s.
ingest-small-values/ingest 128MB/100b rand
time: [1.9001 s 1.9056 s 1.9110 s]
thrpt: [66.982 MiB/s 67.171 MiB/s 67.365 MiB/s]
Benchmarking ingest-small-values/ingest 128MB/100b rand-1024keys: Warming up for 3.0000 s
Warning: Unable to complete 10 samples in 10.0s. You may wish to increase target time to 11.0s.
ingest-small-values/ingest 128MB/100b rand-1024keys
time: [1.0715 s 1.0828 s 1.0937 s]
thrpt: [117.04 MiB/s 118.21 MiB/s 119.46 MiB/s]
ingest-small-values/ingest 128MB/100b seq, no delta
time: [425.49 ms 429.07 ms 432.04 ms]
thrpt: [296.27 MiB/s 298.32 MiB/s 300.83 MiB/s]
Found 1 outliers among 10 measurements (10.00%)
1 (10.00%) low mild
ingest-big-values/ingest 128MB/8k seq
time: [373.03 ms 375.84 ms 379.17 ms]
thrpt: [337.58 MiB/s 340.57 MiB/s 343.13 MiB/s]
Found 1 outliers among 10 measurements (10.00%)
1 (10.00%) high mild
ingest-big-values/ingest 128MB/8k seq, no delta
time: [81.534 ms 82.811 ms 83.364 ms]
thrpt: [1.4994 GiB/s 1.5095 GiB/s 1.5331 GiB/s]
Found 1 outliers among 10 measurements (10.00%)
```
## Problem
Sometimes, a layer is Covered by hasn't yet been evicted from local disk
(e.g. shortly after image layer generation). It is not good use of
resources to download these to a secondary location, as there's a good
chance they will never be read.
This follows the previous change that added layer visibility:
- #8511
Part of epic:
- https://github.com/neondatabase/neon/issues/8398
## Summary of changes
- When generating heatmaps, only include Visible layers
- Update test_secondary_downloads to filter to visible layers when
listing layers from an attached location
## Problem
In staging, we could see that occasionally tenants were wrapping their
pageserver_visible_physical_size metric past zero to 2^64.
This is harmless right now, but will matter more later when we start
using visible size in things like the /utilization endpoint.
## Summary of changes
- Add debug asserts that detect this case. `test_gc_of_remote_layers`
works as a reproducer for this issue once the asserts are added.
- Tighten up the interface around access_stats so that only Layer can
mutate it.
- In Layer, wrap calls to `record_access` in code that will update the
visible size statistic if the access implicitly marks the layer visible
(this was what caused the bug)
- In LayerManager::rewrite_layers, use the proper set_visibility layer
function instead of directly using access_stats (this is an additional
path where metrics could go bad.)
- Removed unused instances of LayerAccessStats in DeltaLayer and
ImageLayer which I noticed while reviewing the code paths that call
record_access.
## Problem
The controller scale test does random migrations. These mutate secondary
locations, and therefore can cause secondary optimizations to happen in
the background, violating the test's expectation that consistency_check
will work as there are no reconciliations running.
Example:
https://neon-github-public-dev.s3.amazonaws.com/reports/main/10247161379/index.html#suites/07874de07c4a1c9effe0d92da7755ebf/6316beacd3fb3060/
## Summary of changes
- Only migrate to existing secondary locations, not randomly picked
nodes, so that we can do a fast reconcile_until_idle (otherwise
reconcile_until_idle is takes a long time to create new secondary
locations).
- Do a reconcile_until_idle before consistency_check.
## Problem
We need to test the logical replication with some external consumers.
## Summary of changes
A test of the logical replication with Debezium as a consumer was added.
---------
Co-authored-by: Alexander Bayandin <alexander@neon.tech>
We don't use it for packaging, and 'poetry install' will soon error
otherwise. Also remove name and version fields as these are not required for
non-packaging mode.
#8600 missed the hunk changing index_part.json informative version.
Include it in this PR, in addition add more non-warning index_part.json
versions to scrubber.
## Problem
We have been maintaining two read paths (legacy and vectored) for a
while now. The legacy read-path was only used for cross validation in some tests.
## Summary of changes
* Tweak all tests that were using the legacy read path to use the
vectored read path instead
* Remove the read path dispatching based on the pageserver configs
* Remove the legacy read path code
We will be able to remove the single blob io code in
`pageserver/src/tenant/blob_io.rs` when https://github.com/neondatabase/neon/issues/7386 is complete.
Closes https://github.com/neondatabase/neon/issues/8005
Currently, we do not have facilities to persistently block GC on a
tenant for whatever reason. We could do a tenant configuration update,
but that is risky for generation numbers and would also be transient.
Introduce a `gc_block` facility in the tenant, which manages per
timeline blocking reasons.
Additionally, add HTTP endpoints for enabling/disabling manual gc
blocking for a specific timeline. For debugging, individual tenant
status now includes a similar string representation logged when GC is
skipped.
Cc: #6994
Add dry-run mode that does not produce any image layer + delta layer. I
will use this code to do some experiments and see how much space we can
reclaim for tenants on staging. Part of
https://github.com/neondatabase/neon/issues/8002
* Add dry-run mode that runs the full compaction process without
updating the layer map. (We never call finish on the writers and the
files will be removed before exiting the function).
* Add compaction statistics and print them at the end of compaction.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
> Currently, long-running LR tests recreate endpoints every night. We'd
like to have along-running buildup of history to exercise the pageserver
in this case (instead of "unit-testing" the same behavior everynight).
Closes#8317
## Summary of changes
- Update Postgres version for replication tests
- Set `BENCHMARK_PROJECT_ID_PUB`/`BENCHMARK_PROJECT_ID_SUB` env vars to
projects that were created for this purpose
---------
Co-authored-by: Sasha Krassovsky <krassovskysasha@gmail.com>
Currently if `GET
/v1/tenant/x/timeline/y?force-await-initial-logical-size=true` is
requested for a root timeline created within the current pageserver
session, the request handler panics hitting the debug assertion. These
timelines will always have an accurate (at initdb import) calculated
logical size. Fix is to never attempt prioritizing timeline size
calculation if we already have an exact value.
Split off from #8528.
## Problem
In some cases, a deadlock between `build-and-test` and
`trigger-e2e-tests` workflows can happen:
```
Build and Test
Canceling since a deadlock for concurrency group 'Build and Test-8600/merge-anysha' was detected between 'top level workflow' and 'trigger-e2e-tests'
```
I don't understand the reason completely, probably `${{ github.workflow
}}` got evaluated to the same value and somehow caused the issue.
We don't need to limit concurrency for `trigger-e2e-tests`
workflow.
See
https://neondb.slack.com/archives/C059ZC138NR/p1722869486708179?thread_ts=1722869027.960029&cid=C059ZC138NR
## Problem
We don't trigger e2e tests for draft PRs, but we do trigger them once a
PR is in the "Ready for review" state.
Sometimes, a PR can be marked as "Ready for review" before we finish
image building. In such cases, triggering e2e tests fails.
## Summary of changes
- Make `trigger-e2e-tests` job poll status of `promote-images` job from
the build-and-test workflow for the last commit. And trigger only if the
status is `success`
- Remove explicit image checking from the workflow
- Add `concurrency` for `triggere-e2e-tests` workflow to make it
possible to cancel jobs in progress (if PR moves from "Draft" to "Ready
for review" several times in a row)
## Problem
PR #7992 was merged without correspondent changes in Postgres submodules
and this is why test_oid_overflow.py is failed now.
## Summary of changes
Bump Postgres versions
## Checklist before requesting a review
- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
## Problem
There is an unused safekeeper option `partial_backup_enabled`.
`partial_backup_enabled` was implemented in #6530, but this option was
always turned into enabled in #8022.
If you intended to keep this option for a specific reason, I will close
this PR.
## Summary of changes
I removed an unused safekeeper option `partial_backup_enabled`.
part of https://github.com/neondatabase/neon/issues/8002
## Summary of changes
Add a `SplitImageWriter` that automatically splits image layer based on
estimated target image layer size. This does not consider compression
and we might need a better metrics.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
Co-authored-by: Arpad Müller <arpad-m@users.noreply.github.com>
We need both compaction and gc lock for gc-compaction. The lock order
should be the same everywhere, otherwise there could be a deadlock where
A waits for B and B waits for A.
We also had a double-lock issue. The compaction lock gets acquired in
the outer `compact` function. Note that the unit tests directly call
`compact_with_gc`, and therefore not triggering the issue.
## Summary of changes
Ensure all places acquire compact lock and then gc lock. Remove an extra
compact lock acqusition.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
Currently, our backward compatibility tests only look one release back.
That means, for example, that when we switch on image layer compression
by default, we'll test reading of uncompressed layers for one release,
and then stop doing it. When we make an index_part.json format change,
we'll test against the old format for a week, then stop (unless we write
separate unit tests for each old format).
The reality in the field is that data in old formats will continue to
exist for weeks/months/years. When we make major format changes, we
should retain examples of the old format data, and continuously verify
that the latest code can still read them.
This test uses contents from a new path in the public S3 bucket,
`compatibility-data-snapshots/`. It is populated by hand. The first
important artifact is one from before we switch on compression, so that
we will keep testing reads of uncompressed data. We will generate more
artifacts ahead of other key changes, like when we update remote storage
format for archival timelines.
Closes: https://github.com/neondatabase/cloud/issues/15576
This commit tries to fix regular load spikes on staging, caused by too
many eviction and partial upload operations running at the same time.
Usually it was hapenning after restart, for partial backup the load was
delayed.
- Add a semaphore for evictions (2 permits by default)
- Rename `resident_since` to `evict_not_before` and smooth out the curve
by using random duration
- Use random duration in partial uploads as well
related to https://github.com/neondatabase/neon/issues/6338
some discussion in
https://neondb.slack.com/archives/C033RQ5SPDH/p1720601531744029
Makes `flush_frozen_layer` add a barrier to the upload queue and makes
it wait for that barrier to be reached until it lets the flushing be
completed.
This gives us backpressure and ensures that writes can't build up in an
unbounded fashion.
Fixes#7317
Chaos injection bridges the gap between automated testing (where we do
lots of different things with small, short-lived tenants), and staging
(where we do many fewer things, but with larger, long-lived tenants).
This PR adds a first type of chaos which isn't really very chaotic: it's
live migration of tenants between healthy pageservers. This nevertheless
provides continuous checks that things like clean, prompt shutdown of
tenants works for realistically deployed pageservers with realistically
large tenants.
## Problem
Previously, when we do a timeline deletion, shards will delete layers
that belong to an ancestor. That is not a correctness issue, because
when we delete a timeline, we're always deleting it from all shards, and
destroying data for that timeline is clearly fine.
However, there exists a race where one shard might start doing this
deletion while another shard has not yet received the deletion request,
and might try to access an ancestral layer. This creates ambiguity over
the "all layers referenced by my index should always exist" invariant,
which is important to detecting and reporting corruption.
Now that we have a GC mode for clearing up ancestral layers, we can rely
on that to clean up such layers, and avoid deleting them right away.
This makes things easier to reason about: there are now no cases where a
shard will delete a layer that belongs to a ShardIndex other than
itself.
## Summary of changes
- Modify behavior of RemoteTimelineClient::delete_all
- Add `test_scrubber_physical_gc_timeline_deletion` to exercise this
case
- Tweak AWS SDK config in the scrubber to enable retries. Motivated by
seeing the test for this feature encounter some transient "service
error" S3 errors (which are probably nothing to do with the changes in
this PR)
## Problem
`allure_attach_from_dir` method might create `tar.zst` archives even
if `--alluredir` is not set (i.e. Allure results collection is disabled)
## Summary of changes
- Don't run `allure_attach_from_dir` if `--alluredir` is not set
part of https://github.com/neondatabase/neon/issues/8002
Due to the limitation of the current layer map implementation, we cannot
directly replace a layer. It's interpreted as an insert and a deletion,
and there will be file exist error when renaming the newly-created layer
to replace the old layer. We work around that by changing the end key of
the image layer. A long-term fix would involve a refactor around the
layer file naming. For delta layers, we simply skip layers with the same
key range produced, though it is possible to add an extra key as an
alternative solution.
* The image layer range for the layers generated from gc-compaction will
be Key::MIN..(Key..MAX-1), to avoid being recognized as an L0 delta
layer.
* Skip existing layers if it turns out that we need to generate a layer
with the same persistent key in the same generation.
Note that it is possible that the newly-generated layer has different
content from the existing layer. For example, when the user drops a
retain_lsn, the compaction could have combined or dropped some records,
therefore creating a smaller layer than the existing one. We discard the
"optimized" layer for now because we cannot deal with such rewrites
within the same generation.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
Co-authored-by: Christian Schwarz <christian@neon.tech>
## Problem
We recently added a "visibility" state to layers, but nothing
initializes it.
Part of:
- #8398
## Summary of changes
- Add a dependency on `range-set-blaze`, which is used as a fast
incrementally updated alternative to KeySpace. We could also use this to
replace the internals of KeySpaceRandomAccum if we wanted to. Writing a
type that does this kind of "BtreeMap & merge overlapping entries" thing
isn't super complicated, but no reason to write this ourselves when
there's a third party impl available.
- Add a function to layermap to calculate visibilities for each layer
- Add a function to Timeline to call into layermap and then apply these
visibilities to the Layer objects.
- Invoke the calculation during startup, after image layer creations,
and when removing branches. Branch removal and image layer creation are
the two ways that a layer can go from Visible to Covered.
- Add unit test & benchmark for the visibility calculation
- Expose `pageserver_visible_physical_size` metric, which should always
be <= `pageserver_remote_physical_size`.
- This metric will feed into the /v1/utilization endpoint later: the
visible size indicates how much space we would like to use on this
pageserver for this tenant.
- When `pageserver_visible_physical_size` is greater than
`pageserver_resident_physical_size`, this is a sign that the tenant has
long-idle branches, which result in layers that are visible in
principle, but not used in practice.
This does not keep visibility hints up to date in all cases:
particularly, when creating a child timeline, any previously covered
layers will not get marked Visible until they are accessed.
Updates after image layer creation could be implemented as more of a
special case, but this would require more new code: the existing depth
calculation code doesn't maintain+yield the list of deltas that would be
covered by an image layer.
## Performance
This operation is done rarely (at startup and at timeline deletion), so
needs to be efficient but not ultra-fast.
There is a new `visibility` bench that measures runtime for a synthetic
100k layers case (`sequential`) and a real layer map (`real_map`) with
~26k layers.
The benchmark shows runtimes of single digit milliseconds (on a ryzen
7950). This confirms that the runtime shouldn't be a problem at startup
(as we already incur S3-level latencies there), but that it's slow
enough that we definitely shouldn't call it more often than necessary,
and it may be worthwhile to optimize further later (things like: when
removing a branch, only bother scanning layers below the branchpoint)
```
visibility/sequential time: [4.5087 ms 4.5894 ms 4.6775 ms]
change: [+2.0826% +3.9097% +5.8995%] (p = 0.00 < 0.05)
Performance has regressed.
Found 24 outliers among 100 measurements (24.00%)
2 (2.00%) high mild
22 (22.00%) high severe
min: 0/1696070, max: 93/1C0887F0
visibility/real_map time: [7.0796 ms 7.0832 ms 7.0871 ms]
change: [+0.3900% +0.4505% +0.5164%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 4 outliers among 100 measurements (4.00%)
3 (3.00%) high mild
1 (1.00%) high severe
min: 0/1696070, max: 93/1C0887F0
visibility/real_map_many_branches
time: [4.5285 ms 4.5355 ms 4.5434 ms]
change: [-1.0012% -0.8004% -0.5969%] (p = 0.00 < 0.05)
Change within noise threshold.
```
Before, we had four versions of linux-raw-sys in our dependency graph:
```
linux-raw-sys@0.1.4
linux-raw-sys@0.3.8
linux-raw-sys@0.4.13
linux-raw-sys@0.6.4
```
now it's only two:
```
linux-raw-sys@0.4.13
linux-raw-sys@0.6.4
```
The changes in this PR are minimal. In order to get to its state one
only has to update procfs in Cargo.toml to 0.16 and do `cargo update -p
tempfile -p is-terminal -p prometheus`.
# Motivation
The working theory for hung systemd during PS deploy
(https://github.com/neondatabase/cloud/issues/11387) is that leftover
walredo processes trigger a race condition.
In https://github.com/neondatabase/neon/pull/8150 I arranged that a
clean Tenant shutdown does actually kill its walredo processes.
But many prod machines don't manage to shut down all their tenants until
the 10s systemd timeout hits and, presumably, triggers the race
condition in systemd / the Linux kernel that causes the frozen systemd
# Solution
This PR bolts on a rather ugly mechanism to shut down tenant managers
out of order 8s after we've received the SIGTERM from systemd.
# Changes
- add a global registry of `Weak<WalRedoManager>`
- add a special thread spawned during `shutdown_pageserver` that sleeps
for 8s, then shuts down all redo managers in the registry and prevents
new redo managers from being created
- propagate the new failure mode of tenant spawning throughout the code
base
- make sure shut down tenant manager results in
PageReconstructError::Cancelled so that if Timeline::get calls come in
after the shutdown, they do the right thing
## Problem
In https://github.com/neondatabase/neon/pull/8241 I've accidentally
removed `create-test-report` dependency on `benchmarks` job
## Summary of changes
- Run `create-test-report` after `benchmarks` job
Uses the newly added APIs from #8541 named `stream_tenants_generic` and
`stream_objects_with_retries` and extends them with
`list_objects_with_retries_generic` and
`stream_tenant_timelines_generic` to migrate the `find-garbage` command
of the scrubber to `GenericRemoteStorage`.
Part of https://github.com/neondatabase/neon/issues/7547
## Problem
This code was confusing, untested and covered:
- an impossible case, where intent state is AttacheStale (we never do
this)
- a rare edge case (going from AttachedMulti to Attached), which we were
not testing, and in any case the pageserver internally does the same
Tenant reset in this transition as it would do if we incremented
generation.
Closes: https://github.com/neondatabase/neon/issues/8367
## Summary of changes
- Simplify the logic to only skip incrementing the generation if the
location already has the expected generation and the exact same mode.
In some cases, we can get a negative metric for replication_delay_bytes.
My best guess from all the research I've done is that we evaluate
pg_last_wal_receive_lsn() before pg_last_wal_replay_lsn(), and that by
the time everything is said and done, the replay LSN has advanced past
the receive LSN. In this case, our lag can effectively be modeled as
0 due to the speed of the WAL reception and replay.
Since the introduction of sharding, the protocol handling loop in
`handle_pagerequests` cannot know anymore which concrete
`Tenant`/`Timeline` object any of the incoming `PagestreamFeMessage`
resolves to.
In fact, one message might resolve to one `Tenant`/`Timeline` while
the next one may resolve to another one.
To avoid going to tenant manager, we added the `shard_timelines` which
acted as an ever-growing cache that held timeline gate guards open for
the lifetime of the connection.
The consequence of holding the gate guards open was that we had to be
sensitive to every cached `Timeline::cancel` on each interaction with
the network connection, so that Timeline shutdown would not have to wait
for network connection interaction.
We can do better than that, meaning more efficiency & better
abstraction.
I proposed a sketch for it in
* https://github.com/neondatabase/neon/pull/8286
and this PR implements an evolution of that sketch.
The main idea is is that `mod page_service` shall be solely concerned
with the following:
1. receiving requests by speaking the protocol / pagestream subprotocol
2. dispatching the request to a corresponding method on the correct
shard/`Timeline` object
3. sending response by speaking the protocol / pagestream subprotocol.
The cancellation sensitivity responsibilities are clear cut:
* while in `page_service` code, sensitivity to page_service cancellation
is sufficient
* while in `Timeline` code, sensitivity to `Timeline::cancel` is
sufficient
To enforce these responsibilities, we introduce the notion of a
`timeline::handle::Handle` to a `Timeline` object that is checked out
from a `timeline::handle::Cache` for **each request**.
The `Handle` derefs to `Timeline` and is supposed to be used for a
single async method invocation on `Timeline`.
See the lengthy doc comment in `mod handle` for details of the design.
part of https://github.com/neondatabase/neon/issues/8002
For child branches, we will pull the image of the modified keys from the
parant into the child branch, which creates a full history for
generating key retention. If there are not enough delta keys, the image
won't be wrote eventually, and we will only keep the deltas inside the
child branch. We could avoid the wasteful work to pull the image from
the parent if we can know the number of deltas in advance, in the future
(currently we always pull image for all modified keys in the child
branch)
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
We run regression tests on `release` & `debug` builds for each of the
three supported Postgres versions (6 in total).
With upcoming ARM support and Postgres 17, the number of jobs will jump
to 16, which is a lot.
See the internal discussion here:
https://neondb.slack.com/archives/C033A2WE6BZ/p1722365908404329
## Summary of changes
- Run `regress-tests` job in debug builds only with the latest Postgres
version
- Do not do `debug` builds on release branches
part of https://github.com/neondatabase/neon/issues/8184
# Problem
We want to bypass PS PageCache for all data block reads, but
`compact_level0_phase1` currently uses `ValueRef::load` to load the WAL
records from delta layers.
Internally, that maps to `FileBlockReader:read_blk` which hits the
PageCache
[here](e78341e1c2/pageserver/src/tenant/block_io.rs (L229-L236)).
# Solution
This PR adds a mode for `compact_level0_phase1` that uses the
`MergeIterator` for reading the `Value`s from the delta layer files.
`MergeIterator` is a streaming k-merge that uses vectored blob_io under
the hood, which bypasses the PS PageCache for data blocks.
Other notable changes:
* change the `DiskBtreeReader::into_stream` to buffer the node, instead
of holding a `PageCache` `PageReadGuard`.
* Without this, we run out of page cache slots in
`test_pageserver_compaction_smoke`.
* Generally, `PageReadGuard`s aren't supposed to be held across await
points, so, this is a general bugfix.
# Testing / Validation / Performance
`MergeIterator` has not yet been used in production; it's being
developed as part of
* https://github.com/neondatabase/neon/issues/8002
Therefore, this PR adds a validation mode that compares the existing
approach's value iterator with the new approach's stream output, item by
item.
If they're not identical, we log a warning / fail the unit/regression
test.
To avoid flooding the logs, we apply a global rate limit of once per 10
seconds.
In any case, we use the existing approach's value.
Expected performance impact that will be monitored in staging / nightly
benchmarks / eventually pre-prod:
* with validation:
* increased CPU usage
* ~doubled VirtualFile read bytes/second metric
* no change in disk IO usage because the kernel page cache will likely
have the pages buffered on the second read
* without validation:
* slightly higher DRAM usage because each iterator participating in the
k-merge has a dedicated buffer (as opposed to before, where compactions
would rely on the PS PageCaceh as a shared evicting buffer)
* less disk IO if previously there were repeat PageCache misses (likely
case on a busy production Pageserver)
* lower CPU usage: PageCache out of the picture, fewer syscalls are made
(vectored blob io batches reads)
# Rollout
The new code is used with validation mode enabled-by-default.
This gets us validation everywhere by default, specifically in
- Rust unit tests
- Python tests
- Nightly pagebench (shouldn't really matter)
- Staging
Before the next release, I'll merge the following aws.git PR that
configures prod to continue using the existing behavior:
* https://github.com/neondatabase/aws/pull/1663
# Interactions With Other Features
This work & rollout should complete before Direct IO is enabled because
Direct IO would double the IOPS & latency for each compaction read
(#8240).
# Future Work
The streaming k-merge's memory usage is proportional to the amount of
memory per participating layer.
But `compact_level0_phase1` still loads all keys into memory for
`all_keys_iter`.
Thus, it continues to have active memory usage proportional to the
number of keys involved in the compaction.
Future work should replace `all_keys_iter` with a streaming keys
iterator.
This PR has a draft in its first commit, which I later reverted because
it's not necessary to achieve the goal of this PR / issue #8184.
Change Azure storage configuration to point to new variables/secrets. They have
the `_NEW` suffix in order not to disrupt any tests while we complete the
switch.
Part of #8128, followup to #8480. closes#8421.
Enable scrubber to optionally post metadata scan health results to
storage controller.
Signed-off-by: Yuchen Liang <yuchen@neon.tech>
Part of #8128, followed by #8502.
## Problem
Currently we lack mechanism to alert unhealthy `scan_metadata` status if
we start running this scrubber command as part of a cronjob. With the
storage controller client introduced to storage scrubber in #8196, it is
viable to set up alert by storing health status in the storage
controller database.
We intentionally do not store the full output to the database as the
json blobs potentially makes the table really huge. Instead, only a
health status and a timestamp recording the last time metadata health
status is posted on a tenant shard.
Signed-off-by: Yuchen Liang <yuchen@neon.tech>
This tests the ability to push into ACR using OIDC. Proved it worked by running slightly modified YAML.
In `promote-images` we push the following images `neon compute-tools {vm-,}compute-node-{v14,v15,v16}` into `neoneastus2`.
https://github.com/neondatabase/cloud/issues/14640
## Problem
We don't allow regular end-users to use `k8s-pod` provisioner,
but we still use it in nightly benchmarks
## Summary of changes
- Remove `provisioner` input from `neon-create-project` action, use
`k8s-neonvm` as a default provioner
- Change `neon-` platform prefix to `neonvm-`
- Remove `neon-captest-freetier` and `neon-captest-new` as we already
have their `neonvm` counterparts
Add two new functions `stream_objects_with_retries` and
`stream_tenants_generic` and use them in the `find-large-objects`
subcommand, migrating it to `remote_storage`.
Also adds the `size` field to the `ListingObject` struct.
Part of #7547
If compression is enabled, we currently try compressing each image
larger than a specific size and if the compressed version is smaller, we
write that one, otherwise we use the uncompressed image. However, this
might sometimes be a wasteful process, if there is a substantial amount
of images that don't compress well.
The compression metrics added in #8420
`pageserver_compression_image_in_bytes_total` and
`pageserver_compression_image_out_bytes_total` are well designed for
answering the question how space efficient the total compression process
is end-to-end, which helps one to decide whether to enable it or not.
To answer the question of how much waste there is in terms of trial
compression, so CPU time, we add two metrics:
* one about the images that have been trial-compressed (considered), and
* one about the images where the compressed image has actually been
written (chosen).
There is different ways of weighting them, like for example one could
look at the count, or the compressed data. But the main contributor to
compression CPU usage is amount of data processed, so we weight the
images by their *uncompressed* size. In other words, the two metrics
are:
* `pageserver_compression_image_in_bytes_considered`
* `pageserver_compression_image_in_bytes_chosen`
Part of #5431
## Problem
Old storage buckets can contain a lot of tenants that aren't known to
the control plane at all, because they belonged to test jobs that get
their control plane state cleaned up shortly after running.
In general, it's somewhat unsafe to purge these, as it's hard to
distinguish "control plane doesn't know about this, so it's garbage"
from "control plane said it didn't know about this, which is a bug in
the scrubber, control plane, or API URL configured".
However, the most common case is that we see only a small husk of a
tenant in S3 from a specific old behavior of the software, for example:
- We had a bug where heatmaps weren't deleted on tenant delete
- When WAL DR was first deployed, we didn't delete initdb.tar.zst on
tenant deletion
## Summary of changes
- Add a KnownBug variant for the garbage reason
- Include such cases in the "safe" deletion mode (`--mode=deleted`)
- Add code that inspects tenants missing in control plane to identify
cases of known bugs (this is kind of slow, but should go away once we've
cleaned all these up)
- Add an additional `-min-age` safety check similar to physical GC,
where even if everything indicates objects aren't needed, we won't
delete something that has been modified too recently.
---------
Co-authored-by: Yuchen Liang <70461588+yliang412@users.noreply.github.com>
Co-authored-by: Arpad Müller <arpad-m@users.noreply.github.com>
## Problem
The secondary download HTTP API is meant to return 200 if the download
is complete, and 202 if it is still in progress. In #8198 the download
implementation was changed to drop out with success early if it
over-runs a time budget, which resulted in 200 responses for incomplete
downloads.
This breaks storcon_cli's "tenant-warmup" command, which uses the OK
status to indicate download complete.
## Summary of changes
- Only return 200 if we get an Ok() _and_ the progress stats indicate
the download is complete.
## Problem
We need to test logical replication with 3rd-party tools regularly.
## Summary of changes
Added a test using ClickHouse as a client
Co-authored-by: Alexander Bayandin <alexander@neon.tech>
Uses the Stream based `list_streaming` function added by #8457 in tenant
deletion, as suggested in https://github.com/neondatabase/neon/pull/7932#issuecomment-2150480180 .
We don't have to worry about retries, as the function is wrapped inside
an outer retry block. If there is a retryable error either during the
listing or during deletion, we just do a fresh start.
Also adds `+ Send` bounds as they are required by the
`delete_tenant_remote` function.
## Problem
After https://github.com/neondatabase/neon/pull/7990 `regress_test` job
started to fail with an error:
```
...
File "/__w/neon/neon/test_runner/fixtures/benchmark_fixture.py", line 485, in pytest_terminal_summary
terminalreporter.write(f"{test_report.head_line}.{recorded_property['name']}: ")
TypeError: 'bool' object is not subscriptable
```
https://github.com/neondatabase/neon/actions/runs/10125750938/job/28002582582
It happens because the current implementation doesn't expect pytest's
`user_properties` can be used for anything else but benchmarks (and
https://github.com/neondatabase/neon/pull/7990 started to use it for
tracking `preserve_database_files` parameter)
## Summary of changes
- Make NeonBenchmarker use only records with`neon_benchmarker_` prefix
## Problem
There's a `NeonEnvBuilder#preserve_database_files` parameter that allows
you to keep database files for debugging purposes (by default, files get
cleaned up), but there's no way to get these files from a CI run.
This PR adds handling of `NeonEnvBuilder#preserve_database_files` and
adds the compressed test output directory to Allure reports (for tests
with this parameter enabled).
Ref https://github.com/neondatabase/neon/issues/6967
## Summary of changes
- Compress and add the whole test output directory to Allure reports
- Currently works only with `neon_env_builder` fixture
- Remove `preserve_database_files = True` from sharding tests as
unneeded
---------
Co-authored-by: Christian Schwarz <christian@neon.tech>
Persists whether a timeline is archived or not in `index_part.json`. We
only return success if the upload has actually worked successfully.
Also introduces a new `index_part.json` version number.
Fixes#8459
Part of #8088
close https://github.com/neondatabase/neon/issues/8435
## Summary of changes
If L0 compaction did not include all L0 layers, skip image generation.
There are multiple possible solutions to the original issue, i.e., an
alternative is to wrap the partial L0 compaction in a loop until it
compacts all L0 layers. However, considering that we should weight all
tenants equally, the current solution can ensure everyone gets a chance
to run compaction, and those who write too much won't get a chance to
create image layers. This creates a natural backpressure feedback that
they get a slower read due to no image layers are created, slowing down
their writes, and eventually compaction could keep up with their writes
+ generate image layers.
Consider deployment, we should add an alert on "skipping image layer
generation", so that we won't run into the case that image layers are
not generated => incidents again.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
Problem
-------
wait_lsn timeouts result in a user-facing errors like
```
$ /tmp/neon/pg_install/v16/bin/pgbench -s3424 -i -I dtGvp user=neondb_owner dbname=neondb host=ep-tiny-wave-w23owa37.eastus2.azure.neon.build sslmode=require options='-cstatement_timeout=0 '
dropping old tables...
NOTICE: table "pgbench_accounts" does not exist, skipping
NOTICE: table "pgbench_branches" does not exist, skipping
NOTICE: table "pgbench_history" does not exist, skipping
NOTICE: table "pgbench_tellers" does not exist, skipping
creating tables...
generating data (server-side)...
vacuuming...
pgbench: error: query failed: ERROR: [NEON_SMGR] [shard 0] could not read block 214338 in rel 1663/16389/16839.0 from page server at lsn C/E1C12828
DETAIL: page server returned error: LSN timeout: Timed out while waiting for WAL record at LSN C/E1418528 to arrive, last_record_lsn 6/999D9CA8 disk consistent LSN=6/999D9CA8, WalReceiver status: (update 2024-07-25 08:30:07): connecting to node 25, safekeeper candidates (id|update_time|commit_lsn): [(21|08:30:16|C/E1C129E0), (23|08:30:16|C/E1C129E0), (25|08:30:17|C/E1C129E0)]
CONTEXT: while scanning block 214338 of relation "public.pgbench_accounts"
pgbench: detail: Query was: vacuum analyze pgbench_accounts
```
Solution
--------
Its better to be slow than to fail the queries.
If the app has a deadline, it can use `statement_timeout`.
In the long term, we want to eliminate wait_lsn timeout.
In the short term (this PR), we bump the wait_lsn timeout to
a larger value to reduce the frequency at which these wait_lsn timeouts
occur.
We will observe SLOs and specifically
`pageserver_wait_lsn_seconds_bucket`
before we eliminate the timeout completely.
## Problem
We are missing the step-down primitive required to implement rolling
restarts of the storage controller.
## Summary of changes
Add `/control/v1/step_down` endpoint which puts the storage controller
into a state where it rejects
all API requests apart from `/control/v1/step_down`, `/status` and
`/metrics`. When receiving the request,
storage controller cancels all pending reconciles and waits for them to
exit gracefully. The response contains
a snapshot of the in-memory observed state.
Related:
* https://github.com/neondatabase/cloud/issues/14701
* https://github.com/neondatabase/neon/issues/7797
* https://github.com/neondatabase/neon/pull/8310
## Problem
Vectored get is already enabled in all prod regions without validation.
The pageserver defaults
are out of sync however.
## Summary of changes
Update the pageserver defaults to match the prod config. Also means that
when running tests locally,
people don't have to use the env vars to get the prod config.
## Problem
This is an experiment to see if 16x concurrency is actually helping, or
if it's just giving us very noisy results. If the total runtime with a
lower concurrency is similar, then a lower concurrency is preferable to
reduce the impact of resource-hungry tests running concurrently.
## Problem
This test relies on writing image layers before the split. It can fail
to do so durably if the image layers are written ahead of the remote
consistent LSN, so we should have been doing a checkpoint rather than
just a compaction
## Problem
The scrubber would like to check the highest mtime in a tenant's objects
as a safety check during purges. It recently switched to use
GenericRemoteStorage, so we need to expose that in the listing methods.
## Summary of changes
- In Listing.keys, return a ListingObject{} including a last_modified
field, instead of a RemotePath
---------
Co-authored-by: Arpad Müller <arpad-m@users.noreply.github.com>
## Problem
follow up for #8475
## Summary of changes
Using own private docker registry in `cache-from` and `cache-to`
settings in docker build-push actions
There is a race condition between timeline shutdown and the split task.
Timeline shutdown first shuts down the upload queue, and only then fires
the cancellation token. A parallel running timeline split operation
might thus encounter a cancelled upload queue before the cancellation
token is fired, and print a noisy error.
Fix this by mapping `anyhow::Error{ NotInitialized::ShuttingDown }) to
`FlushLayerError::Cancelled` instead of `FlushLayerError::Other(_)`.
Fixes#8496
update pg_jsonschema extension to v 0.3.1
update pg_graphql extension to v1.5.7
update pgx_ulid extension to v0.1.5
update pg_tiktoken extension, patch Cargo.toml to use new pgrx
This pull request (should) fix the failure of test_gc_feedback. See the
explanation in the newly-added test case.
Part of https://github.com/neondatabase/neon/issues/8002
Allow incomplete history for the compaction algorithm.
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
Storcon shutdown did not produce a clean observed state. This is not a
problem at the moment, but we will need to stop all reconciles with
clean observed state for rolling restarts.
I tried to test this by collecting the observed state during shutdown
and comparing it with the in-memory observed
state, but it doesn't work because a lot of tests use the cursed attach
hook to create tenants directly through the ps.
## Summary of Changes
Rework storcon shutdown as follows:
* Reconcilers get a separate cancellation token which is a child token
of the global `Service::cancel`.
* Reconcilers get a separate gate
* Add a mechanism to drain the reconciler result queue before
* Put all of this together into a clean shutdown sequence
Related https://github.com/neondatabase/cloud/issues/14701
## Problem
This test was destabilized by
https://github.com/neondatabase/neon/pull/8431. The threshold is
arbitrary & failures are still quite close to it. At a high level the
test is asserting "eviction was approximately fair to these tenants",
which appears to still be the case when the abs diff between ratios is
slightly higher at ~0.6-0.7.
## Summary of changes
- Change threshold from 0.06 to 0.065. Based on the last ~10 failures
that should be sufficient.
## Problem
Currently, tests may have a scrub during teardown if they ask for it,
but most tests don't request it. To detect "unknown unknowns", let's run
it at the end of every test where possible. This is similar to asserting
that there are no errors in the log at the end of tests.
## Summary of changes
- Remove explicit `enable_scrub_on_exit`
- Always scrub if remote storage is an S3Storage.
## Problem
Re-attach blocks the pageserver http server from starting up. Hence, it
can't reply to heartbeats
until that's done. This makes the storage controller mark the node
off-line (not good). We worked
around this by setting the interval after which nodes are marked offline
to 5 minutes. This isn't a
long term solution.
## Summary of changes
* Introduce a new `NodeAvailability` state: `WarmingUp`. This state
models the following time interval:
* From receiving the re-attach request until the pageserver replies to
the first heartbeat post re-attach
* The heartbeat delta generator becomes aware of this state and uses a
separate longer interval
* Flag `max-warming-up-interval` now models the longer timeout and
`max-offline-interval` the shorter one to
match the names of the states
Closes https://github.com/neondatabase/neon/issues/7552
## Problem
The rds-aurora endpoint connection cannot be reached from GitHub action
runners.
Temporarily remove this DBMS from the pgbench comparison runs.
## Summary of changes
On Saturday we normally run Neon in comparison with AWS RDS-Postgres and
AWS RDS-Aurora.
Remove Aurora until we have a working setup
Before this PR
1.The circuit breaker would trip on CompactionError::Shutdown. That's
wrong, we want to ignore those cases.
2. remote timeline client shutdown would not be mapped to
CompactionError::Shutdown in all circumstances.
We observed this in staging, see
https://neondb.slack.com/archives/C033RQ5SPDH/p1721829745384449
This PR fixes (1) with a simple `match` statement, and (2) by switching
a bunch of `anyhow` usage over to distinguished errors that ultimately
get mapped to `CompactionError::Shutdown`.
I removed the implicit `#[from]` conversion from `anyhow::Error` to
`CompactionError::Other` to discover all the places that were mapping
remote timeline client shutdown to `anyhow::Error`.
In my opinion `#[from]` is an antipattern and we should avoid it,
especially for `anyhow::Error`. If some callee is going to return
anyhow, the very least the caller should to is to acknowledge, through a
`map_err(MyError::Other)` that they're conflating different failure
reasons.
## Problem
Jobs `check-linux-arm-build` and `check-codestyle-rust-arm` (from
`.github/workflows/neon_extra_builds.yml`) duplicate `build-neon` and
`check-codestyle-rust` jobs in the main pipeline.
## Summary of changes
- Move `check-linux-arm-build` and `check-codestyle-rust-arm` from extra
builds to the main pipeline
By default git does not find a nice hunk header with rust. New(er)
versions ship with a handy xfuncname pattern, so lets enable that for
all developers.
Example of how this should help:
39046172ab
## Problem
PR that modified compaction raced with PR that modified the GcInfo
structure
## Summary of changes
Fix it
Co-authored-by: Vlad Lazar <vlalazar.vlad@gmail.com>
## Problem
The in-memory layer vectored read was very slow in some conditions
(walingest::test_large_rel) test. Upon profiling, I realised that 80% of
the time was spent building up the binary heap of reads. This stage
isn't actually needed.
## Summary of changes
Remove the planning stage as we never took advantage of it in order to
merge reads. There should be no functional change from this patch.
## Problem
- `build-and-test` workflow is pretty big
- jobs that depend on the matrix job don't start before all variations
are done. I.e. `regress-tests` depend on `build-neon`, but we can't
start `regress-tests` on the release configuration until `build-neon` is
done on release **and debug** configurations. This will be more visible
once we add ARM to the matrix.
## Summary of changes
- Move jobs related to building (`build-neon`) and testing
(`regress-tests`) to a separate job
## Problem
The current bucket based rate limiter is not very intuitive and has some
bad failure cases.
## Summary of changes
Switches from fixed interval buckets to leaky bucket impl. A single
bucket per endpoint,
drains over time. Drains by checking the time since the last check, and
draining tokens en-masse. Garbage collection works similar to before, it
drains a shard (1/64th of the set) every 2048 checks, and it only
removes buckets that are empty.
To be compatible with the existing config, I've faffed to make it take
the min and the max rps of each as the sustained rps and the max bucket
size which should be roughly equivalent.
## Problem
Previously, Timeline::gc_info was only updated in a batch operation at
the start of GC. That means that timelines didn't generally have
accurate information about who their children were before the first GC,
or between GC cycles.
Knowledge of child branches is important for calculating layer
visibility in #8398
## Summary of changes
- Split out part of refresh_gc_info into initialize_gc_info, which is
now called early in startup
- Include TimelineId in retain_lsns so that we can later add/remove the
LSNs for particular children
- When timelines are added/removed, update their parent's retain_lsns
## Problem
In `test_basebackup_with_high_slru_count`, the pageserver is sometimes
mysteriously hanging on startup, having been started+stopped earlier in
the test setup while populating template tenant data.
- #7586
We can't see why this is hanging in this particular test. The test does
some weird stuff though, like attaching a load of broken tenants and
then doing a SIGQUIT kill of a pageserver.
## Summary of changes
- Attach tenants normally instead of doing a failpoint dance to attach
them as broken
- Shut the pageserver down gracefully during init instead of using
immediate mode
- Remove the "sequential" variant of the unstable test, as this is going
away soon anyway
- Log before trying to acquire lock file, so that if it hangs we have a
clearer sense of if that's really where it's hanging. It seems like it
is, but that code does a non-blocking flock so it's surprising.
Implements the TODO from #8466 about retries: now the user of the stream
returned by `list_streaming` is able to obtain the next item in the
stream as often as they want, and retry it if it is an error.
Also adds extends the test for paginated listing to include a dedicated
test for `list_streaming`.
follow-up of #8466fixes#8457
part of #7547
---------
Co-authored-by: Joonas Koivunen <joonas@neon.tech>
## Problem
LayerAccessStats contains a lot of detail that we don't use: short
histories of most recent accesses, specifics on what kind of task
accessed a layer, etc. This is all stored inside a Mutex, which is
locked every time something accesses a layer.
## Summary of changes
- Store timestamps at a very low resolution (to the nearest second),
sufficient for use on the timescales of eviction.
- Pack access time and last residence change time into a single u64
- Use the high bits of the u64 for other flags, including the new layer
visibility concept.
- Simplify the external-facing model for access stats to just include
what we now track.
Note that the `HistoryBufferWithDropCounter` is removed here because it
is no longer used. I do not dislike this type, we just happen not to use
it for anything else at present.
Co-authored-by: Christian Schwarz <christian@neon.tech>
## Problem
While investigating problem with test_subscriber_restart flukyness, I
found out that this test is not passed at all for PG 14/15 at MacOS
(while working for PG16).
## Summary of changes
Rewrite async connect state machine exactly in the same way as in
Vanilla: call `WaitLatchOrSocket` with `WL_SOCKETR_WRTEABLE` before
calling `PQconnectPoll`.
Please notice that most likely it will not fix flukyness of
test_subscriber_restart.
## Checklist before requesting a review
- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
This adds the ability to list many prefixes in a streaming fashion to
both the `RemoteStorage` trait as well as `GenericRemoteStorage`.
* The `list` function of the `RemoteStorage` trait is implemented by
default in terms of `list_streaming`.
* For the production users (S3, Azure), `list_streaming` is implemented
and the default `list` implementation is used.
* For `LocalFs`, we keep the `list` implementation and make
`list_streaming` call it.
The `list_streaming` function is implemented for both S3 and Azure.
A TODO for later is retries, which the scrubber currently has while the
`list_streaming` implementations lack them.
part of #8457 and #7547
part of https://github.com/neondatabase/neon/issues/8002
The main thing in this pull request is the new `generate_key_retention`
function. It decides which deltas to retain and generate images for a
given key based on its history + retain_lsn + horizon.
On that, we generate a flat single level of delta layers over all deltas
included in the compaction. In the future, we can decide whether to
split them over the LSN axis as described in the RFC.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
Co-authored-by: Christian Schwarz <christian@neon.tech>
## Problem
`test_change_pageserver` stops pageservers in a way that can overlap
with the controller's heartbeats: the controller can get a heartbeat
success and then immediately find the node unavailable. This particular
situation triggers a log that isn't in our current allow-list of
messages for nodes offline
Example:
https://neon-github-public-dev.s3.amazonaws.com/reports/pr-8339/10048487700/index.html#testresult/19678f27810231df/retries
## Summary of changes
- Add the message to the allow list
## Problem
Postgres is using `access()` function in `GetNewRelFileNumber` to check
if assigned relfilenumber is not used for any other relation. This check
will not work in Neon, because we do not have all files in local
storage.
## Summary of changes
Use smgrexists() instead which will check at page server if such
relfilenode is used.
## Checklist before requesting a review
- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
---------
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
## Problem
As described in https://github.com/neondatabase/neon/issues/8398, layer
visibility is a new hint that will help us manage disk space more
efficiently.
## Summary of changes
- Introduce LayerVisibilityHint and store it as part of access stats
- Automatically mark a layer visible if it is accessed, or when it is
created.
The impact on the access stats size will be reversed in
https://github.com/neondatabase/neon/pull/8431
This is functionally a no-op change: subsequent PRs will add the logic
that sets layers to Covered, and which uses the layer visibility as an
input to eviction and heatmap generation.
---------
Co-authored-by: Joonas Koivunen <joonas@neon.tech>
## Problem
This test sometimes found that ancestors were getting cleaned up before
it had done any compaction.
Compaction was happening implicitly via Workload.
Example:
https://neon-github-public-dev.s3.amazonaws.com/reports/pr-8298/10032173390/index.html#testresult/fb04786402f80822/retries
## Summary of changes
- Set upload=False when writing data after shard split, to avoid doing a
checkpoint
- Add a checkpoint_period & explicit wait for uploads so that we ensure
data lands in S3 without doing a checkpoint
In general, replace:
* 'lfc_approximate_working_set_size' with
* 'lfc_approximate_working_set_size_windows'
For the "main" metrics that are actually scraped and used internally,
the old one is just marked as deprecated.
For the "autoscaling" metrics, we're not currently using the old one, so
we can get away with just replacing it.
Also, for the user-visible metrics we'll only store & expose a few
different time windows, to avoid making the UI overly busy or bloating
our internal metrics storage.
But for the autoscaling-related scraper, we aren't storing the metrics,
and it's useful to be able to programmatically operate on the trendline
of how WSS increases (or doesn't!) with window size. So there, we can
just output datapoints for each minute.
Part of neondatabase/autoscaling#872
See also https://www.notion.so/neondatabase/cca38138fadd45eaa753d81b859490c6
Backup tools such as `tar` and `restic` recognize this.
More info: https://bford.info/cachedir/
NB: cargo _should_ create the tag file in the `target/` directory
but doesn't if the directory already exists, which happens frequently
if rust-analyzer is launched by your IDE before you can type
`cargo build`. Hence, create the file manually here.
=> https://github.com/rust-lang/cargo/issues/14281
## Motivation & Context
We want to move away from `task_mgr` towards explicit tracking of child
tasks.
This PR is extracted from https://github.com/neondatabase/neon/pull/8339
where I refactor `PageRequestHandler` to not depend on task_mgr anymore.
## Changes
This PR refactors all global tasks but `PageRequestHandler` to use some
combination of `JoinHandle`/`JoinSet` + `CancellationToken`.
The `task_mgr::spawn(.., shutdown_process_on_error)` functionality is
preserved through the new `exit_on_panic_or_error` wrapper.
Some global tasks were not using it before, but as of this PR, they are.
The rationale is that all global tasks are relevant for correct
operation of the overall Neon system in one way or another.
## Future Work
After #8339, we can make `task_mgr::spawn` require a `TenantId` instead
of an `Option<TenantId>` which concludes this step of cleanup work and
will help discourage future usage of task_mgr for global tasks.
Part of #8128.
## Problem
Scrubber uses `scan_metadata` command to flag metadata inconsistencies.
To trust it at scale, we need to make sure the errors we emit is a
reflection of real scenario. One check performed in the scrubber is to
see whether layers listed in the latest `index_part.json` is present in
object listing. Currently, the scrubber does not robustly handle the
case where objects are uploaded/deleted during the scan.
## Summary of changes
**Condition for success:** An object in the index is (1) in the object
listing we acquire from S3 or (2) found in a HeadObject request (new
object).
- Add in the `HeadObject` requests for the layers missing from the
object listing.
- Keep the order of first getting the object listing and then
downloading the layers.
- Update check to only consider shards with highest shard count.
- Skip analyzing a timeline if `deleted_at` tombstone is marked in
`index_part.json`.
- Add new test to see if scrubber actually detect the metadata
inconsistency.
_Misc_
- A timeline with no ancestor should always have some layers.
- Removed experimental histograms
_Caveat_
- Ancestor layer is not cleaned until #8308 is implemented. If ancestor
layers reference non-existing layers in the index, the scrubber will
emit false positives.
Signed-off-by: Yuchen Liang <yuchen@neon.tech>
Starts using the `remote_storage` crate in the S3 scrubber for the
`PurgeGarbage` subcommand.
The `remote_storage` crate is generic over various backends and thus
using it gives us the ability to run the scrubber against all supported
backends.
Start with the `PurgeGarbage` subcommand as it doesn't use
`stream_tenants`.
Part of #7547.
## Problem
This test predates the storage controller. It stops pageservers and
reconfigures computes, but that races with the storage controller's node
failure detection, which can result in restarting nodes not getting the
attachments they expect, and the test failing
## Summary of changes
- Configure the storage controller to use a compute notify hook that
does nothing, so that it cannot interfere with the test's configuration
of computes.
- Instead of using the attach hook, just notify the storage controller
that nodes are offline, and reconcile tenants so that they will
automatically be attached to the other node.
## Problem
Deployed pageserver configurations are all like this:
```
disk_usage_based_eviction:
max_usage_pct: 85
min_avail_bytes: 0
period: "10s"
eviction_order:
type: "RelativeAccessed"
args:
highest_layer_count_loses_first: true
```
But we're maintaining this optional absolute order eviction, with test
cases etc.
## Summary of changes
- Remove absolute order eviction. Make the default eviction policy the
same as how we really deploy pageservers.
## Problem
We have two No.34 RFC.
## Summary of changes
## Checklist before requesting a review
- [x] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
On Azure we need to use username-password authentication in proxy for
regional redis client.
## Summary of changes
This adds `redis_auth_type` to the config with default value of "irsa".
Not specifying it will enforce the `regional_redis_client` to be
configured with IRSA redis (as it's done now).
If "plain" is specified, then the regional client is condifigured with
`redis_notifications`, consuming username:password auth from URI. We
plan to do that for the Azure cloud.
Configuring `regional_redis_client` is required now, there is no opt-out
from configuring it.
https://github.com/neondatabase/cloud/issues/14462
PR #8299 has switched the storage scrubber to use
`DefaultCredentialsChain`. Now we do this for `remote_storage`, as it
allows us to use `remote_storage` from inside kubernetes. Most of the
diff is due to `GenericRemoteStorage::from_config` becoming `async fn`.
This adds an archival_config endpoint to the pageserver. Currently it
has no effect, and always "works", but later the intent is that it will
make a timeline archived/unarchived.
- [x] add yml spec
- [x] add endpoint handler
Part of https://github.com/neondatabase/neon/issues/8088
## Problem
There are some swagger errors in `pageserver/src/http/openapi_spec.yml`
```
Error 431 15000 Object includes not allowed fields
Error 569 3100401 should always have a 'required'
Error 569 15000 Object includes not allowed fields
Error 1111 10037 properties members must be schemas
```
## Summary of changes
Fixed the above errors.
## Problem
After a shard split, the pageserver leaves the ancestor shard's content
in place. It may be referenced by child shards, but eventually child
shards will de-reference most ancestor layers as they write their own
data and do GC. We would like to eventually clean up those ancestor
layers to reclaim space.
## Summary of changes
- Extend the physical GC command with `--mode=full`, which includes
cleaning up unreferenced ancestor shard layers
- Add test `test_scrubber_physical_gc_ancestors`
- Remove colored log output: in testing this is irritating ANSI code
spam in logs, and in interactive use doesn't add much.
- Refactor storage controller API client code out of storcon_client into
a `storage_controller/client` crate
- During physical GC of ancestors, call into the storage controller to
check that the latest shards seen in S3 reflect the latest state of the
tenant, and there is no shard split in progress.
We're removing the usage of this long-meaningless config field in
https://github.com/neondatabase/aws/pull/1599
Once that PR has been deployed to staging and prod, we can merge this
PR.
## Problem
My prior PR https://github.com/neondatabase/neon/pull/8422
caused leftovers in the GitHub action runner work directory with root
permission.
As an example see here
https://github.com/neondatabase/neon/actions/runs/10001857641/job/27646237324#step:3:37
To work-around we install vanilla postgres as non-root using deb
packages in /home/nonroot user directory
## Summary of changes
- since we cannot use root we install the deb pkgs directly and create
symbolic links for psql, pgbench and libs in expected places
- continue jobs an aws even if azure jobs fail (because this region is
currently unreliable)
Successor of #8288 , just enable zstd in tests. Also adds a test that
creates easily compressable data.
Part of #5431
---------
Co-authored-by: John Spray <john@neon.tech>
Co-authored-by: Joonas Koivunen <joonas@neon.tech>
The error means that manager exited earlier than `ResidenceGuard` and
it's not unexpected with current deletion implementation. This commit
changes log level to reduse noise.
Use the k-merge iterator in the compaction process to reduce memory
footprint.
part of https://github.com/neondatabase/neon/issues/8002
## Summary of changes
* refactor the bottom-most compaction code to use k-merge iterator
* add Send bound on some structs as it is used across the await points
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
Co-authored-by: Arpad Müller <arpad-m@users.noreply.github.com>
We have an issue that some partial uploaded segments can be actually
missing in remote storage. I found this issue when was looking at the
logs in staging, and it can be triggered by failed uploads:
1. Code tries to upload `SEG_TERM_LSN_LSN_sk5.partial`, but receives
error from S3
2. The failed attempt is saved to `segments` vec
3. After some time, the code tries to upload
`SEG_TERM_LSN_LSN_sk5.partial` again
4. This time the upload is successful and code calls `gc()` to delete
previous uploads
5. Since new object and old object share the same name, uploaded data
gets deleted from remote storage
This commit fixes the issue by patching `gc()` not to delete objects
with the same name as currently uploaded.
---------
Co-authored-by: Arpad Müller <arpad-m@users.noreply.github.com>
## Problem
Ahead of enabling eviction in the field, where it will become the
normal/default mode, let's enable it by default throughout our tests in
case any issues become visible there.
## Summary of changes
- Make default `extra_opts` for safekeepers enable offload & deletion
- Set low timeouts in `extra_opts` so that tests running for tens of
seconds have a chance to hit some of these background operations.
## Problem
These tests time out ~1 in 50 runs when in debug mode.
There is no indication of a real issue: they're just wrappers that have
large numbers of individual tests contained within on pytest case.
## Summary of changes
- Bump pg_regress timeout from 600 to 900s
- Bump test_isolation timeout from 300s (default) to 600s
In future it would be nice to break out these tests to run individual
cases (or batches thereof) as separate tests, rather than this monolith.
## Problem
This test would occasionally fail its metric check. This could happen in
the rare case that the nodes had all been restarted before their most
recent eviction.
The metric check was added in
https://github.com/neondatabase/neon/pull/8348
## Summary of changes
- Check metrics before each restart, accumulate into a bool that we
assert on at the end of the test
When `NeonEnv.from_repo_dir` was introduced, storage controller stored
its
state exclusively `attachments.json`.
Since then, it has moved to using Postgres, which stores its state in
`storage_controller_db`.
But `NeonEnv.from_repo_dir` wasn't adjusted to do this.
This PR rectifies the situation.
Context for this is failures in
`test_pageserver_characterize_throughput_with_n_tenants`
CF:
https://neondb.slack.com/archives/C033RQ5SPDH/p1721035799502239?thread_ts=1720901332.293769&cid=C033RQ5SPDH
Notably, `from_repo_dir` is also used by the backwards- and
forwards-compatibility.
Thus, the changes in this PR affect those tests as well.
However, it turns out that the compatibility snapshot already contains
the `storage_controller_db`.
Thus, it should just work and in fact we can remove hacks like
`fixup_storage_controller`.
Follow-ups created as part of this work:
* https://github.com/neondatabase/neon/issues/8399
* https://github.com/neondatabase/neon/issues/8400
## Problem
There are something wrong in the comment of
`control_plane/src/broker.rs` and `control_plane/src/pageserver.rs`
## Summary of changes
Fixed the comment about component name and their data path in
`control_plane/src/broker.rs` and `control_plane/src/pageserver.rs`.
## Problem
We lack insight into:
- How much of a tenant's physical size is image vs. delta layers
- Average sizes of image vs. delta layers
- Total layer counts per timeline, indicating size of index_part object
As well as general observability love, this is motivated by
https://github.com/neondatabase/neon/issues/6738, where we need to
define some sensible thresholds for storage amplification, and using
total physical size may not work well (if someone does a lot of DROPs
then it's legitimate for the physical-synthetic ratio to be huge), but
the ratio between image layer size and delta layer size may be a better
indicator of whether we're generating unreasonable quantities of image
layers.
## Summary of changes
- Add pageserver_layer_bytes and pageserver_layer_count metrics,
labelled by timeline and `kind` (delta or image)
- Add & subtract these with LayerInner's lifetime.
I'm intentionally avoiding using a generic metric RAII guard object, to
avoid bloating LayerInner: it already has all the information it needs
to update metric on new+drop.
This test reproduces the case of a writer creating a deep stack of L0
layers. It uses realistic layer sizes and writes several gigabytes of
data, therefore runs as a performance test although it is validating
memory footprint rather than performance per se.
It acts a regression test for two recent fixes:
- https://github.com/neondatabase/neon/pull/8401
- https://github.com/neondatabase/neon/pull/8391
In future it will demonstrate the larger improvement of using a k-merge
iterator for L0 compaction (#8184)
This test can be extended to enforce limits on the memory consumption of
other housekeeping steps, by restarting the pageserver and then running
other things to do the same "how much did RSS increase" measurement.
Existing tenants and some selection of layers might produce duplicated
keys. Add tests to ensure the k-merge iterator handles it correctly. We
also enforced ordering of the k-merge iterator to put images before
deltas.
part of https://github.com/neondatabase/neon/issues/8002
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
Co-authored-by: Arpad Müller <arpad-m@users.noreply.github.com>
## Problem
We want to run performance tests on all supported cloud providers.
We want to run most tests on the postgres version which is default for
new projects in production, currently (July 24) this is postgres version
16
## Summary of changes
- change default postgres version for some (performance) tests to 16
(which is our default for new projects in prod anyhow)
- add azure region to pgbench_compare jobs
- add azure region to pgvector benchmarking jobs
- re-used project `weathered-snowflake-88107345` was prepared with 1
million embeddings running on 7 minCU 7 maxCU in azure region to compare
with AWS region (pgvector indexing and hnsw queries)
- see job pgbench-pgvector
- Note we now have a 11 environments combinations where we run
pgbench-compare and 5 are for k8s-pod (deprecated) which we can remove
in the future once auto-scaling team approves.
## Logs
A current run with the changes from this pull request is running here
https://github.com/neondatabase/neon/actions/runs/9972096222
Note that we currently expect some failures due to
- https://github.com/neondatabase/neon/issues/8275
- instability of projects on azure region
## Problem
When a tenant creates a new timeline that they will treat as their
'main' history,
it is awkward to permanently retain an 'old main' timeline as its
ancestor. Currently
this is necessary because it is forbidden to delete a timeline which has
descendents.
## Summary of changes
A new pageserver API is proposed to 'adopt' data from a parent timeline
into
one of its children, such that the link between ancestor and child can
be severed,
leaving the parent in a state where it may then be deleted.
---------
Co-authored-by: Joonas Koivunen <joonas@neon.tech>
## Problem
ValueRef is an unnecessarily large structure, because it carries a
cursor. L0 compaction currently instantiates gigabytes of these under
some circumstances.
## Summary of changes
- Carry a ref to the parent layer instead of a cursor, and construct a
cursor on demand.
This reduces RSS high watermark during L0 compaction by about 20%.
## Problem
The `evictions_with_low_residence_duration` is used as an indicator of
cache thrashing. However, there are situations where it is quite
legitimate to only have a short residence during compaction, where a
delta is downloaded, used to generate an image layer, and then
discarded. This can lead to false positive alerts.
## Summary of changes
- Only track low residence duration for layers that have been accessed
at least once (compaction doesn't count as an access). This will give us
a metric that indicates thrashing on layers that the _user_ is using,
rather than those we're downloading for housekeeping purposes.
Once we add "layer visibility" as an explicit property of layers, this
can also be used as a cleaner condition (residence of non-visible layers
should never be alertable)
## Problem
close https://github.com/neondatabase/neon/issues/8389
## Summary of changes
A quick mitigation for tenants with fast writes. We compact at most 60
delta layers at a time, expecting a memory footprint of 15GB. We will
pick the oldest 60 L0 layers.
This should be a relatively safe change so no test is added. Question is
whether to make this parameter configurable via tenant config.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
Co-authored-by: John Spray <john@neon.tech>
- `horizon` is a confusing term, it's not at all obvious that this means
space-based retention limit, rather than the total GC history limit.
Rename to `GcCutoffs::space`.
- `pitr` is less confusing, but still an unecessary level of indirection
from what we really mean: a time-based condition. The fact that we use
that that time-history for Point In Time Recovery doesn't mean we have
to refer to time as "pitr" everywhere. Rename to `GcCutoffs::time`.
As described in #8385, the likely source for flakiness in
test_tenant_creation_fails is the following sequence of events:
1. test instructs the storage controller to create the tenant
2. storage controller adds the tenant and persists it to the database.
issues a creation request
3. the pageserver restarts with the failpoint disabled
4. storage controller's background reconciliation still wants to create
the tenant
5. pageserver gets new request to create the tenant from background
reconciliation
This commit just avoids the storage controller entirely. It has its own
set of issues, as the re-attach request will obviously not include the
tenant, but it's still useful to test for non-existence of the tenant.
The generation is also not optional any more during tenant attachment.
If you omit it, the pageserver yields an error. We change the signature
of `tenant_attach` to reflect that.
Alternative to #8385Fixes#8266
## Problem
This structure was in an Arc<> unnecessarily, making it harder to reason
about its lifetime (i.e. it was superficially possible for LayerManager
to outlive timeline, even though no code used it that way)
## Summary of changes
- Remove the Arc<>
Right now timeline detach ancestor reports an error (409, "no ancestor")
on a new attempt after successful completion. This makes it troublesome
for storage controller retries. Fix it to respond with `200 OK` as if
the operation had just completed quickly.
Additionally, the returned timeline identifiers in the 200 OK response
are now ordered so that responses between different nodes for error
comparison are done by the storage controller added in #8353.
Design-wise, this PR introduces a new strategy for accessing the latest
uploaded IndexPart:
`RemoteTimelineClient::initialized_upload_queue(&self) ->
Result<UploadQueueAccessor<'_>, NotInitialized>`. It should be a more
scalable way to query the latest uploaded `IndexPart` than to add a
query method for each question directly on `RemoteTimelineClient`.
GC blocking will need to be introduced to make the operation fully
idempotent. However, it is idempotent for the cases demonstrated by
tests.
Cc: #6994
## Problem
Pageserver GC uses a size-based condition (GC "horizon" in addition to
time-based "PITR").
Eventually we plan to retire the size-based condition:
https://github.com/neondatabase/neon/issues/6374
Currently, we always apply the more conservative of the two, meaning
that tenants always retain at least 64MB of history (default horizon),
even after a very long time has passed. This is particularly acute in
cases where someone has dropped tables/databases, and then leaves a
database idle: the horizon can prevent GCing very large quantities of
historical data (we already account for this in synthetic size by
ignoring gc horizon).
We're not entirely removing GC horizon right now because we don't want
to 100% rely on standby_horizon for robustness of physical replication,
but we can tweak our logic to avoid retaining that 64MB LSN length
indefinitely.
## Summary of changes
- Rework `Timeline::find_gc_cutoffs`, with new logic:
- If there is no PITR set, then use `DEFAULT_PITR_INTERVAL` (1 week) to
calculate a time threshold. Retain either the horizon or up to that
thresholds, whichever requires less data.
- When there is a PITR set, and we have unambiguously resolved the
timestamp to an LSN, then ignore the GC horizon entirely. For typical
PITRs (1 day, 1 week), this will still easily retain enough data to
avoid stressing read only replicas.
The key property we end up with, whether a PITR is set or not, is that
after enough time has passed, our GC cutoff on an idle timeline will
catch up with the last_record_lsn.
Using `DEFAULT_PITR_INTERVAL` is a bit of an arbitrary hack, but this
feels like it isn't really worth the noise of exposing in TenantConfig.
We could just make it a different named constant though. The end-end
state will be that there is no gc_horizon at all, and that tenants with
pitr_interval=0 would truly retain no history, so this constant would go
away.
Currently storage controller does not support forwarding timeline detach
ancestor requests to pageservers. Add support for forwarding `PUT
.../:tenant_id/timelines/:timeline_id/detach_ancestor`. Implement the
support mostly as is, because the timeline detach ancestor will be made
(mostly) idempotent in future PR.
Cc: #6994
## Problem
Right now if there are too many running xacts to be restored from CLOG
at replica startup,
then replica is not trying to restore them and wait for non-overflown
running-xacs WAL record from primary.
But if primary is not active, then replica will not start at all.
Too many running xacts can be caused by transactions with large number
of subtractions.
But right now it can be also cause by two reasons:
- Lack of shutdown checkpoint which updates `oldestRunningXid` (because
of immediate shutdown)
- nextXid alignment on 1024 boundary (which cause loosing ~1k XIDs on
each restart)
Both problems are somehow addressed now.
But we have existed customers with "sparse" CLOG and lack of
checkpoints.
To be able to start RO replicas for such customers I suggest to add GUC
which allows replica to start even in case of subxacts overflow.
## Summary of changes
Add `neon.running_xacts_overflow_policy` with the following values:
- ignore: restore from CLOG last N XIDs and accept connections
- skip: do not restore any XIDs from CXLOGbut still accept connections
- wait: wait non-overflown running xacts record from primary node
## Checklist before requesting a review
- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
---------
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
## Problem
This test would sometimes violate the min resident size during disk
eviction and fail due to the generate warning log.
Disk usage candidate collection only takes into account active tenants.
However, the statvfs call takes into account the entire tenants
directory, which includes tenants which haven't become active yet.
After re-starting the pageserver, disk usage eviction may kick in
*before* both tenants have become active. Hence, the logic will try to satisfy
thedisk usage requirements by evicting everything belonging to the active
tenant, and hence violating the tenant minimum resident size.
## Summary of changes
Allow the warning
## Problem
new clippy warnings on nightly.
## Summary of changes
broken up each commit by warning type.
1. Remove some unnecessary refs.
2. In edition 2024, inference will default to `!` and not `()`.
3. Clippy complains about doc comment indentation
4. Fix `Trait + ?Sized` where `Trait: Sized`.
5. diesel_derives triggering `non_local_defintions`
## Problem
We already back off on compaction retries, but the impact of a failing
compaction can be so great that backing off up to 300s isn't enough. The
impact is consuming a lot of I/O+CPU in the case of image layer
generation for large tenants, and potentially also leaking disk space.
Compaction failures are extremely rare and almost always indicate a bug,
frequently a bug that will not let compaction to proceed until it is
fixed.
Related: https://github.com/neondatabase/neon/issues/6738
## Summary of changes
- Introduce a CircuitBreaker type
- Add a circuit breaker for compaction, with a policy that after 5
failures, compaction will not be attempted again for 24 hours.
- Add metrics that we can alert on: any >0 value for
`pageserver_circuit_breaker_broken_total` should generate an alert.
- Add a test that checks this works as intended.
Couple notes to reviewers:
- Circuit breakers are intrinsically a defense-in-depth measure: this is
not the solution to any underlying issues, it is just a general
mitigation for "unknown unknowns" that might be encountered in future.
- This PR isn't primarily about writing a perfect CircuitBreaker type:
the one in this PR is meant to be just enough to mitigate issues in
compaction, and make it easy to monitor/alert on these failures. We can
refine this type in future as/when we want to use it elsewhere.
Implement decompression of images for vectored reads.
This doesn't implement support for still treating blobs as uncompressed
with the bits we reserved for compression, as we have removed that
functionality in #8300 anyways.
Part of #5431
We need to pass on the configured compression param during image layer
generation.
This was an oversight of #8106, and the likely cause why #8288 didn't
bring any interesting regressions.
Part of https://github.com/neondatabase/neon/issues/5431
## Problem
I need `neon_superuser` to be allowed to create snapshots for
replication tests
## Summary of changes
Adds a migration that grants these functions to neon_superuser
Rewrite streaming vectored read planner to be a separate struct. The API
is designed to produce batches around `max_read_size` instead of exactly
less than that so that `handle_XX` returns one batch a time.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
In anticipation of later adding a really nice drain+delete API, I
initially only added an intentionally basic `/drop` API that is just
about usable for deleting nodes in a pinch, but requires some ugly
storage controller restarts to persuade it to restart secondaries.
## Summary of changes
I started making a few tiny fixes, and ended up writing the delete
API...
- Quality of life nit: ordering of node + tenant listings in storcon_cli
- Papercut: Fix the attach_hook using the wrong operation type for
reporting slow locks
- Make Service::spawn tolerate `generation_pageserver` columns that
point to nonexistent node IDs. I started out thinking of this as a
general resilience thing, but when implementing the delete API I
realized it was actually a legitimate end state after the delete API is
called (as that API doesn't wait for all reconciles to succeed).
- Add a `DELETE` API for nodes, which does not gracefully drain, but
does reschedule everything. This becomes safe to use when the system is
in any state, but will incur availability gaps for any tenants that
weren't already live-migrated away. If tenants have already been
drained, this becomes a totally clean + safe way to decom a node.
- Add a test and a storcon_cli wrapper for it
This is meant to be a robust initial API that lets us remove nodes
without doing ugly things like restarting the storage controller -- it's
not quite a totally graceful node-draining routine yet. There's more
work in https://github.com/neondatabase/neon/issues/8333 to get to our
end-end state.
## Problem
Follow up to https://github.com/neondatabase/neon/pull/8335, to improve
observability of how many evict/restores we are doing.
## Summary of changes
- Add `safekeeper_eviction_events_started_total` and
`safekeeper_eviction_events_completed_total`, with a "kind" label of
evict or restore. This gives us rates, and also ability to calculate how
many are in progress.
- Generalize SafekeeperMetrics test type to use the same helpers as
pageserver, and enable querying any metric.
- Read the new metrics at the end of the eviction test.
## Problem
SeqWait::would_wait_for returns Ok in the case when we would not wait
for the sequence number and Err otherwise.
ReconcilerWaiter::get_status uses it the wrong way around. This can
cause the storage controller to go into a busy loop
and make it look unavailable to the k8s controller.
## Summary of changes
Use `SeqWait::would_wait_for` correctly.
`trace_read_requests` is a per `Tenant`-object option.
But the `handle_pagerequests` loop doesn't know which
`Tenant` object (i.e., which shard) the request is for.
The remaining use of the `Tenant` object is to check `tenant.cancel`.
That check is incorrect [if the pageserver hosts multiple
shards](https://github.com/neondatabase/neon/issues/7427#issuecomment-2220577518).
I'll fix that in a future PR where I completely eliminate the holding
of `Tenant/Timeline` objects across requests.
See [my code RFC](https://github.com/neondatabase/neon/pull/8286) for
the
high level idea.
Note that we can always bring the tracing functionality if we need it.
But since it's actually about logging the `page_service` wire bytes,
it should be a `page_service`-level config option, not per-Tenant.
And for enabling tracing on a single connection, we can implement
a `set pageserver_trace_connection;` option.
Set core rmilit to ulimited in compute_ctl, so that all child processes
inherit it. We could also set rlimit in relevant startup script, but
that way we would depend on external setup and might inadvertently
disable it again (core dumping worked in pods, but not in VMs with
inittab-based startup).
## Problem
- The condition for eviction is not time-based: it is possible for a
timeline to be restored in response to a client, that client times out,
and then as soon as the timeline is restored it is immediately evicted
again.
- There is no delay on eviction at startup of the safekeeper, so when it
starts up and sees many idle timelines, it does many evictions which
will likely be immediately restored when someone uses the timeline.
## Summary of changes
- Add `eviction_min_resident` parameter, and use it in
`ready_for_eviction` to avoid evictions if the timeline has been
resident for less than this period.
- This also implicitly delays evictions at startup for
`eviction_min_resident`
- Set this to a very low number for the existing eviction test, which
expects immediate eviction.
The default period is 15 minutes. The general reasoning for that is that
in the worst case where we thrash ~10k timelines on one safekeeper,
downloading 16MB for each one, we should set a period that would not
overwhelm the node's bandwidth.
Part of https://github.com/neondatabase/neon/issues/8002. This pull
request adds a k-merge iterator for bottom-most compaction.
## Summary of changes
* Added back lsn_range / key_range in delta layer inner. This was
removed due to https://github.com/neondatabase/neon/pull/8050, but added
back because iterators need that information to process lazy loading.
* Added lazy-loading k-merge iterator.
* Added iterator wrapper as a unified iterator type for image+delta
iterator.
The current status and test should cover the use case for L0 compaction
so that the L0 compaction process can bypass page cache and have a fixed
amount of memory usage. The next step is to integrate this with the new
bottom-most compaction.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
Co-authored-by: Christian Schwarz <christian@neon.tech>
Removes the `ImageCompressionAlgorithm::DisabledNoDecompress` variant.
We now assume any blob with the specific bits set is actually a
compressed blob.
The `ImageCompressionAlgorithm::Disabled` variant still remains and is
the new default.
Reverts large parts of #8238 , as originally intended in that PR.
Part of #5431
## Problem
This test incorrectly assumed that a post-split compaction would only
drop content. This was easily destabilized by any changes to image
generation rules.
## Summary of changes
- Before split, do a full image layer generation pass, to guarantee that
post-split compaction should only drop data, never create it.
- Fix the force_image_layer_creation mode of compaction that we use from
tests like this: previously it would try and generate image layers even
if one already existed with the same layer key, which caused compaction
to fail.
## Problem
#7809 - we do not support sslnegotiation=direct
#7810 - we do not support negotiating down the protocol extensions.
## Summary of changes
1. Same as postgres, check the first startup packet byte for tls header
`0x16`, and check the ALPN.
2. Tell clients using protocol >3.0 to downgrade
I want to fix bugs in `page_service`
([issue](https://github.com/neondatabase/neon/issues/7427)) and the
`import basebackup` / `import wal` stand in the way / make the
refactoring more complicated.
We don't use these methods anyway in practice, but, there have been some
objections to removing the functionality completely.
So, this PR preserves the existing functionality but moves it into the
HTTP management API.
Note that I don't try to fix existing bugs in the code, specifically not
fixing
* it only ever worked correctly for unsharded tenants
* it doesn't clean up on error
All errors are mapped to `ApiError::InternalServerError`.
## Problem
Slack thread:
https://neondb.slack.com/archives/C033RQ5SPDH/p1720511577862519
We're seeing OOMs in staging on a pageserver that has
l0_flush.mode=Direct enabled.
There's a strong correlation between jumps in `maxrss_kb` and
`pageserver_timeline_ephemeral_bytes`, so, it's quite likely that
l0_flush.mode=Direct is the culprit.
Notably, the expected max memory usage on that staging server by the
l0_flush.mode=Direct is ~2GiB but we're seeing as much as 24GiB max RSS
before the OOM kill.
One hypothesis is that we're dropping the semaphore permit before all
the dirtied pages have been flushed to disk. (The flushing to disk
likely happens in the fsync inside the `.finish()` call, because we're
using ext4 in data=ordered mode).
## Summary of changes
Hold the permit until after we're done with `.finish()`.
part of https://github.com/neondatabase/cloud/issues/14024, k8s does not
always have a volume available for logging, and I'm running into weird
permission errors... While I could spend time figuring out how to create
temp directories for logging, I think it would be better to just disable
file logging as k8s containers are ephemeral and we cannot retrieve
anything on the fs after the container gets removed.
## Summary of changes
`PAGESERVER_DISABLE_FILE_LOGGING=1` -> file logging disabled
Signed-off-by: Alex Chi Z <chi@neon.tech>
This tweaks the rows-to-JSON rendering logic in order to avoid
allocating 0-sized temporary vectors and later growing them
to insert elements.
As the exact size is known in advance, both vectors can be built
with an exact capacity upfront. This will avoid further vector
growing/reallocation in the rendering hotpath.
Signed-off-by: Luca BRUNO <lucab@lucabruno.net>
## Problem
In https://github.com/neondatabase/neon/pull/8161, we changed the path
to Neon artefacts by adding commit sha to it, but we missed adding these
changes to `promote-compatibility-data` job that we use for
backward/forward- compatibility testing.
## Summary of changes
- Add commit sha to `promote-compatibility-data`
## Summary of changes
Increase the `assert_size_approx_equal` threshold to avoid flakiness of
`test_lsn_lease_size`. Still needs more investigation to fully resolve
#8293.
- Also set `autovacuum=off` for the endpoint we are running in the test.
Signed-off-by: Yuchen Liang <yuchen@neon.tech>
## Problem
`test_timeline_size_quota_on_startup` assumed that writing data beyond
the size limit would always be blocked. This is not so: the limit is
only enforced if feedback makes it back from the pageserver to the
safekeeper + compute.
Closes: https://github.com/neondatabase/neon/issues/6562
## Summary of changes
- Modify the test to wait for the pageserver to catch up. The size limit
was never actually being enforced robustly, the original version of this
test was just writing much more than 30MB and about 98% of the time
getting lucky such that the feedback happened to arrive before the tests
for loop was done.
- If the test fails, log the logical size as seen by the pageserver.
## Problem
Debug-mode runs of test_pg_regress are rather slow since
https://github.com/neondatabase/neon/pull/8105, and occasionally exceed
their 600s timeout.
## Summary of changes
- Use 8MiB layer files, avoiding large ephemeral layers
On a hetzner AX102, this takes the runtime from 230s to 190s. Which
hopefully will be enough to get the runtime on github runners more
reliably below its 600s timeout.
This has the side benefit of exercising more of the pageserver stack
(including compaction) under a workload that exercises a more diverse
set of postgres functionality than most of our tests.
## Problem
We currently use 'immediate' mode in the most commonly used shutdown
path, when the control plane calls a `compute_ctl` API to terminate
Postgres inside compute without waiting for the actual pod / VM
termination. Yet, 'immediate' shutdown doesn't create a shutdown
checkpoint and ROs have bad times figuring out the list of running xacts
during next start.
## Summary of changes
Use 'fast' mode, which creates a shutdown checkpoint that is important
for ROs to get a list of running xacts faster instead of going through
the CLOG. On the control plane side, we poll this `compute_ctl`
termination API for 10s, it should be enough as we don't really write
any data at checkpoint time. If it times out, we anyway switch to the
slow k8s-based termination.
See https://www.postgresql.org/docs/current/server-shutdown.html for the
list of modes and signals.
The default VM shutdown hook already uses `fast` mode, see [1]
[1]
c9fd8d7693/vm-image-spec.yaml (L30-L31)
Related to #6211
## Problem
LSN Leases introduced in #8084 is a new API that is made shard-aware
from day 1. To support ephemeral endpoint in #7994 without linking
Postgres C API against `compute_ctl`, part of the sharding needs to
reside in `utils`.
## Summary of changes
- Create a new `shard` module in utils crate.
- Move more interface related part of tenant sharding API to utils and
re-export them in pageserver_api.
Signed-off-by: Yuchen Liang <yuchen@neon.tech>
## Problem
Rarely, a dbdir entry can exist with no `relmap_file_key` data. This
causes compaction to fail, because it assumes that if the database
exists, then so does the relmap file.
Basebackup already handled this using a boolean to record whether such a
key exists, but `collect_keyspace` didn't.
## Summary of changes
- Respect the flag for whether a relfilemap exists in collect_keyspace
- The reproducer for this issue will merge separately in
https://github.com/neondatabase/neon/pull/8232
These tests will help verify that replication, both physical and
logical, works as expected in Neon.
Co-authored-by: Sasha Krassovsky <sasha@neon.tech>
Allows a process to run without blocking program execution, which can be
useful for certain test scenarios.
Co-authored-by: Sasha Krassovsky <sasha@neon.tech>
## Problem
- Resident memory on long running pageserver processes tends to climb:
memory fragmentation is suspected.
- Total resident memory may be a limiting factor for running on smaller
nodes.
## Summary of changes
- As a low-energy experiment, switch the pageserver to use jemalloc (not
a net-new dependency, proxy already use it)
- Decide at end of week whether to revert before next release.
## Problem
Sparse keyspaces were constructed with ranges out of order: this didn't break things obviously, but meant that users of KeySpace functions that assume ordering would assert out.
Closes https://github.com/neondatabase/neon/issues/8277
## Summary of changes
make sure the sparse keyspace has ordered keyspace parts
The find-large-objects scrubber subcommand is quite fast if you run it
in an environment with low latency to the S3 bucket (say an EC2 instance
in the same region). However, the higher the latency gets, the slower
the command becomes. Therefore, add a concurrency param and make it
parallelized. This doesn't change that general relationship, but at
least lets us do multiple requests in parallel and therefore hopefully
faster.
Running with concurrency of 64 (default):
```
2024-07-05T17:30:22.882959Z INFO lazy_load_identity [...]
[...]
2024-07-05T17:30:28.289853Z INFO Scanned 500 shards. [...]
```
With concurrency of 1, simulating state before this PR:
```
2024-07-05T17:31:43.375153Z INFO lazy_load_identity [...]
[...]
2024-07-05T17:33:51.987092Z INFO Scanned 500 shards. [...]
```
In other words, to list 500 shards, speed is increased from 2:08 minutes
to 6 seconds.
Follow-up of #8257, part of #5431
Improve parsing of the `ImageCompressionAlgorithm` enum to allow level
customization like `zstd(1)`, as strum only takes `Default::default()`,
i.e. `None` as the level.
Part of #5431
## Problem
test_subscriber_restart has quit large failure rate'
https://neonprod.grafana.net/d/fddp4rvg7k2dcf/regression-test-failures?orgId=1&var-test_name=test_subscriber_restart&var-max_count=100&var-restrict=false
I can be caused by too small timeout (5 seconds) to wait until changes
are propagated.
Related to #8097
## Summary of changes
Increase timeout to 30 seconds.
## Checklist before requesting a review
- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
## Problem
We want to be able to test how our infrastructure reacts on segfaults in
Postgres (for example, we collect cores, and get some required
logs/metrics, etc)
## Summary of changes
- Add `trigger_segfauls` function to `neon_test_utils` to trigger a
segfault in Postgres
- Add `trigger_panic` function to `neon_test_utils` to trigger SIGABRT
(by using `elog(PANIC, ...))
- Fix cleanup logic in regression tests in endpoint crashed
## Problem
Assume a timeline with the following workload: very slow ingest of
updates to a small number of keys that fit within the same partition (as decided by
`KeySpace::partition`). These tenants will create small L0 layers since due to time
based rolling, and, consequently, the L1 layers will also be small.
Currently, by default, we need to ingest 512 MiB of WAL before checking
if an image layer is required. This scheme works fine under the assumption that L1s are roughly of
checkpoint distance size, but as the first paragraph explained, that's not the case for all workloads.
## Summary of changes
Check if new image layers are required at least once every checkpoint timeout interval.
## Problem
Safekeepers left running for a long time use a lot of memory (up to the
point of OOMing, on small nodes) for deleted timelines, because the
`Timeline` struct is kept alive as a guard against recreating deleted
timelines.
Closes: https://github.com/neondatabase/neon/issues/6810
## Summary of changes
- Create separate tombstones that just record a ttid and when the
timeline was deleted.
- Add a periodic housekeeping task that cleans up tombstones older than
a hardcoded TTL (24h)
I think this also makes https://github.com/neondatabase/neon/pull/6766
un-needed, as the tombstone is also checked during deletion.
I considered making the overall timeline map use an enum type containing
active or deleted, but having a separate map of tombstones avoids
bloating that map, so that calls like `get()` can still go straight to a
timeline without having to walk a hashmap that also contains tombstones.
## Problem
This test directly manages locations on pageservers and configuration of
an endpoint. However, it did not switch off the parts of the storage
controller that attempt to do the same: occasionally, the test would
fail in a strange way such as a compute failing to accept a
reconfiguration request.
## Summary of changes
- Wire up the storage controller's compute notification hook to a no-op
handler
- Configure the tenant's scheduling policy to Stop.
## Problem
See #7466
## Summary of changes
Implement algorithm descried in
https://hal.science/hal-00465313/document
Now new GUC is added:
`neon.wss_max_duration` which specifies size of sliding window (in
seconds). Default value is 1 hour.
It is possible to request estimation of working set sizes (within this
window using new function
`approximate_working_set_size_seconds`. Old function
`approximate_working_set_size` is preserved for backward compatibility.
But its scope is also limited by `neon.wss_max_duration`.
Version of Neon extension is changed to 1.4
## Checklist before requesting a review
- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
---------
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
Co-authored-by: Matthias van de Meent <matthias@neon.tech>
Part of #7497, closes#8071. (accidentally closed#8208, reopened here)
## Problem
After the changes in #8084, we need synthetic size to also account for
leased LSNs so that users do not get free retention by running a small
ephemeral endpoint for a long time.
## Summary of changes
This PR integrates LSN leases into the synthetic size calculation. We
model leases as read-only branches started at the leased LSN (except it
does not have a timeline id).
Other changes:
- Add new unit tests testing whether a lease behaves like a read-only
branch.
- Change `/size_debug` response to include lease point in the SVG
visualization.
- Fix `/lsn_lease` HTTP API to do proper parsing for POST.
Signed-off-by: Yuchen Liang <yuchen@neon.tech>
Co-authored-by: Joonas Koivunen <joonas@neon.tech>
Co-authored-by: Christian Schwarz <christian@neon.tech>
Adds a find-large-objects subcommand to the scrubber to allow listing
layer objects larger than a specific size.
To be used like:
```
AWS_PROFILE=dev REGION=us-east-2 BUCKET=neon-dev-storage-us-east-2 cargo run -p storage_scrubber -- find-large-objects --min-size 250000000 --ignore-deltas
```
Part of #5431
## Problem
When generations were new, these messages were an important way of
noticing if something unexpected was going on. We found some real issues
when investigating tests that unexpectedly tripped them.
At time has gone on, this code is now pretty battle-tested, and as we do
more live migrations etc, it's fairly normal to see the occasional
message from a node with a stale generation.
At this point the cognitive load on developers to selectively allow-list
these logs outweighs the benefit of having them at warn severity.
Closes: https://github.com/neondatabase/neon/issues/8080
## Summary of changes
- Downgrade "Dropped remote consistent LSN updates" and "Dropping stale
deletions" messages to INFO
- Remove all the allow-list entries for these logs.
## Problem
`pg-clients` workflow looks different from the main `build-and-test`
workflow for historical reasons (it was my very first task at Neon, and
back then I wasn't really familiar with the rest of the CI pipelines).
This PR unifies `pg-clients` workflow with `build-and-test`
## Summary of changes
- Rename `pg_clients.yml` to `pg-clients.yml`
- Run the workflow on changes in relevant files
- Create Allure report for tests
- Send slack notifications to `#on-call-qa-staging-stream` channel
(instead of `#on-call-staging-stream`)
- Update Client libraries once we're here
## Problem
I'd like to keep this in the tree since it might be useful in prod as
well. It's a bit too noisy as is and missing the lsn.
## Summary of changes
Add an lsn field and and increase the rate limit duration.
## Problem
Currently, if you need to rename a job and the job is listed in [branch
protection
rules](https://github.com/neondatabase/neon/settings/branch_protection_rules),
the PR won't be allowed to merge.
## Summary of changes
- Add `conclusion` job that fails if any of its dependencies don't
finish successfully
## Problem
If there's a quota error, it makes sense to cache it for a short window
of time. Many clients do not handle database connection errors
gracefully, so just spam retry 🤡
## Summary of changes
Updates the node_info cache to support storing console errors. Store
console errors if they cannot be retried (using our own heuristic.
should only trigger for quota exceeded errors).
## Problem
The metrics we have today aren't convenient for planning around the
impact of timeline archival on costs.
Closes: https://github.com/neondatabase/neon/issues/8108
## Summary of changes
- Add metric `pageserver_archive_size`, which indicates the logical
bytes of data which we would expect to write into an archived branch.
- Add metric `pageserver_pitr_history_size`, which indicates the
distance between last_record_lsn and the PITR cutoff.
These metrics are somewhat temporary: when we implement #8088 and
associated consumption metric changes, these will reach a final form.
For now, an "archived" branch is just any branch outside of its parent's
PITR window: later, archival will become an explicit state (which will
_usually_ correspond to falling outside the parent's PITR window).
The overall volume of timeline metrics is something to watch, but we are
removing many more in https://github.com/neondatabase/neon/pull/8245
than this PR is adding.
I'd like to add some constraints to the layer map we generate in tests.
(1) is the layer map that the current compaction algorithm will produce.
There is a property that for all delta layer, all delta layer overlaps
with it on the LSN axis will have the same LSN range.
(2) is the layer map that cannot be produced with the legacy compaction
algorithm.
(3) is the layer map that will be produced by the future
tiered-compaction algorithm. The current validator does not allow that
but we can modify the algorithm to allow it in the future.
## Summary of changes
Add a validator to check if the layer map is valid and refactor the test
cases to include delta layer start/end LSN.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
Co-authored-by: Christian Schwarz <christian@neon.tech>
## Problem
We record detailed histograms for all page_service op types, which
mostly aren't very interesting, but make our prometheus scrapes huge.
Closes: #8223
## Summary of changes
- Only track GetPageAtLsn histograms on a per-timeline granularity. For
all other operation types, rely on existing node-wide histograms.
we want to run some specific pagebench test cases on dedicated hardware
to get reproducible results
run1: 1 client per tenant => characterize throughput with n tenants.
- 500 tenants
- scale 13 (200 MB database)
- 1 hour duration
- ca 380 GB layer snapshot files
run2.singleclient: 1 client per tenant => characterize latencies
run2.manyclient: N clients per tenant => characterize throughput
scalability within one tenant.
- 1 tenant with 1 client for latencies
- 1 tenant with 64 clients because typically for a high number of
connections we recommend the connection pooler
which by default uses 64 connections (for scalability)
- scale 136 (2048 MB database)
- 20 minutes each
PR #8106 was created with the assumption that no blob is larger than
`256 MiB`. Due to #7852 we have checking for *writes* of blobs larger
than that limit, but we didn't have checking for *reads* of such large
blobs: in theory, we could be reading these blobs every day but we just
don't happen to write the blobs for some reason.
Therefore, we now add a warning for *reads* of such large blobs as well.
To make deploying compression less dangerous, we therefore only assume a
blob is compressed if the compression setting is present in the config.
This also means that we can't back out of compression once we enabled
it.
Part of https://github.com/neondatabase/neon/issues/5431
## Problem
test_location_conf_churn fails on log errors when it tries to shutdown a
pageserver immediately after starting a tenant attach, like this:
https://neon-github-public-dev.s3.amazonaws.com/reports/pr-8224/9761000525/index.html#/testresult/15fb6beca5c7327c
```
shutdown:shutdown{tenant_id=35f5c55eb34e7e5e12288c5d8ab8b909 shard_id=0000}:timeline_shutdown{timeline_id=30936747043353a98661735ad09cbbfe shutdown_mode=FreezeAndFlush}: failed to freeze and flush: cannot flush frozen layers when flush_loop is not running, state is Exited\n')
```
This is happening because Tenant::shutdown fires its cancellation token
early if the tenant is not fully attached by the time shutdown is
called, so the flush loop is shutdown by the time we try and flush.
## Summary of changes
- In the early-cancellation case, also set the shutdown mode to Hard to
skip trying to do a flush that will fail.
## Problem
GitHub Actions complain that we use actions that depend on deprecated
Node 16:
```
Node.js 16 actions are deprecated. Please update the following actions to use Node.js 20: docker/setup-buildx-action@v2
```
But also, the latest `docker/setup-buildx-action` fails with the following
error:
```
/nvme/actions-runner/_work/_actions/docker/setup-buildx-action/v3/webpack:/docker-setup-buildx/node_modules/@actions/cache/lib/cache.js:175
throw new Error(`Path Validation Error: Path(s) specified in the action for caching do(es) not exist, hence no cache is being saved.`);
^
Error: Path Validation Error: Path(s) specified in the action for caching do(es) not exist, hence no cache is being saved.
at Object.rejected (/nvme/actions-runner/_work/_actions/docker/setup-buildx-action/v3/webpack:/docker-setup-buildx/node_modules/@actions/cache/lib/cache.js:175:1)
at Generator.next (<anonymous>)
at fulfilled (/nvme/actions-runner/_work/_actions/docker/setup-buildx-action/v3/webpack:/docker-setup-buildx/node_modules/@actions/cache/lib/cache.js:29:1)
```
We can work this around by setting `cache-binary: false` for `uses:
docker/setup-buildx-action@v3`
## Summary of changes
- Update `docker/setup-buildx-action` from `v2` to `v3`, set
`cache-binary: false`
- Update `docker/login-action` from `v2` to `v3`
- Update `docker/build-push-action` from `v4`/`v5` to `v6`
All the code to ensure the WAL record lands at a page boundary was
unnecessary for reproducing the original problem. In fact, it's a pretty
basic test that checks that outbound replication (= neon as publisher)
still works after restarting the endpoint. It just used to be very
broken before commit 5ceccdc7de, which also added this test.
To verify that:
1. Check out commit f3af5f4660 (because the next commit, 7dd58e1449,
fixed the same bug in a different way, making it infeasible to revert
the bug fix in an easy way)
2. Revert the bug fix from commit 5ceccdc7de with this:
```
diff --git a/pgxn/neon/walproposer_pg.c b/pgxn/neon/walproposer_pg.c
index 7debb6325..9f03bbd99 100644
--- a/pgxn/neon/walproposer_pg.c
+++ b/pgxn/neon/walproposer_pg.c
@@ -1437,8 +1437,10 @@ XLogWalPropWrite(WalProposer *wp, char *buf, Size nbytes, XLogRecPtr recptr)
*
* https://github.com/neondatabase/neon/issues/5749
*/
+#if 0
if (!wp->config->syncSafekeepers)
XLogUpdateWalBuffers(buf, recptr, nbytes);
+#endif
while (nbytes > 0)
{
```
3. Run the test_wal_page_boundary_start regression test. It fails, as
expected
4. Apply this commit to the test, and run it again. It still fails, with
the same error mentioned in issue #5749:
```
PG:2024-06-30 20:49:08.805 GMT [1248196] STATEMENT: START_REPLICATION SLOT "sub1" LOGICAL 0/0 (proto_version '4', origin 'any', publication_names '"pub1"')
PG:2024-06-30 21:37:52.567 GMT [1467972] LOG: starting logical decoding for slot "sub1"
PG:2024-06-30 21:37:52.567 GMT [1467972] DETAIL: Streaming transactions committing after 0/1532330, reading WAL from 0/1531C78.
PG:2024-06-30 21:37:52.567 GMT [1467972] STATEMENT: START_REPLICATION SLOT "sub1" LOGICAL 0/0 (proto_version '4', origin 'any', publication_names '"pub1"')
PG:2024-06-30 21:37:52.567 GMT [1467972] LOG: logical decoding found consistent point at 0/1531C78
PG:2024-06-30 21:37:52.567 GMT [1467972] DETAIL: There are no running transactions.
PG:2024-06-30 21:37:52.567 GMT [1467972] STATEMENT: START_REPLICATION SLOT "sub1" LOGICAL 0/0 (proto_version '4', origin 'any', publication_names '"pub1"')
PG:2024-06-30 21:37:52.568 GMT [1467972] ERROR: could not find record while sending logically-decoded data: invalid contrecord length 312 (expected 6) at 0/1533FD8
```
## Problem
See https://github.com/neondatabase/cloud/issues/14289
and PR #8210
## Summary of changes
Add test for problems fixed in #8210
## Checklist before requesting a review
- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
---------
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
## Problem
Some of the Nightly benchmarks fail with the error
```
+ /tmp/neon/pg_install/v14/bin/pgbench --version
/tmp/neon/pg_install/v14/bin/pgbench: error while loading shared libraries: libpq.so.5: cannot open shared object file: No such file or directory
```
Originally, we added the `pgbench --version` call to check that
`pgbench` is installed and to fail earlier if it's not.
The failure happens because we don't have `LD_LIBRARY_PATH` set for
every job, and it also affects `psql` command.
We can move it to `actions/run-python-test-set` so as not to duplicate
code (as it already have `LD_LIBRARY_PATH` set).
## Summary of changes
- Remove `pgbench --version` call
- Move `psql` commands to common `actions/run-python-test-set`
part of https://github.com/neondatabase/neon/issues/7418
# Motivation
(reproducing #7418)
When we do an `InMemoryLayer::write_to_disk`, there is a tremendous
amount of random read I/O, as deltas from the ephemeral file (written in
LSN order) are written out to the delta layer in key order.
In benchmarks (https://github.com/neondatabase/neon/pull/7409) we can
see that this delta layer writing phase is substantially more expensive
than the initial ingest of data, and that within the delta layer write a
significant amount of the CPU time is spent traversing the page cache.
# High-Level Changes
Add a new mode for L0 flush that works as follows:
* Read the full ephemeral file into memory -- layers are much smaller
than total memory, so this is afforable
* Do all the random reads directly from this in memory buffer instead of
using blob IO/page cache/disk reads.
* Add a semaphore to limit how many timelines may concurrently do this
(limit peak memory).
* Make the semaphore configurable via PS config.
# Implementation Details
The new `BlobReaderRef::Slice` is a temporary hack until we can ditch
`blob_io` for `InMemoryLayer` => Plan for this is laid out in
https://github.com/neondatabase/neon/issues/8183
# Correctness
The correctness of this change is quite obvious to me: we do what we did
before (`blob_io`) but read from memory instead of going to disk.
The highest bug potential is in doing owned-buffers IO. I refactored the
API a bit in preliminary PR
https://github.com/neondatabase/neon/pull/8186 to make it less
error-prone, but still, careful review is requested.
# Performance
I manually measured single-client ingest performance from `pgbench -i
...`.
Full report:
https://neondatabase.notion.site/2024-06-28-benchmarking-l0-flush-performance-e98cff3807f94cb38f2054d8c818fe84?pvs=4
tl;dr:
* no speed improvements during ingest, but
* significantly lower pressure on PS PageCache (eviction rate drops to
1/3)
* (that's why I'm working on this)
* noticable but modestly lower CPU time
This is good enough for merging this PR because the changes require
opt-in.
We'll do more testing in staging & pre-prod.
# Stability / Monitoring
**memory consumption**: there's no _hard_ limit on max `InMemoryLayer`
size (aka "checkpoint distance") , hence there's no hard limit on the
memory allocation we do for flushing. In practice, we a) [log a
warning](23827c6b0d/pageserver/src/tenant/timeline.rs (L5741-L5743))
when we flush oversized layers, so we'd know which tenant is to blame
and b) if we were to put a hard limit in place, we would have to decide
what to do if there is an InMemoryLayer that exceeds the limit.
It seems like a better option to guarantee a max size for frozen layer,
dependent on `checkpoint_distance`. Then limit concurrency based on
that.
**metrics**: we do have the
[flush_time_histo](23827c6b0d/pageserver/src/tenant/timeline.rs (L3725-L3726)),
but that includes the wait time for the semaphore. We could add a
separate metric for the time spent after acquiring the semaphore, so one
can infer the wait time. Seems unnecessary at this point, though.
Add support for reading and writing zstd-compressed blobs for use in
image layer generation, but maybe one day useful also for delta layers.
The reading of them is unconditional while the writing is controlled by
the `image_compression` config variable allowing for experiments.
For the on-disk format, we re-use some of the bitpatterns we currently
keep reserved for blobs larger than 256 MiB. This assumes that we have
never ever written any such large blobs to image layers.
After the preparation in #7852, we now are unable to read blobs with a
size larger than 256 MiB (or write them).
A non-goal of this PR is to come up with good heuristics of when to
compress a bitpattern. This is left for future work.
Parts of the PR were inspired by #7091.
cc #7879
Part of #5431
## Problem
At high percentiles we see more than 800 layers being visited by the
read path. We need the tenant/timeline to investigate.
## Summary of changes
Add a rate limited log line when the average number of layers visited
per key is in the last specified histogram bucket.
I plan to use this to identify tenants in us-east-2 staging that exhibit
this behaviour. Will revert before next week's release.
Before this PR, during timeline shutdown, we'd occasionally see
log lines like this one:
```
2024-06-26T18:28:11.063402Z INFO initial_size_calculation{tenant_id=$TENANT,shard_id=0000 timeline_id=$TIMELINE}:logical_size_calculation_task:get_or_maybe_download{layer=000000000000000000000000000000000000-000000067F0001A3950001C1630100000000__0000000D88265898}: layer file download failed, and caller has been cancelled: Cancelled, shutting down
Stack backtrace:
0: <core::result::Result<T,F> as core::ops::try_trait::FromResidual<core::result::Result<core::convert::Infallible,E>>>::from_residual
at /rustc/129f3b9964af4d4a709d1383930ade12dfe7c081/library/core/src/result.rs:1964:27
pageserver::tenant::remote_timeline_client::RemoteTimelineClient::download_layer_file::{{closure}}
at /home/nonroot/pageserver/src/tenant/remote_timeline_client.rs:531:13
pageserver::tenant::storage_layer::layer::LayerInner::download_and_init::{{closure}}
at /home/nonroot/pageserver/src/tenant/storage_layer/layer.rs:1136:14
pageserver::tenant::storage_layer::layer::LayerInner::download_init_and_wait::{{closure}}::{{closure}}
at /home/nonroot/pageserver/src/tenant/storage_layer/layer.rs:1082:74
```
We can eliminate the anyhow backtrace with no loss of information
because the conversion to anyhow::Error happens in exactly one place.
refs #7427
## Problem
Tenant attachment has error paths for failures to write local
configuration, but these types of local storage I/O errors should be
considered fatal for the process. Related thread on an earlier PR that
touched this code:
https://github.com/neondatabase/neon/pull/7947#discussion_r1655134114
## Summary of changes
- Make errors writing tenant config fatal (abort process)
- When reading tenant config, make all I/O errors except ENOENT fatal
- Replace use of bare anyhow errors with `LoadConfigError`
Before this PR, `RemoteStorageConfig::from_toml` would support
deserializing an
empty `{}` TOML inline table to a `None`, otherwise try `Some()`.
We can instead let
* in proxy: let clap derive handle the Option
* in PS & SK: assume that if the field is specified, it must be a valid
RemtoeStorageConfig
(This PR started with a much simpler goal of factoring out the
`deserialize_item` function because I need that in another PR).
## Problem
See https://github.com/neondatabase/cloud/issues/14289
## Summary of changes
Check connection status after calling PQconnectStartParams
## Checklist before requesting a review
- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
---------
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
This makes it much more convenient to use in the common case that you
want to flush all the WAL. (Passing pg_current_wal_insert_lsn() as the
argument doesn't work for the same reasons as explained in the comments:
we need to be back off to the beginning of a page if the previous record
ended at page boundary.)
I plan to use this to fix the issue that Arseny Sher called out at
https://github.com/neondatabase/neon/pull/7288#discussion_r1660063852
## Problem
We use `build-tools` image as a base image to build other images, and it
has a pretty old `libpq-dev` installed (v13; it wasn't that old until I
removed system Postgres 14 from `build-tools` image in
https://github.com/neondatabase/neon/pull/6540)
## Summary of changes
- Remove `libpq-dev` from `build-tools` image
- Set `LD_LIBRARY_PATH` for tests (for different Postgres binaries that
we use, like psql and pgbench)
- Set `PQ_LIB_DIR` to build Storage Controller
- Set `LD_LIBRARY_PATH`/`DYLD_LIBRARY_PATH` in the Storage Controller
where it calls Postgres binaries
## Problem
We lack visibility of how much local disk space is used by secondary
tenant locations
Close: https://github.com/neondatabase/neon/issues/8181
## Summary of changes
- Add `pageserver_secondary_resident_physical_size`, tagged by tenant
- Register & de-register label sets from SecondaryTenant
- Add+use wrappers in SecondaryDetail that update metrics when
adding+removing layers/timelines
We have one pretty serious MVCC visibility bug with hot standby
replicas. We incorrectly treat any transactions that are in progress
in the primary, when the standby is started, as aborted. That can
break MVCC for queries running concurrently in the standby. It can
also lead to hint bits being set incorrectly, and that damage can last
until the replica is restarted.
The fundamental bug was that we treated any replica start as starting
from a shut down server. The fix for that is straightforward: we need
to set 'wasShutdown = false' in InitWalRecovery() (see changes in the
postgres repo).
However, that introduces a new problem: with wasShutdown = false, the
standby will not open up for queries until it receives a running-xacts
WAL record from the primary. That's correct, and that's how Postgres
hot standby always works. But it's a problem for Neon, because:
* It changes the historical behavior for existing users. Currently,
the standby immediately opens up for queries, so if they now need to
wait, we can breka existing use cases that were working fine
(assuming you don't hit the MVCC issues).
* The problem is much worse for Neon than it is for standalone
PostgreSQL, because in Neon, we can start a replica from an
arbitrary LSN. In standalone PostgreSQL, the replica always starts
WAL replay from a checkpoint record, and the primary arranges things
so that there is always a running-xacts record soon after each
checkpoint record. You can still hit this issue with PostgreSQL if
you have a transaction with lots of subtransactions running in the
primary, but it's pretty rare in practice.
To mitigate that, we introduce another way to collect the
running-xacts information at startup, without waiting for the
running-xacts WAL record: We can the CLOG for XIDs that haven't been
marked as committed or aborted. It has limitations with
subtransactions too, but should mitigate the problem for most users.
See https://github.com/neondatabase/neon/issues/7236.
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
This makes it much more convenient to use in the common case that you
want to flush all the WAL. (Passing pg_current_wal_insert_lsn() as the
argument doesn't work for the same reasons as explained in the
comments: we need to be back off to the beginning of a page if the
previous record ended at page boundary.)
I plan to use this to fix the issue that Arseny Sher called out at
https://github.com/neondatabase/neon/pull/7288#discussion_r1660063852
The 'running' boolean was replaced with a semaphore in commit
f0e2bb79b2, but this initialization was missed. Remove it so that if a
test tries to access it, you get an error rather than always claiming
that the endpoint is not running.
Spotted by Arseny at
https://github.com/neondatabase/neon/pull/7288#discussion_r1660068657
Whenever we see an XLOG_MULTIXACT_CREATE_ID WAL record, we need to
update the nextMulti and NextMultiOffset fields in the pageserver's
copy of the CheckPoint struct, to cover the new multi-XID. In
PostgreSQL, this is done by updating an in-memory struct during WAL
replay, but because in Neon you can start a compute node at any LSN,
we need to have an up-to-date value pre-calculated in the pageserver
at all times. We do the same for nextXid.
However, we had a bug in WAL ingestion code that does that: the
multi-XIDs will wrap around at 2^32, just like XIDs, so we need to do
the comparisons in a wraparound-aware fashion.
Fix that, and add tests.
Fixes issue #6520
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
## Problem
At the start of do_tenant_shard_split, we drop any secondary location
for the parent shards. The reconciler uses presence of secondary
locations as a condition for enabling heatmaps.
On the pageserver, child shards inherit their configuration from
parents, but the storage controller assumes the child's ObservedState is
the same as the parent's config from the prepare phase. The result is
that some child shards end up with inaccurate ObservedState, and until
something next migrates or restarts, those tenant shards aren't
uploading heatmaps, so their secondary locations are downloading
everything that was resident at the moment of the split (including
ancestor layers which are often cleaned up shortly after the split).
Closes: https://github.com/neondatabase/neon/issues/8189
## Summary of changes
- Use PlacementPolicy to control enablement of heatmap upload, rather
than the literal presence of secondaries in IntentState: this way we
avoid switching them off during shard split
- test: during tenant split test, assert that the child shards have
heatmap uploads enabled.
## Problem
Very long running downloads can be wasteful, because the heatmap they're
using is outdated after a few minutes.
Closes: https://github.com/neondatabase/neon/issues/8182
## Summary of changes
- Impose a deadline on timeline downloads, using the same period as we
use for scheduling, and returning an UpdateError::Restart when it is
reached. This restart will involve waiting for a scheduling interval,
but that's a good thing: it helps let other tenants proceed.
- Refactor download_timeline so that the part where we update the state
for local layers is done even if we fall out of the layer download loop
with an error: this is important, especially for big tenants, because
only layers in the SecondaryDetail state will be considered for
eviction.
part of https://github.com/neondatabase/neon/issues/7418
I reviewed how the VirtualFile API's `read` methods look like and came
to the conclusion that we've been using `IoBufMut` / `BoundedBufMut` /
`Slice` wrong.
This patch rectifies the situation.
# Change 1: take `tokio_epoll_uring::Slice` in the read APIs
Before, we took an `IoBufMut`, which is too low of a primitive and while
it _seems_ convenient to be able to pass in a `Vec<u8>` without any
fuzz, it's actually very unclear at the callsite that we're going to
fill up that `Vec` up to its `capacity()`, because that's what
`IoBuf::bytes_total()` returns and that's what
`VirtualFile::read_exact_at` fills.
By passing a `Slice` instead, a caller that "just wants to read into a
`Vec`" is forced to be explicit about it, adding either `slice_full()`
or `slice(x..y)`, and these methods panic if the read is outside of the
bounds of the `Vec::capacity()`.
Last, passing slices is more similar to what the `std::io` APIs look
like.
# Change 2: fix UB in `virtual_file_io_engine=std-fs`
While reviewing call sites, I noticed that the
`io_engine::IoEngine::read_at` method for `StdFs` mode has been
constructing an `&mut[u8]` from raw parts that were uninitialized.
We then used `std::fs::File::read_exact` to initialize that memory, but,
IIUC we must not even be constructing an `&mut[u8]` where some of the
memory isn't initialized.
So, stop doing that and add a helper ext trait on `Slice` to do the
zero-initialization.
# Change 3: eliminate `read_exact_at_n`
The `read_exact_at_n` doesn't make sense because the caller can just
1. `slice = buf.slice()` the exact memory it wants to fill
2. `slice = read_exact_at(slice)`
3. `buf = slice.into_inner()`
Again, the `std::io` APIs specify the length of the read via the Rust
slice length.
We should do the same for the owned buffers IO APIs, i.e., via
`Slice::bytes_total()`.
# Change 4: simplify filling of `PageWriteGuard`
The `PageWriteGuardBuf::init_up_to` was never necessary.
Remove it. See changes to doc comment for more details.
---
Reviewers should probably look at the added test case first, it
illustrates my case a bit.
## Problem
For some time, we have created tenants with calls to location_conf. The
legacy "POST /v1/tenant" path was only used in some tests.
## Summary of changes
- Remove the API
- Relocate TenantCreateRequest to the controller API file (this used to
be used in both pageserver and controller APIs)
- Rewrite tenant_create test helper to use location_config API, as
control plane and storage controller do
- Update docker-compose test script to create tenants with
location_config API (this small commit is also present in
https://github.com/neondatabase/neon/pull/7947)
After https://github.com/neondatabase/neon/pull/8022 was deployed to
staging, I noticed many cases of timeouts. After inspecting the logs, I
realized that some operations are taking ~20 seconds and they're doing
while holding shared state lock. Usually it happens right after
redeploy, because compute reconnections put high load on disks. This
commit tries to improve observability around slow operations.
Non-observability changes:
- `TimelineState::finish_change` now skips update if nothing has changed
- `wal_residence_guard()` timeout is set to 30s
While investigating Pageserver logs from the cases where systemd hangs
during shutdown (https://github.com/neondatabase/cloud/issues/11387), I
noticed that even if Pageserver shuts down cleanly[^1], there are
lingering walredo processes.
[^1]: Meaning, pageserver finishes its shutdown procedure and calls
`exit(0)` on its own terms, instead of hitting the systemd unit's
`TimeoutSec=` limit and getting SIGKILLed.
While systemd should never lock up like it does, maybe we can avoid
hitting that bug by cleaning up properly.
Changes
-------
This PR adds a shutdown method to `WalRedoManager` and hooks it up to
tenant shutdown.
We keep track of intent to shutdown through the new `enum
ProcessOnceCell` stored inside the pre-existing `redo_process` field.
A gate is added to keep track of running processes, using the new type
`struct Process`.
Future Work
-----------
Requests that don't need the redo process will not observe the shutdown
(see doc comment).
Doing so would be nice for completeness sake, but doesn't provide much
benefit because `Tenant` and `Timeline` already shut down all walredo
users.
Testing
-------
I did manual testing to confirm that the problem exists before this PR
and that it's gone after.
Setup:
* `neon_local` with a single tenant, create some data using `pgbench`
* ensure walredo process is running, not pid
* watch `strace -e kill,wait4 -f -p "$(pgrep pageserver)"`
* `neon_local pageserver stop`
With this PR, we always observe
```
$ strace -e kill,wait4 -f -p "$(pgrep pageserver)"
...
[pid 591120] --- SIGTERM {si_signo=SIGTERM, si_code=SI_USER, si_pid=591215, si_uid=1000} ---
[pid 591134] kill(591174, SIGKILL) = 0
[pid 591134] wait4(591174, <unfinished ...>
[pid 591142] --- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_KILLED, si_pid=591174, si_uid=1000, si_status=SIGKILL, si_utime=0, si_stime=0} ---
[pid 591134] <... wait4 resumed>[{WIFSIGNALED(s) && WTERMSIG(s) == SIGKILL}], 0, NULL) = 591174
...
+++ exited with 0 +++
```
Before this PR, we'd usually observe just
```
...
[pid 596239] --- SIGTERM {si_signo=SIGTERM, si_code=SI_USER, si_pid=596455, si_uid=1000} ---
...
+++ exited with 0 +++
```
Refs
----
refs https://github.com/neondatabase/cloud/issues/11387
## Problem
Background node operations take a long time for loaded nodes.
## Summary of changes
Increase number of concurrent reconciles an operation is allowed to
spawn.
This should make drain and fill operations faster and the new value is
still well below the total limit of concurrent reconciles.
## Problem
The cache keys that we use on CI are the same for X64 and ARM64
(`runner.arch`)
## Summary of changes
- Include `runner.arch` along with `runner.os` into cache keys
- Add --safekeepers option to neon_local reconfigure
- Add it to python Endpoint reconfigure
- Implement config reload in walproposer by restarting the whole bgw when
safekeeper list changes.
ref https://github.com/neondatabase/neon/issues/6341
## Problem
The fill requirement was not taken into account when looking through the
shards of a given node to fill from.
## Summary of Changes
Ensure that we do not fill a node past the recommendation from
`Scheduler::compute_fill_requirement`.
Also, modify the "neon_local timeline import" command so that it
doesn't create the endpoint any more. I don't see any reason to bundle
that in the same command, the "timeline create" and "timeline branch"
commands don't do that either.
I plan to add more tests similar to 'test_import_at_2bil', this will
help to reduce the copy-pasting.
## Problem
In https://github.com/neondatabase/neon/pull/5299, the new config-v1
tenant config file was added to hold the LocationConf type. We left the
old config file in place for forward compat, and because running without
generations (therefore without LocationConf) as still useful before the
storage controller was ready for prime-time.
Closes: https://github.com/neondatabase/neon/issues/5388
## Summary of changes
- Remove code for reading and writing the legacy config file
- Remove Generation::Broken: it was unused.
- Treat missing config file on disk as an error loading a tenant, rather
than defaulting it. We can now remove LocationConf::default, and thereby
guarantee that we never construct a tenant with a None generation.
- Update some comments + add some assertions to clarify that
Generation::None is only used in layer metadata, not in the state of a
running tenant.
- Update docker compose test to create tenants with a generation
Fixes https://github.com/neondatabase/neon/issues/6337
Add safekeeper support to switch between `Present` and
`Offloaded(flush_lsn)` states. The offloading is disabled by default,
but can be controlled using new cmdline arguments:
```
--enable-offload
Enable automatic switching to offloaded state
--delete-offloaded-wal
Delete local WAL files after offloading. When disabled, they will be left on disk
--control-file-save-interval <CONTROL_FILE_SAVE_INTERVAL>
Pending updates to control file will be automatically saved after this interval [default: 300s]
```
Manager watches state updates and detects when there are no actvity on
the timeline and actual partial backup upload in remote storage. When
all conditions are met, the state can be switched to offloaded.
In `timeline.rs` there is `StateSK` enum to support switching between
states. When offloaded, code can access only control file structure and
cannot use `SafeKeeper` to accept new WAL.
`FullAccessTimeline` is now renamed to `WalResidentTimeline`. This
struct contains guard to notify manager about active tasks requiring
on-disk WAL access. All guards are issued by the manager, all requests
are sent via channel using `ManagerCtl`. When manager receives request
to issue a guard, it unevicts timeline if it's currently evicted.
Fixed a bug in partial WAL backup, it used `term` instead of
`last_log_term` previously.
After this commit is merged, next step is to roll this change out, as in
issue #6338.
## Problem
We have seen numerous segfault and memory corruption issue for clients
using libpq and openssl 3.2.2. I don't know if this is a bug in openssl
or libpq. Downgrading to 1.1.1w fixes the issues for the storage
controller and pgbench.
## Summary of Changes:
Use openssl 1.1.1w instead of 3.2.2
I was looking for metrics on how many computes are still using protocol
version 1 and 2. This provides counters for that as "pagestream" and
"pagestream_v2" commands, but also all the other commands. The new
metrics are global for the whole pageserver instance rather than
per-tenant, so the additional metrics bloat should be fairly small.
I saw this compiler warning on my laptop:
pgxn/neon_walredo/walredoproc.c:178:10: warning: using the result of an
assignment as a condition without parentheses [-Wparentheses]
if (err = close_range_syscall(3, ~0U, 0)) {
~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
pgxn/neon_walredo/walredoproc.c:178:10: note: place parentheses around
the assignment to silence this warning
if (err = close_range_syscall(3, ~0U, 0)) {
^
( )
pgxn/neon_walredo/walredoproc.c:178:10: note: use '==' to turn this
assignment into an equality comparison
if (err = close_range_syscall(3, ~0U, 0)) {
^
==
1 warning generated.
I'm not sure what compiler version or options cause that, but it's a
good warning. Write the call a little differently, to avoid the warning
and to make it a little more clear anyway. (The 'err' variable wasn't
used for anything, so I'm surprised we were not seeing a compiler
warning on the unused value, too.)
## Problem
We install Postgres 14 in `build-tools` image, but we don't need
it. We use Postgres binaries, which we build ourselves.
## Summary of changes
- Remove Postgresql 14 installation from `build-tools` image
## Problem
Hard to debug the disconnection reason currently.
## Summary of changes
Keep track of error-direction, and therefore error source (client vs
compute) during passthrough.
A follow-up on https://github.com/neondatabase/neon/pull/8103/.
Previously, main branch fails with:
```
assertion `left == right` failed
left: b"value 3@0x10@0x30@0x28@0x40"
right: b"value 3@0x10@0x28@0x30@0x40"
```
This gets fixed after #8103 gets merged.
Signed-off-by: Alex Chi Z <chi@neon.tech>
This was a half-finished mechanism to allow a replica to enter hot
standby mode sooner, without waiting for a running-xacts record. It had
issues, and we are working on a better mechanism to replace it.
The control plane might still set the flag in the spec file, but
compute_ctl will simply ignore it.
In future we may want to run periodic tests on dedicated cloud instances
that are not GitHub action runners.
To allow these to download artifact binaries for a specific commit hash
we want to make the search by commit hash possible and prefix the S3
objects with
`artifacts/${GITHUB_SHA}/${GITHUB_RUN_ID}/${GITHUB_RUN_ATTEMPT}`
---------
Co-authored-by: Alexander Bayandin <alexander@neon.tech>
Adds manual compaction trigger; add gc compaction to test_gc_feedback
Part of https://github.com/neondatabase/neon/issues/8002
```
test_gc_feedback[debug-pg15].logical_size: 50 Mb
test_gc_feedback[debug-pg15].physical_size: 2269 Mb
test_gc_feedback[debug-pg15].physical/logical ratio: 44.5302
test_gc_feedback[debug-pg15].max_total_num_of_deltas: 7
test_gc_feedback[debug-pg15].max_num_of_deltas_above_image: 2
test_gc_feedback[debug-pg15].logical_size_after_bottom_most_compaction: 50 Mb
test_gc_feedback[debug-pg15].physical_size_after_bottom_most_compaction: 287 Mb
test_gc_feedback[debug-pg15].physical/logical ratio after bottom_most_compaction: 5.6312
test_gc_feedback[debug-pg15].max_total_num_of_deltas_after_bottom_most_compaction: 4
test_gc_feedback[debug-pg15].max_num_of_deltas_above_image_after_bottom_most_compaction: 1
```
## Summary of changes
* Add the manual compaction trigger
* Use in test_gc_feedback
* Add a guard to avoid running it with retain_lsns
* Fix: Do `schedule_compaction_update` after compaction
* Fix: Supply deltas in the correct order to reconstruct value
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
part of https://github.com/neondatabase/neon/issues/8002
## Summary of changes
This pull request adds the image layer iterator. It buffers a fixed
amount of key-value pairs in memory, and give developer an iterator
abstraction over the image layer. Once the buffer is exhausted, it will
issue 1 I/O to fetch the next batch.
Due to the Rust lifetime mysteries, the `get_stream_from` API has been
refactored to `into_stream` and consumes `self`.
Delta layer iterator implementation will be similar, therefore I'll add
it after this pull request gets merged.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
1. Proxy is retrying errors from cplane that shouldn't be retried
2. ~~Proxy is not using the retry_after_ms value~~
## Summary of changes
1. Correct the could_retry impl for ConsoleError.
2. ~~Update could_retry interface to support returning a fixed wait
duration.~~
## Problem
These APIs have been deprecated for some time, but were still used from
test code.
Closes: https://github.com/neondatabase/neon/issues/4282
## Summary of changes
- It is still convenient to do a "tenant_attach" from a test without
having to write out a location_conf body, so those test methods have
been retained with implementations that call through to their
location_conf equivalent.
Part of #7497, closes#8120.
## Summary of changes
This PR adds a metric to track the number of valid leases after `GCInfo`
gets refreshed each time.
Besides this metric, we should also track disk space and synthetic size
(after #8071 is closed) to make sure leases are used properly.
Signed-off-by: Yuchen Liang <yuchen@neon.tech>
## Problem
Storage controller runs into memory corruption issue on the drain/fill
code paths.
## Summary of changes
Update db related depdencies in the unlikely case that the issue was
fixed in diesel.
We did not recover the subtransaction IDs of prepared transactions when
starting a hot standby from a shutdown checkpoint. As a result, such
subtransactions were considered as aborted, rather than in-progress.
That would lead to hint bits being set incorrectly, and the
subtransactions suddenly becoming visible to old snapshots when the
prepared transaction was committed.
To fix, update pg_subtrans with prepared transactions's subxids when
starting hot standby from a shutdown checkpoint. The snapshots taken
from that state need to be marked as "suboverflowed", so that we also
check the pg_subtrans.
Discussion:
https://www.postgresql.org/message-id/6b852e98-2d49-4ca1-9e95-db419a2696e0%40iki.fi
NEON: cherry-picked from the upstream thread ahead of time, to unblock
https://github.com/neondatabase/neon/pull/7288. I expect this to be
committed to upstream in the next few days, superseding this. NOTE: I
did not include the new regression test on v15 and v14 branches, because
the test would need some adapting, and we don't run the perl tests on
Neon anyway.
Part of #7497, closes#8072.
## Problem
Currently the `get_lsn_by_timestamp` and branch creation pageserver APIs do not provide a pleasant client experience where the looked-up LSN might be GC-ed between the two API calls.
This PR attempts to prevent common races between GC and branch creation by making use of LSN leases provided in #8084. A lease can be optionally granted to a looked-up LSN. With the lease, GC will not touch layers needed to reconstruct all pages at this LSN for the duration of the lease.
Signed-off-by: Yuchen Liang <yuchen@neon.tech>
## Problem
- `test_storage_controller_many_tenants` can fail with warnings in the
storage controller about tenant creation holding a lock for too long,
because this test stresses the machine running the test with many
concurrent timeline creations
- `test_tenant_delete_smoke` can fail when synthetic remote storage
errors show up
## Summary of changes
- tolerate warnings about slow timeline creation in
test_storage_controller_many_tenants
- tolerate both possible errors during error_tolerant_delete
## Problem
This command was used in the very early days of sharding, before the
storage controller had anti-affinity + scheduling optimization to spread
out shards.
## Summary of changes
- Remove `storcon_cli tenant-scatter`
Part of https://github.com/neondatabase/neon/issues/8002
This pull request adds tests for bottom-most gc-compaction with delta
records. Also fixed a bug in the compaction process that creates
overlapping delta layers by force splitting at the original delta layer
boundary.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
We now have a `vX` number in the file name, i.e.,
`000000067F0000000400000B150100000000-000000067F0000000400000D350100000000__00000000014B7AC8-v1-00000001`
The related pull request for new-style path was merged a month ago
https://github.com/neondatabase/neon/pull/7660
## Summary of changes
Fixed the draw timeline dir command to handle it.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
Previously in safekeeper code, new segment file was opened without
truncate option. I don't think there is a reason to do it, this commit
replaces it with `File::create` to make it simpler and remove
`clippy::suspicious_open_options` linter warning.
## Problem
This test could occasionally trigger a "removing local file ... because
it has unexpected length log" when using the
`compact-shard-ancestors-persistent` failpoint is in use, which is
unexpected because that failpoint stops the process when the remote
metadata is in sync with local files.
It was because there are two shards on the same pageserver, and while
the one being compacted explicitly stops at the failpoint, another shard
was compacting in the background and failing at an unclean point. The
test intends to disable background compaction, but was mistakenly
revoking the value of `compaction_period` when it updated
`pitr_interval`.
Example failure:
https://neon-github-public-dev.s3.amazonaws.com/reports/pr-8123/9602976462/index.html#/testresult/7dd6165da7daef40
## Summary of changes
- Update `TENANT_CONF` in the test to use properly typed values, so that
it is usable in pageserver APIs as well as via neon_local.
- When updating tenant config with `pitr_interval`, retain the overrides
from the start of the test, so that there won't be any background
compaction going on during the test.
## Problem
`test_sharding_autosplit` is occasionally failing on warnings about
shard splits taking longer than expected (`Exclusive lock by ShardSplit
was held for`...)
It's not obvious which part is taking the time (I suspect remote storage
uploads).
Example:
https://neon-github-public-dev.s3.amazonaws.com/reports/main/9618788427/index.html#testresult/b395294d5bdeb783/
## Summary of changes
- Since shard splits are infrequent events, we can afford to be very
chatty: add a bunch of info-level logging throughout the process.
#8082 removed the legacy deletion path, but retained code for completing
deletions that were started before a pageserver restart. This PR cleans
up that remaining code, and removes all the pageserver code that dealt
with tenant deletion markers and resuming tenant deletions.
The release at https://github.com/neondatabase/neon/pull/8138 contains
https://github.com/neondatabase/neon/pull/8082, so we can now merge this
to `main`
Moves `RemoteStorageConfig` and related structs and functions into a
dedicated module. Also implements `Serialize` for the config structs
(requested in #8126).
Follow-up of #8126
## Problem
Fixes https://github.com/neondatabase/neon/issues/1287
## Summary of changes
tokio-postgres now supports arbitrary server params through the
`param(key, value)` method. Some keys are special so we explicitly
filter them out.
Adds a `Deserialize` impl to `RemoteStorageConfig`. We thus achieve the
same as #7743 but with less repetitive code, by deriving `Deserialize`
impls on `S3Config`, `AzureConfig`, and `RemoteStorageConfig`. The
disadvantage is less useful error messages.
The git history of this PR contains a state where we go via an
intermediate representation, leveraging the `serde_json` crate,
without it ever being actual json though.
Also, the PR adds deserialization tests.
Alternative to #7743 .
## Problem
While adapting the storage controller scale test to do graceful rolling
restarts via
drain and fill, I noticed that secondaries are also being rescheduled,
which, in turn,
caused the storage controller to optimise attachments.
## Summary of changes
* Introduce a transactional looking rescheduling primitive (i.e. "try to
schedule to this
secondary, but leave everything as is if you can't")
* Use it for the drain and fill stages to avoid calling into
`Scheduler::schedule` and having
secondaries move around.
## Problem
For testing, the storage controller has a built-in hack that loads
neon_local endpoint config from disk, and uses it to reconfigure
endpoints when the attached pageserver changes.
Some tests that stop an endpoint while the storage controller is running
could occasionally fail on log errors from the controller trying to use
its special test-mode calls into neon local Endpoint.
Example:
https://neon-github-public-dev.s3.amazonaws.com/reports/pr-8117/9592392425/index.html#/testresult/9d2bb8623d0d53f8
## Summary of changes
- Give NotifyError an explicit NeonLocal variant, to avoid munging these
into generic 500s (I don't want to ignore 500s in general)
- Allow-list errors related to the local notification hook.
The expectation is that tests using endpoints/workloads should be
independently checking that those endpoints work: if neon_local
generates an error inside the storage controller, that's ignorable.
## Problem
There's no way to cancel drain and fill operations.
## Summary of changes
Implement HTTP endpoints to allow cancelling of background operations.
When the operationis cancelled successfully, the node scheduling policy will revert to
`Active`.
In #7957 we enabled deletion without attachment, but retained the
old-style deletion (return 202, delete in background) for attached
tenants. In this PR, we remove the old-style deletion path, such that if
the tenant delete API is invoked while a tenant is detached, it is
simply detached before completing the deletion.
This intentionally doesn't rip out all the old deletion code: in case a
deletion was in progress at time of upgrade, we keep around the code for
finishing it for one release cycle. The rest of the code removal happens
in https://github.com/neondatabase/neon/pull/8091
Now that deletion will always be via the new path, the new path is also
updated to use some retries around remote storage operations, to
tripping up the control plane with 500s if S3 has an intermittent issue.
This is a preparation for #8022, to make the PR both backwards and
foward compatible.
This commit adds `eviction_state` field to control file. Adds support
for reading it, but writes control file in old format where possible, to
keep the disk format forward compatible.
Note: in `patch_control_file`, new field gets serialized to json like
this:
- `"eviction_state": "Present"`
- `"eviction_state": {"Offloaded": "0/8F"}`
## Problem
see https://github.com/neondatabase/neon/issues/8070
## Summary of changes
the neon_local subcommands to
- start neon
- start pageserver
- start safekeeper
- start storage controller
get a new option -t=xx or --start-timeout=xx which allows to specify a
longer timeout in seconds we wait for the process start.
This is useful in test cases where the pageserver has to read a lot of
layer data, like in pagebench test cases.
In addition we exploit the new timeout option in the python test
infrastructure (python fixtures) and modify the flaky testcase to
increase the timeout from 10 seconds to 1 minute.
Example from the test execution
```bash
RUST_BACKTRACE=1 NEON_ENV_BUILDER_USE_OVERLAYFS_FOR_SNAPSHOTS=1 DEFAULT_PG_VERSION=15 BUILD_TYPE=release ./scripts/pytest test_runner/performance/pageserver/pagebench/test_pageserver_max_throughput_getpage_at_latest_lsn.py
...
2024-06-19 09:29:34.590 INFO [neon_fixtures.py:1513] Running command "/instance_store/neon/target/release/neon_local storage_controller start --start-timeout=60s"
2024-06-19 09:29:36.365 INFO [broker.py:34] starting storage_broker to listen incoming connections at "127.0.0.1:15001"
2024-06-19 09:29:36.365 INFO [neon_fixtures.py:1513] Running command "/instance_store/neon/target/release/neon_local pageserver start --id=1 --start-timeout=60s"
2024-06-19 09:29:36.366 INFO [neon_fixtures.py:1513] Running command "/instance_store/neon/target/release/neon_local safekeeper start 1 --start-timeout=60s"
```
## Problem
These APIs have be unused for some time. They were superseded by
/location_conf: the equivalent of ignoring a tenant is now to put it in
secondary mode.
## Summary of changes
- Remove APIs
- Remove tests & helpers that used them
- Remove error variants that are no longer needed.
## Problem
The fill planner introduced in
https://github.com/neondatabase/neon/pull/8014 selects
tenant shards to promote strictly based on attached shard count load
(tenant shards on
nodes with the most attached shard counts are considered first). This
approach runs the
risk of migrating too many shards belonging to the same tenant on the
same primary node.
This is bad for availability and causes extra reconciles via the storage
controller's background
optimisations.
Also see
https://github.com/neondatabase/neon/pull/8014#discussion_r1642456241.
## Summary of changes
Refine the fill plan to avoid promoting too many shards belonging to the
same tenant on the same node.
We allow for `max(1, shard_count / node_count)` shards belonging to the
same tenant to be promoted.
## Problem
When creating a new shard the storage controller schedules via
Scheduler::schedule_shard. This does not take into account the number of
attached shards. What it does take into account is the node affinity:
when a shard is scheduled, all its nodes (primaries and secondaries) get
their affinity incremented.
For two node clusters and shards with one secondary we have a
pathological case where all primaries are scheduled on the same node.
Now that we track the count of attached shards per node, this is trivial
to fix. Still, the "proper" fix is to use the pageserver's utilization
score.
Closes https://github.com/neondatabase/neon/issues/8041
## Summary of changes
Use attached shard count when deciding which node to schedule a fresh
shard on.
part of Epic https://github.com/neondatabase/neon/issues/7386
# Motivation
The materialized page cache adds complexity to the code base, which
increases the maintenance burden and risk for subtle and hard to
reproduce bugs such as #8050.
Further, the best hit rate that we currently achieve in production is ca
1% of materialized page cache lookups for
`task_kind=PageRequestHandler`. Other task kinds have hit rates <0.2%.
Last, caching page images in Pageserver rewards under-sized caches in
Computes because reading from Pageserver's materialized page cache over
the network is often sufficiently fast (low hundreds of microseconds).
Such Computes should upscale their local caches to fit their working
set, rather than repeatedly requesting the same page from Pageserver.
Some more discussion and context in internal thread
https://neondb.slack.com/archives/C033RQ5SPDH/p1718714037708459
# Changes
This PR removes the materialized page cache code & metrics.
The infrastructure for different key kinds in `PageCache` is left in
place, even though the "Immutable" key kind is the only remaining one.
This can be further simplified in a future commit.
Some tests started failing because their total runtime was dependent on
high materialized page cache hit rates. This test makes them
fixed-runtime or raises pytest timeouts:
* test_local_file_cache_unlink
* test_physical_replication
* test_pg_regress
# Performance
I focussed on ensuring that this PR will not result in a performance
regression in prod.
* **getpage** requests: our production metrics have shown the
materialized page cache to be irrelevant (low hit rate). Also,
Pageserver is the wrong place to cache page images, it should happen in
compute.
* **ingest** (`task_kind=WalReceiverConnectionHandler`): prod metrics
show 0 percent hit rate, so, removing will not be a regression.
* **get_lsn_by_timestamp**: important API for branch creation, used by
control pane. The clog pages that this code uses are not
materialize-page-cached because they're not 8k. No risk of introducing a
regression here.
We will watch the various nightly benchmarks closely for more results
before shipping to prod.
## Problem
There are a couple of small CI cleanups that seem too small for dedicated PRs
## Summary of changes
- Create release PR with the title that matches the title in the description
- Tune error message for disallowing `ubuntu-latest` to explicitly
mention what to do
- Remove junit output from pytest, we use allure instead
## Problem
```
Unable to find image 'neondatabase/neon:9583413584' locally
docker: Error response from daemon: toomanyrequests: You have reached your pull rate limit. You may increase the limit by authenticating and upgrading: https://www.docker.com/increase-rate-limit.
```
## Summary of changes
- add `docker/login-action@v3` for `test-images` job
## Problem
Ahem, let's try this again.
https://github.com/neondatabase/neon/pull/8110 had a spooky failure in
test_multi_attach where a call to Endpoint.stop() timed out waiting for
a lock, even though we can see an earlier call completing and releasing
the lock. I suspect something weird is going on with the way pytest runs
tests across processes, or use of asyncio perhaps.
Anyway: the simplest fix is to just use a semaphore instead: if we don't
lock we can't deadlock.
## Summary of changes
- Make Endpoint.running a semaphore, where we add a unit to its counter
when starting the process and atomically decrement it when stopping.
Fixes https://github.com/neondatabase/neon/issues/7897
## Problem
`shard->delay_us` was potentially uninitialized when we connect to PS,
as it wasn't set to a non-0 value until we've first connected to the
shard's pageserver.
That caused the exponential backoff to use an initial value (multiplier)
of 0 for the first connection attempt to that pageserver, thus causing a
hot retry loop with connection attempts to the pageserver without
significant delay. That in turn caused attemmpts to reconnect to quickly
fail, rather than showing the expected 'wait until pageserver is
available' behaviour.
## Summary of changes
We initialize shard->delay_us before connection initialization if we
notice it is not initialized yet.
## Problem
`test_pageserver_max_throughput_getpage_at_latest_lsn` is a pagebench
testcase which creates several tenants/timelines to verify pageserver
performance.
The test swaps environments around in the tenant duplication stage, so
the storage controller uses two separate db instances (one in the
duplication stage and another one in the benchmarking stage).
In the benchmarking stage, the storage controller starts without any
knowledge of nodes, but with knowledge of tenants (via
attachments.json). When we re-attach and attempt to update the scheduler
stats, the scheduler rightfully complains
about the node not being known. The setup should preserve the storage
controller across the two envs, but i think it's fine to just allow list
the error in this case.
## Summary of changes
add the error message
`2024-06-19T09:38:27.866085Z ERROR Scheduler missing node 1``
to the list of allowed errors for storage_controller
## Problem
Some tasks are using around upwards of 10KB of memory at all times,
sometimes having buffers that swing them up to 30MB.
## Summary of changes
Split some of the async tasks in selective places and box them as
appropriate to try and reduce the constant memory usage. Especially in
the locations where the large future is only a small part of the total
runtime of the task.
Also, reduces the size of the CopyBuffer buffer size from 8KB to 1KB.
In my local testing and in staging this had a minor improvement. sadly
not the improvement I was hoping for :/ Might have more impact in
production
## Problem
Tests using the `Workload` helper would occasionally fail in a strange
way, where the endpoint appears to try and stop twice concurrently, and
the second stop fails because the pidfile is already gone.
`test_multi_attach` suffered from this.
Workload has a `__del__` that stops the endpoint, and python is
destroying this object in a different thread than NeonEnv.stop is
called, resulting in racing stop() calls. Endpoint has a `running`
attribute that avoids calling neon_local's stop twice, but that doesn't
help in the concurrent case.
## Summary of changes
- Make `Endpoint.stop` thread safe with a simple lock held across the
updates to `running` and the actual act of stopping it.
One could also work around this by letting Workload.endpoint outlive the
Workload, or making Workload a context manager, but this change feels
most robust, as it avoids all test code having to know that it must not
try and stop an endpoint from a destructor.
Before this PR, storage controller and broker would run in the
PWD of neon_local, i.e., most likely the checkout of neon.git.
With this PR, the shared infrastructure for background processes
sets the PWD.
Benefits:
* easy listing of processes in a repo dir using `lsof`, see added
comment in the code
* coredumps go in the right directory (next to the process)
* generally matching common expectations, I think
Changes:
* set the working directory in `background_process` module
* drive-by: fix reliance of storage_controller on NEON_REPO_DIR being
set by neon_local for the local compute hook to work correctly
## Problem
We have this set of test utilities which duplicate a tenant by copying
everything that's in remote storage and then attaching a tenant to the
pageserver and storage controller. When the "copied tenants" are created
on the storage controller, they start off from generation number 0. This
means that they can't see anything past that generation.
This issues has existed ever since generation numbers have been
introduced, but we've largely been lucky
for the generation to stay stable during the template tenant creation.
## Summary of Changes
Extend the storage controller debug attach hook to accept a generation
override. Use that in the tenant duplication logic to set the generation
number to something greater than the naturally reached generation. This
allows the tenants to see all layer files.
## Problem
Pageserver restarts cause read availablity downtime for tenants. See
`Motivation` section in the
[RFC](https://github.com/neondatabase/neon/pull/7704).
## Summary of changes
* Introduce a new `NodeSchedulingPolicy`: `PauseForRestart`
* Implement the first take of drain and fill algorithms
* Add a node status endpoint which can be polled to figure out when an
operation is done
The implementation follows the RFC, so it might be useful to peek at it
as you're reviewing.
Since the PR is rather chunky, I've made sure all commits build (with
warnings), so you can
review by commit if you prefer that.
RFC: https://github.com/neondatabase/neon/pull/7704
Related https://github.com/neondatabase/neon/issues/7387
## Problem
We want to have all released images in production ECR repository
## Summary of changes
Copy all docker images to production ECR repository
cc: https://github.com/neondatabase/cloud/issues/10177
The new image iterator and delta iterator uses an iterator-based API.
https://github.com/neondatabase/neon/pull/8006 / part of
https://github.com/neondatabase/neon/issues/8002
This requires the underlying thing (the btree) to have an iterator API,
and the iterator should have a type name so that it can be stored
somewhere.
```rust
pub struct DeltaLayerIterator {
index_iterator: BTreeIterator
}
```
versus
```rust
pub struct DeltaLayerIterator {
index_iterator: impl Stream<....>
}
```
(this requires nightly flag and still buggy in the Rust compiler)
There are multiple ways to achieve this:
1. Either write a BTreeIterator from scratch that provides `async next`.
This is the most efficient way to do that.
2. Or wrap the current `get_stream` API, which is the current approach
in the pull request.
In the future, we should do (1), and the `get_stream` API should be
refactored to use the iterator API. With (2), we have to wrap the
`get_stream` API with `Pin<Box<dyn Stream>>`, where we have the overhead
of dynamic dispatch. However, (2) needs a rewrite of the `visit`
function, which would take some time to write and review. I'd like to
define this iterator API first and work on a real iterator API later.
## Summary of changes
Add `DiskBtreeIterator` and related tests.
Signed-off-by: Alex Chi Z <chi@neon.tech>
Part of #7497, extracts from #7996, closes#8063.
## Problem
With the LSN lease API introduced in
https://github.com/neondatabase/neon/issues/7808, we want to implement
the real lease logic so that GC will
keep all the layers needed to reconstruct all pages at all the leased
LSNs with valid leases at a given time.
Signed-off-by: Yuchen Liang <yuchen@neon.tech>
- Relation size cache was moved to extension
- the changes in visibilitymap.c and freespace.c became unnecessary
with v16, thanks to changes in upstream code
- WALProposer was moved to extension
- The hack in ReadBuffer_common to not throw an error on unexpected
data beyond EOF was removed in v16 rebase. We haven't seen such
errors, so I guess that was some early issue that was fixed long
time ago.
- The ginfast.c diff was made unnecessary by upstream commit 56b662523f
- Make safekeeper read SAFEKEEPER_AUTH_TOKEN env variable with JWT
token to connect to other safekeepers.
- Set it in neon_local when auth is enabled.
- Create simple rust http client supporting it, and use it in pull_timeline
implementation.
- Enable auth in all pull_timeline tests.
- Make sk http_client() by default generate safekeeper wide token, it makes
easier enabling auth in all tests by default.
- Add /snapshot http endpoing streaming tar archive timeline contents up to
flush_lsn.
- Add check that term doesn't change, corresponding test passes now.
- Also prepares infra to hold off WAL removal during the basebackup.
- Sprinkle fsyncs to persist the pull_timeline result.
ref https://github.com/neondatabase/neon/issues/6340
## Problem
```
ERROR synthetic_size_worker: failed to calculate synthetic size for tenant ae449af30216ac56d2c1173f894b1122: Could not find size at 0/218CA70 in timeline d8da32b5e3e0bf18cfdb560f9de29638\n')
```
e.g.
https://neon-github-public-dev.s3.amazonaws.com/reports/main/9518948590/index.html#/testresult/30a6d1e2471d2775
This test had allow lists but was disrupted by
https://github.com/neondatabase/neon/pull/8051. In that PR, I had kept
an error path in fill_logical_sizes that covered the case where we
couldn't find sizes for some of the segments, but that path could only
be hit in the case that some Timeline was shut down concurrently with a
synthetic size calculation, so it makes sense to just leave the
segment's size None in this case: the subsequent size calculations do
not assume it is Some.
## Summary of changes
- Remove `CalculateSyntheticSizeError::LsnNotFound` and just proceed in
the case where we used to return it
- Remove defunct allow list entries in `test_metric_collection`
## Summary of changes
- Stop logging HealthCheck message passing at INFO level (moved to
DEBUG)
- Stop logging /status accesses at INFO (moved to DEBUG)
- Stop logging most occurances of
`missing config file "compute_ctl_temp_override.conf"`
- Log memory usage only when the data has changed significantly, or if
we've not recently logged the data, rather than always every 2 seconds.
## Problem
We don't rebuild `build-tools` image for changes in a workflow that
builds this image itself
(`.github/workflows/build-build-tools-image.yml`) or in a workflow that
determines which tag to use
(`.github/workflows/check-build-tools-image.yml`)
## Summary of changes
- Use a hash of `Dockerfile.build-tools` and workflow files as a
persistent tag instead of using a commit sha.
## Problem
A period of unavailability for all pageservers in a cluster produced the
following fallout in staging:
all tenants became detached and required manual operation to re-attach.
Manually restarting
the storage controller re-attached all tenants due to a consistency bug.
Turns out there are two related bugs which caused the issue:
1. Pageserver re-attach can be processed before the first heartbeat.
Hence, when handling
the availability delta produced by the heartbeater,
`Node::get_availability_transition` claims
that there's no need to reconfigure the node.
2. We would still attempt to reschedule tenant shards when handling
offline transitions even
if the entire cluster is down. This puts tenant shards into a state
where the reconciler believes
they have to be detached (no pageserver shows up in their intent state).
This is doubly wrong
because we don't mark the tenant shards as detached in the database,
thus causing memory vs
database consistency issues. Luckily, this bug allowed all tenant shards
to re-attach after restart.
## Summary of changes
* For (1), abuse the fact that re-attach requests do not contain an
utilisation score and use that
to differentiate from a node that replied to heartbeats.
* For (2), introduce a special case that skips any rescheduling if the
entire cluster is unavailable.
* Update the storage controller heartbeat test with an extra scenario
where the entire cluster goes
for lunch.
Fixes https://github.com/neondatabase/neon/issues/8044
cargo test (or nextest) might rebuild the binaries with different
features/flags, so do install immediately after the build. Triggered by the
particular case of nextest invocations missing $CARGO_FEATURES, which recompiled
safekeeper without 'testing' feature which made python tests needing
it (failpoints) not run in the CI.
Also add CARGO_FEATURES to the nextest runs anyway because there doesn't seem to
be an important reason not to.
## Problem
halfvec data type was introduced in pgvector 0.7.0 and is popular
because
it allows smaller vectors, smaller indexes and potentially better
performance.
So far we have not tested halfvec in our periodic performance tests.
This PR adds halfvec indexing and halfvec queries to the test.
## Problem
I've bumped `docker/setup-buildx-action` in #8042 because I wasn't able
to reproduce the issue from #7445.
But now the issue appears again in
https://github.com/neondatabase/neon/actions/runs/9514373620/job/26226626923?pr=8059
The steps to reproduce aren't clear, it required
`docker/setup-buildx-action@v3` and rebuilding the image without cache,
probably
## Summary of changes
- Downgrade `docker/setup-buildx-action@v3`
to `docker/setup-buildx-action@v2`
## Problem
rust 1.79 new enabled by default lints
## Summary of changes
* update to rust 1.79
* `s/default_features/default-features/`
* fix proxy dead code.
* fix pageserver dead code.
## Problem
This PR refactors some error handling to avoid log spam on
tenant/timeline shutdown.
- "ignoring failure to find gc cutoffs: timeline shutting down." logs
(https://github.com/neondatabase/neon/issues/8012)
- "synthetic_size_worker: failed to calculate synthetic size for tenant
...: Failed to refresh gc_info before gathering inputs: tenant shutting
down", for example here:
https://neon-github-public-dev.s3.amazonaws.com/reports/pr-8049/9502988669/index.html#suites/3fc871d9ee8127d8501d607e03205abb/1a074a66548bbcea
Closes: https://github.com/neondatabase/neon/issues/8012
## Summary of changes
- Refactor: Add a PageReconstructError variant to GcError: this is the
only kind of error that find_gc_cutoffs can emit.
- Functional change: only ignore shutdown PageReconstructError variant:
for other variants, treat it as a real error
- Refactor: add a structured CalculateSyntheticSizeError type and use it
instead of anyhow::Error in synthetic size calculations
- Functional change: while iterating through timelines gathering logical
sizes, only drop out if the whole tenant is cancelled: individual
timeline cancellations indicate deletion in progress and we can just
ignore those.
## Problem
This test could fail with a timeout waiting for tenant deletions.
Tenant deletions could get tripped up on nodes transitioning from
offline to online at the moment of the deletion. In a previous
reconciliation, the reconciler would skip detaching a particular
location because the node was offline, but then when we do the delete
the node is marked as online and can be picked as the node to use for
issuing a deletion request. This hits the "Unexpectedly still attached
path", which would still work if the caller kept calling DELETE, but if
a caller does a Delete,get,get,get poll, then it doesn't work because
the GET calls fail after we've marked the tenant as detached.
## Summary of changes
Fix the undesirable storage controller behavior highlighted by this test
failure:
- Change tenant deletion flow to _always_ wait for reconciliation to
succeed: it was unsound to proceed and return 202 if something was still
attached, because after the 202 callers can no longer GET the tenant.
Stabilize the test:
- Add a reconcile_until_idle to the test, so that it will not have
reconciliations running in the background while we mark a node online.
This test is not meant to be a chaos test: we should test that kind of
complexity elsewhere.
- This reconcile_until_idle also fixes another failure mode where the
test might see a None for a tenant location because a reconcile was
mutating it
(https://neon-github-public-dev.s3.amazonaws.com/reports/pr-7288/9500177581/index.html#suites/8fc5d1648d2225380766afde7c428d81/4acece42ae00c442/)
It remains the case that a motivated tester could produce a situation
where a DELETE gives a 500, when precisely the wrong node transitions
from offline to available at the precise moment of a deletion (but the
500 is better than returning 202 and then failing all subsequent GETs).
Note that nodes don't go through the offline state during normal
restarts, so this is super rare. We should eventually fix this by making
DELETE to the pageserver implicitly detach the tenant if it's attached,
but that should wait until nobody is using the legacy-style deletes (the
ones that use 202 + polling)
## Problem
We have some amount of outdated action in the CI pipeline, GitHub
complains about some of them.
## Summary of changes
- Update `actions/checkout@1` (a really old one) in
`vm-compute-node-image`
- Update `actions/checkout@3` in `build-build-tools-image`
- Update `docker/setup-buildx-action` in all workflows / jobs, it was
downgraded in https://github.com/neondatabase/neon/pull/7445, but it
it seems it works fine now
This failed once with `relation "test" does not exist` when trying to
run the query on the standby. It's possible that the standby is started
before the CREATE TABLE is processed in the pageserver, and the standby
opens up for queries before it has received the CREATE TABLE transaction
from the primary. To fix, wait for the standby to catch up to the
primary before starting to run the queries.
https://neon-github-public-dev.s3.amazonaws.com/reports/pr-8025/9483658488/index.html
## Problem
Some code paths during secondary mode download are returning Ok() rather
than UpdateError::Cancelled. This is functionally okay, but it means
that the end of TenantDownloader::download has a sanity check that the
progress is 100% on success, and prints a "Correcting drift..." warning
if not. This warning can be emitted in a test, e.g.
https://neon-github-public-dev.s3.amazonaws.com/reports/pr-8049/9503642976/index.html#/testresult/fff1624ba6adae9e.
## Summary of changes
- In secondary download cancellation paths, use
Err(UpdateError::Cancelled) rather than Ok(), so that we drop out of the
download function and do not reach the progress sanity check.
# Problem
Suppose our vectored get starts with an inexact materialized page cache
hit ("cached lsn") that is shadowed by a newer image layer image layer.
Like so:
```
<inmemory layers>
+-+ < delta layer
| |
-|-|----- < image layer
| |
| |
-|-|----- < cached lsn for requested key
+_+
```
The correct visitation order is
1. inmemory layers
2. delta layer records in LSN range `[image_layer.lsn,
oldest_inmemory_layer.lsn_range.start)`
3. image layer
However, the vectored get code, when it visits the delta layer, it
(incorrectly!) returns with state `Complete`.
The reason why it returns is that it calls `on_lsn_advanced` with
`self.lsn_range.start`, i.e., the layer's LSN range.
Instead, it should use `lsn_range.start`, i.e., the LSN range from the
correct visitation order listed above.
# Solution
Use `lsn_range.start` instead of `self.lsn_range.start`.
# Refs
discovered by & fixes https://github.com/neondatabase/neon/issues/6967
Co-authored-by: Vlad Lazar <vlad@neon.tech>
https://github.com/neondatabase/neon/issues/8002
We need mock WAL record to make it easier to write unit tests. This pull
request adds such a record. It has `clear` flag and `append` field. The
tests for legacy-enhanced compaction are not modified yet and will be
part of the next pull request.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
If a standby is started right after switching to a new WAL segment, the
request in the SLRU download request would point to the beginning of the
segment (e.g. 0/5000000), while the not-modified-since LSN would point
to just after the page header (e.g. 0/5000028). It's effectively the
same position, as there cannot be any WAL records in between, but the
pageserver rightly errors out on any request where the request LSN <
not-modified since LSN.
To fix, round down the not-modified since LSN to the beginning of the
page like the request LSN.
Fixes issue #8030
Some test cases add random keys into the timeline, but it is not part of
the `collect_keyspace`, this will cause compaction remove the keys.
The pull request adds a field to supply extra keyspaces during unit
tests.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
Testcase page bench test_pageserver_max_throughput_getpage_at_latest_lsn
had been deactivated because it was flaky.
We now ignore copy fail error messages like in
270d3be507/test_runner/regress/test_pageserver_getpage_throttle.py (L17-L20)
and want to reactivate it to see it it is still flaky
## Summary of changes
- reactivate the test in CI
- ignore CopyFail error message during page bench test cases
## Checklist before requesting a review
- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
## Problem
The previous code would attempt to drain to unavailable or unschedulable
nodes.
## Summary of Changes
Remove such nodes from the list of nodes to fill.
## Problem
The merging of #7818 caused the problem with the docker-compose file.
Running docker compose is now impossible due to the unavailability of
the neon-test-extensions:latest image
## Summary of changes
Fix the problem:
Add the latest tag to the neon-test-extensions image and use the
profiles feature of the docker-compose file to avoid loading the
neon-test-extensions container if it is not needed.
- Fix the dockerhub URLs
- `neondatabase/compute-node` image has been replaced with Postgres
version specific images like `neondatabase/compute-node-v16`
- Use TAG=latest in the example, rather than some old tag. That's a
sensible default for people to copy-past
- For convenience, use a Postgres connection URL in the `psql` example
that also includes the password. That way, there's no need to set up
.pgpass
- Update the image names in `docker ps` example to match what you get
when you follow the example
- Split the first and second parts of the test to two separate tests
- In the first test, disable the aggressive GC, compaction, and
autovacuum. They are only needed by the second test. I'd like to get the
first test to a point that the VM page is never all-zeros. Disabling
autovacuum in the first test is hopefully enough to accomplish that.
- Compare the full page images, don't skip page header. After fixing the
previous point, there should be no discrepancy. LSN still won't match,
though, because of commit 387a36874c.
Fixes issue https://github.com/neondatabase/neon/issues/7984
The S3 scrubber contains "S3" in its name, but we want to make it
generic in terms of which storage is used (#7547). Therefore, rename it
to "storage scrubber", following the naming scheme of already existing
components "storage broker" and "storage controller".
Part of #7547
## Problem
We need the ability to prepare a subset of storage controller managed
pageservers for decommisioning. The storage controller cannot currently
express this in terms of scheduling constraints (it's a pretty special
case, so I'm not sure it even should).
## Summary of Changes
A new `drain` command is added to `storcon_cli`. It takes a set of nodes
to drain and migrates primary attachments outside of said set. Simple
round robing assignment is used under the assumption that nodes outside
of the draining set are evenly balanced.
Note that secondary locations are not migrated. This is fine for
staging, but the migration API will have to be extended for prod in
order to allow migration of secondaries as well.
I've tested this out against a neon local cluster. The immediate use for
this command will be to migrate staging to ARM(Arch64) pageservers.
Related https://github.com/neondatabase/cloud/issues/14029
## Problem
The storage controller does not track the number of shards attached to a
given pageserver. This is a requirement for various scheduling
operations (e.g. draining and filling will use this to figure out if the
cluster is balanced)
## Summary of Changes
Track the number of shards attached to each node.
Related https://github.com/neondatabase/neon/issues/7387
A demo for a building block for compaction. The GC-compaction operation
iterates all layers below/intersect with the GC horizon, and do a full
layer rewrite of all of them. The end result will be image layer
covering the full keyspace at GC-horizon, and a bunch of delta layers
above the GC-horizon. This helps us collect the garbages of the
test_gc_feedback test case to reduce space amplification.
This operation can be manually triggered using an HTTP API or be
triggered based on some metrics. Actual method TBD.
The test is very basic and it's very likely that most part of the
algorithm will be rewritten. I would like to get this merged so that I
can have a basic skeleton for the algorithm and then make incremental
changes.
<img width="924" alt="image"
src="https://github.com/neondatabase/neon/assets/4198311/f3d49f4e-634f-4f56-986d-bfefc6ae6ee2">
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
We need unique tenant harness names in case you want to inspect the
results of the last failing run. We are not using any proc macros to get
the test name as there is no stable way of doing that, and there will
not be one in the future, so we need to fix these duplicates.
Also, clean up the duplicated tests to not mix `?` and `unwrap/assert`.
We've stored metadata as bytes within the `index_part.json` for
long fixed reasons. #7693 added support for reading out normal json
serialization of the `TimelineMetadata`.
Change the serialization to only write `TimelineMetadata` as json for
going forward, keeping the backward compatibility to reading the
metadata as bytes. Because of failure to include `alias = "metadata"` in
#7693, one more follow-up is required to make the switch from the old
name to `"metadata": <json>`, but that affects only the field name in
serialized format.
In documentation and naming, an effort is made to add enough warning
signs around TimelineMetadata so that it will receive no changes in the
future. We can add those fields to `IndexPart` directly instead.
Additionally, the path to cleaning up `metadata.rs` is documented in the
`metadata.rs` module comment. If we must extend `TimelineMetadata`
before that, the duplication suggested in [review comment] is the way to
go.
[review comment]:
https://github.com/neondatabase/neon/pull/7699#pullrequestreview-2107081558
## Problem
We need automated tests of extensions shipped with Neon to detect
possible problems.
## Summary of changes
A new image neon-test-extensions is added. Workflow changes to test the
shipped extensions are added as well.
Currently, the regression tests, shipped with extensions are in use.
Some extensions, i.e. rum, timescaledb, rdkit, postgis, pgx_ulid, pgtap,
pg_tiktoken, pg_jsonschema, pg_graphql, kq_imcx, wal2json_2_5 are
excluded due to problems or absence of internal tests.
---------
Co-authored-by: Alexander Bayandin <alexander@neon.tech>
Co-authored-by: Heikki Linnakangas <heikki@neon.tech>
The new features have deteriorated layer flushing, most recently with
#7927. Changes:
- inline `Timeline::freeze_inmem_layer` to the only caller
- carry the TimelineWriterState guard to the actual point of freezing
the layer
- this allows us to `#[cfg(feature = "testing")]` the assertion added in
#7927
- remove duplicate `flush_frozen_layer` in favor of splitting the
`flush_frozen_layers_and_wait`
- this requires starting the flush loop earlier for `checkpoint_distance
< initdb size` tests
Quite a few existing test cases create their own timelines instead of
using the default one. This pull request highlights that and hopefully
people can write simpler tests in the future.
Signed-off-by: Alex Chi Z <chi@neon.tech>
Co-authored-by: Yuchen Liang <70461588+yliang412@users.noreply.github.com>
As seen with the pgvector 0.7.0 index builds, we can receive large
batches of images, leading to very large L0 layers in the range of 1GB.
These large layers are produced because we are only able to roll the
layer after we have witnessed two different Lsns in a single
`DataDirModification::commit`. As the single Lsn batches of images can
span over multiple `DataDirModification` lifespans, we will rarely get
to write two different Lsns in a single `put_batch` currently.
The solution is to remember the TimelineWriterState instead of eagerly
forgetting it until we really open the next layer or someone else
flushes (while holding the write_guard).
Additional changes are test fixes to avoid "initdb image layer
optimization" or ignoring initdb layers for assertion.
Cc: #7197 because small `checkpoint_distance` will now trigger the
"initdb image layer optimization"
Reverts neondatabase/neon#7956
Rationale: compute incompatibilties
Slack thread:
https://neondb.slack.com/archives/C033RQ5SPDH/p1718011276665839?thread_ts=1718008160.431869&cid=C033RQ5SPDH
Relevant quotes from @hlinnaka
> If we go through with the current release candidate, but the compute
is pinned, people who create new projects will get that warning, which
is silly. To them, it looks like the ICU version was downgraded, because
initdb was run with newer version.
> We should upgrade the ICU version eventually. And when we do that,
users with old projects that use ICU will start to see that warning. I
think that's acceptable, as long as we do homework, notify users, and
communicate that properly.
> When do that, we should to try to upgrade the storage and compute
versions at roughly the same time.
A simple API to collect some statistics after compaction to easily
understand the result.
The tool reads the layer map, and analyze range by range instead of
doing single-key operations, which is more efficient than doing a
benchmark to collect the result. It currently computes two key metrics:
* Latest data access efficiency, which finds how many delta layers /
image layers the system needs to iterate before returning any key in a
key range.
* (Approximate) PiTR efficiency, as in
https://github.com/neondatabase/neon/issues/7770, which is simply the
number of delta files in the range. The reason behind that is, assume no
image layer is created, PiTR efficiency is simply the cost of collect
records from the delta layers, and the replay time. Number of delta
files (or in the future, estimated size of reads) is a simple yet
efficient way of estimating how much effort the page server needs to
reconstruct a page.
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
Due to the upcoming End of Life (EOL) for Debian 11, we need to upgrade
the base OS for Pageservers from Debian 11 to Debian 12 for security
reasons.
When deploying a new Pageserver on Debian 12 with the same binary built
on
Debian 11, we encountered the following errors:
```
could not execute operation: pageserver error, status: 500,
msg: Command failed with status ExitStatus(unix_wait_status(32512)):
/usr/local/neon/v16/bin/initdb: error while loading shared libraries:
libicuuc.so.67: cannot open shared object file: No such file or directory
```
and
```
could not execute operation: pageserver error, status: 500,
msg: Command failed with status ExitStatus(unix_wait_status(32512)):
/usr/local/neon/v14/bin/initdb: error while loading shared libraries:
libssl.so.1.1: cannot open shared object file: No such file or directory
```
These issues occur when creating new projects.
## Summary of changes
- To address these issues, we configured PostgreSQL build to use
statically linked OpenSSL and ICU libraries.
- This resolves the missing shared library errors when running the
binaries on Debian 12.
Closes: https://github.com/neondatabase/cloud/issues/12648
## Checklist before requesting a review
- [x] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [x] Do not forget to reformat commit message to not include the above
checklist
It makes them much easier to reason about, and allows other SQL tooling
to operate on them like language servers, formatters, etc.
I also brought back the removed migrations such that we can more easily
understand what they were. I included a "-- SKIP" comment describing why
those migrations are now skipped. We no longer skip migrations by
checking if it is empty, but instead check to see if the migration
starts with "-- SKIP".
## Problem
We don't carry run-* labels from external contributors' PRs to
ci-run/pr-* PRs. This is not really convenient.
Need to sync labels in approved-for-ci-run workflow.
## Summary of changes
Added the procedure of transition of labels from the original PR
## Checklist before requesting a review
- [x] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
---------
Co-authored-by: Alexander Bayandin <alexander@neon.tech>
In #7927 I needed to fix this test case, but the fixes should be
possible to land irrespective of the layer ingestion code change.
The most important fix is the behavior if an image layer is found: the
assertion message formatting raises a runtime error, which obscures the
fact that we found an image layer.
We had a random sleep in the beginning of partial backup task, which was
needed for the first partial backup deploy. It helped with gradual
upload of segments without causing network overload. Now partial backup
is deployed everywhere, so we don't need this random sleep anymore.
We also had an issue related to this, in which manager task was not shut
down for a long time. The cause of the issue is this random sleep that
didn't take timeline cancellation into account, meanwhile manager task
waited for partial backup to complete.
Fixes https://github.com/neondatabase/neon/issues/7967
currently we warn even by going over a single byte. even that will be
hit much more rarely once #7927 lands, but get this in earlier.
rationale for 2*checkpoint_distance: anything smaller is not really
worth a warn.
we have an global allowed_error for this warning, which still cannot be
removed nor can it be removed with #7927 because of many tests with very
small `checkpoint_distance`.
M-series macOS has different alignments/size for some fields (which I
did not investigate in detail) and therefore this test cannot pass on
macOS. Fixed by using `<=` for the comparison so that we do not test for
an exact match.
observed by @yliang412
Signed-off-by: Alex Chi Z <chi@neon.tech>
This adds retries to the bulk deletion, because if there is a certain
chance n that a request fails, the chance that at least one of the
requests in a chain of requests fails increases exponentially.
We've had similar issues with the S3 DR tests, which in the end yielded
in adding retries at the remote_storage level. Retries at the top level
are not sufficient when one remote_storage "operation" is multiple
network requests in a trench coat, especially when there is no notion of
saving the progress: even if prior deletions had been successful, we'd
still need to get a 404 in order to continue the loop and get to the
point where we failed in the last iteration. Maybe we'll fail again but
before we've even reached it.
Retries at the bottom level avoid this issue because they have the
notion of progress and also when one network operation fails, only that
operation is retried.
First part of #7931.
Related to #7341 tenant deletion will end up shutting down timelines
twice, once before actually starting and the second time when per
timeline deletion is requested. Shutting down TimelineMetrics causes
underflows. Add an atomic boolean and only do the shutdown once.
Closes#7406.
## Problem
When a `get_lsn_by_timestamp` request is cancelled, an anyhow error is
exposed to handle that case, which verbosely logs the error. However, we
don't benefit from having the full backtrace provided by anyhow in this
case.
## Summary of changes
This PR introduces a new `ApiError` type to handle errors caused by
cancelled request more robustly.
- A new enum variant `ApiError::Cancelled`
- Currently the cancelled request is mapped to status code 500.
- Need to handle this error in proxy's `http_util` as well.
- Added a failpoint test to simulate cancelled `get_lsn_by_timestamp`
request.
Signed-off-by: Yuchen Liang <yuchen@neon.tech>
## Problem
As described in #7952, the controller's attempt to reconcile a tenant
before finally deleting it can get hung up waiting for the compute
notification hook to accept updates.
The fact that we try and reconcile a tenant at all during deletion is
part of a more general design issue (#5080), where deletion was
implemented as an operation on attached tenant, requiring the tenant to
be attached in order to delete it, which is not in principle necessary.
Closes: #7952
## Summary of changes
- In the pageserver deletion API, only do the traditional deletion path
if the tenant is attached. If it's secondary, then tear down the
secondary location, and then do a remote delete. If it's not attached at
all, just do the remote delete.
- In the storage controller, instead of ensuring a tenant is attached
before deletion, do a best-effort detach of the tenant, and then call
into some arbitrary pageserver to issue a deletion of remote content.
The pageserver retains its existing delete behavior when invoked on
attached locations. We can remove this later when all users of the API
are updated to either do a detach-before-delete. This will enable
removing the "weird" code paths during startup that sometimes load a
tenant and then immediately delete it, and removing the deletion markers
on tenants.
## Problem
Currently we serialize the `TimelineMetadata` into bytes to put it into
`index_part.json`. This `Vec<u8>` (hopefully `[u8; 512]`) representation
was chosen because of problems serializing TimelineId and Lsn between
different serializers (bincode, json). After #5335, the serialization of
those types became serialization format aware or format agnostic.
We've removed the pageserver local `metadata` file writing in #6769.
## Summary of changes
Allow switching from the current serialization format to plain JSON for
the legacy TimelineMetadata format in the future by adding a competitive
serialization method to the current one
(`crate::tenant::metadata::modern_serde`), which accepts both old bytes
and new plain JSON.
The benefits of this are that dumping the index_part.json with pretty
printing no longer produces more than 500 lines of output, but after
enabling it produces lines only proportional to the layer count, like:
```json
{
"version": ???,
"layer_metadata": { ... },
"disk_consistent_lsn": "0/15FD5D8",
"legacy_metadata": {
"disk_consistent_lsn": "0/15FD5D8",
"prev_record_lsn": "0/15FD5A0",
"ancestor_timeline": null,
"ancestor_lsn": "0/0",
"latest_gc_cutoff_lsn": "0/149FD18",
"initdb_lsn": "0/149FD18",
"pg_version": 15
}
}
```
In the future, I propose we completely stop using this legacy metadata
type and wasting time trying to come up with another version numbering
scheme in addition to the informative-only one already found in
`index_part.json`, and go ahead with storing metadata or feature flags
on the `index_part.json` itself.
#7699 is the "one release after" changes which starts to produce
metadata in the index_part.json as json.
fixes https://github.com/neondatabase/neon/issues/7790 (duplicating most
of the issue description here for posterity)
# Background
From the time before always-authoritative `index_part.json`, we had to
handle duplicate layers. See the RFC for an illustration of how
duplicate layers could happen:
a8e6d259cb/docs/rfcs/027-crash-consistent-layer-map-through-index-part.md (L41-L50)
As of #5198 , we should not be exposed to that problem anymore.
# Problem 1
We still have
1. [code in
Pageserver](82960b2175/pageserver/src/tenant/timeline.rs (L4502-L4521))
than handles duplicate layers
2. [tests in the test
suite](d9dcbffac3/test_runner/regress/test_duplicate_layers.py (L15))
that demonstrates the problem using a failpoint
However, the test in the test suite doesn't use the failpoint to induce
a crash that could legitimately happen in production.
What is does instead is to return early with an `Ok()`, so that the code
in Pageserver that handles duplicate layers (item 1) actually gets
exercised.
That "return early" would be a bug in the routine if it happened in
production.
So, the tests in the test suite are tests for their own sake, but don't
serve to actually regress-test any production behavior.
# Problem 2
Further, if production code _did_ (it nowawdays doesn't!) create a
duplicate layer, the code in Pageserver that handles the condition (item
1 above) is too little and too late:
* the code handles it by discarding the newer `struct Layer`; that's
good.
* however, on disk, we have already overwritten the old with the new
layer file
* the fact that we do it atomically doesn't matter because ...
* if the new layer file is not bit-identical, then we have a cache
coherency problem
* PS PageCache block cache: caches old bit battern
* blob_io offsets stored in variables, based on pre-overwrite bit
pattern / offsets
* => reading based on these offsets from the new file might yield
different data than before
# Solution
- Remove the test suite code pertaining to Problem 1
- Move & rename test suite code that actually tests RFC-27
crash-consistent layer map.
- Remove the Pageserver code that handles duplicate layers too late
(Problem 1)
- Use `RENAME_NOREPLACE` to prevent over-rename the file during
`.finish()`, bail with an error if it happens (Problem 2)
- This bailing prevents the caller from even trying to insert into the
layer map, as they don't even get a `struct Layer` at hand.
- Add `abort`s in the place where we have the layer map lock and check
for duplicates (Problem 2)
- Note again, we can't reach there because we bail from `.finish()` much
earlier in the code.
- Share the logic to clean up after failed `.finish()` between image
layers and delta layers (drive-by cleanup)
- This exposed that test `image_layer_rewrite` was overwriting layer
files in place. Fix the test.
# Future Work
This PR adds a new failure scenario that was previously "papered over"
by the overwriting of layers:
1. Start a compaction that will produce 3 layers: A, B, C
2. Layer A is `finish()`ed successfully.
3. Layer B fails mid-way at some `put_value()`.
4. Compaction bails out, sleeps 20s.
5. Some disk space gets freed in the meantime.
6. Compaction wakes from sleep, another iteration starts, it attempts to
write Layer A again. But the `.finish()` **fails because A already
exists on disk**.
The failure in step 5 is new with this PR, and it **causes the
compaction to get stuck**.
Before, it would silently overwrite the file and "successfully" complete
the second iteration.
The mitigation for this is to `/reset` the tenant.
## Problem
This was an oversight when adding heatmaps: because they are at the top
level of the tenant, they aren't included in the catch-all list & delete
that happens for timeline paths.
This doesn't break anything, but it leaves behind a few kilobytes of
garbage in the S3 bucket after a tenant is deleted, generating work for
the scrubber.
## Summary of changes
- During deletion, explicitly remove the heatmap file
- In test_tenant_delete_smoke, upload a heatmap so that the test would
fail its "remote storage empty after delete" check if we didn't delete
it.
RemoteTimelineClient maintains a copy of "next IndexPart" as a number of
fields which are like an IndexPart but this is not immediately obvious.
Instead of multiple fields, maintain a `dirty` ("next IndexPart") and
`clean` ("uploaded IndexPart") fields.
Additional cleanup:
- rename `IndexPart::disk_consistent_lsn` accessor
`duplicated_disk_consistent_lsn`
- no one except scrubber should be looking at it, even scrubber is a
stretch
- remove usage elsewhere (pagectl used by tests, metadata scan endpoint)
- serialize index part *before* the index upload operation
- avoid upload operation being retried because of serialization error
- serialization error is fatal anyway for timeline -- it can only make
transient local progress after that, at least the error is bubbled up
now
- gather exploded IndexPart fields into single actual
`UploadQueueInitialized::dirty` of which the uploaded snapshot is
serialized
- implement the long wished monotonicity check with the `clean`
IndexPart with an assertion which is not expected to fire
Continued work from #7860 towards next step of #6994.
in build logs we get a lot of lines for building the compute node images
because of verbose tar unpack. we know the sha256 so we don't need to
log the contents. my hope is that this will allow us more reliably use
the github live updating log view.
I suspected a wakeup could be lost with
`remote_storage::support::DownloadStream` if the cancellation and inner
stream wakeups happen simultaneously. The next poll would only return
the cancellation error without setting the wakeup. There is no lost
wakeup because the single future for getting the cancellation error is
consumed when the value is ready, and a new future is created for the
*next* value. The new future is always polled. Similarly, if only the
`Stream::poll_next` is being used after a `Some(_)` value has been
yielded, it makes no sense to have an expectation of a wakeup for the
*(N+1)th* stream value already set because when a value is wanted,
`Stream::poll_next` will be called.
A test is added to show that the above is true.
Additionally, there was a question of these cancellations and timeouts
flowing to attached or secondary tenant downloads. A test is added to
show that this, in fact, happens.
Lastly, a warning message is logged when a download stream is polled
after a timeout or cancellation error (currently unexpected) so we can
rule it out while troubleshooting.
## Problem
Page LSN is not set while VM update.
May be reason of test_vm_bits flukyness.
Buit more serious issues can be also caused by wrong LSN.
Related: https://github.com/neondatabase/neon/pull/7935
## Summary of changes
- In `apply_in_neon`, set the LSN bytes when applying records of type
`ClearVisibilityMapFlags`
## Problem
There was a bug in dynamic rate limiter, which exhausted CPU in proxy
and proxy wasn't able to accept any connections.
## Summary of changes
1. `if self.available > 1` -> `if self.available >= 1`
2. remove `timeout_at` to use just timeout
3. remove potential infinite loops which can exhaust CPUs.
## Problem
CreateImageLayersError and CompactionError had proper From
implementations, but compact_legacy was explicitly squashing all image
layer errors into an anyhow::Error anyway.
This led to errors like:
```
Error processing HTTP request: InternalServerError(timeline shutting down
Stack backtrace:
0: <<anyhow::Error as core::convert::From<pageserver::tenant::timeline::CreateImageLayersError>>::from as core::ops::function::FnOnce<(pageserver::tenant::timeline::CreateImageLayersError,)>>::call_once
at /rustc/9b00956e56009bab2aa15d7bff10916599e3d6d6/library/core/src/ops/function.rs:250:5
1: <core::result::Result<alloc::vec::Vec<pageserver::tenant::storage_layer::layer::ResidentLayer>, pageserver::tenant::timeline::CreateImageLayersError>>::map_err::<anyhow::Error, <anyhow::Error as core::convert::From<pageserver::tenant::timeline::CreateImageLayersError>>::from>
at /rustc/9b00956e56009bab2aa15d7bff10916599e3d6d6/library/core/src/result.rs:829:27
2: <pageserver::tenant::timeline::Timeline>::compact_legacy::{closure#0}
at pageserver/src/tenant/timeline/compaction.rs:125:36
3: <pageserver::tenant::timeline::Timeline>::compact::{closure#0}
at pageserver/src/tenant/timeline.rs:1719:84
4: pageserver::http::routes::timeline_checkpoint_handler::{closure#0}::{closure#0}
```
Closes: https://github.com/neondatabase/neon/issues/7861
The storage controller has 'drop' APIs for tenants and nodes, for use in
situations where something weird has happened:
- node-drop is useful until we implement proper node decom, or if we
have a partially provisioned node that somehow gets registered with the
storage controller but is then dead.
- tenant-drop is useful if we accidentally add a tenant that shouldn't
be there at all, or if we want to make the controller forget about a
tenant without deleting its data. For example, if one uses the
tenant-warmup command with a bad tenant ID and needs to clean that up.
The drop commands require an `--unsafe` parameter, to reduce the chance
that someone incorrectly assumes these are the normal/clean ways to
delete things.
This PR also adds a convenience command for setting the time based
eviction parameters on a tenant. This is useful when onboarding an
existing tenant that has high resident size due to storage amplification
in compaction: setting a lower time based eviction threshold brings down
the resident size ahead of doing a shard split.
Store logical replication origin in KV storage
## Problem
See #6977
## Summary of changes
* Extract origin_lsn from commit WAl record
* Add ReplOrigin key to KV storage and store origin_lsn
* In basebackup replace snapshot origin_lsn with last committed
origin_lsn at basebackup LSN
## Checklist before requesting a review
- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
Co-authored-by: Alex Chi Z <chi@neon.tech>
## Problem
Currently, we leave `index_part.json` objects from old generations
behind each time a pageserver restarts or a tenant is migrated. This
doesn't break anything, but it's annoying when a tenant has been around
for a long time and starts to accumulate 10s-100s of these.
Partially implements: #7043
## Summary of changes
- Add a new `pageserver-physical-gc` command to `s3_scrubber`
The name is a bit of a mouthful, but I think it makes sense:
- GC is the accurate term for what we are doing here: removing data that
takes up storage but can never be accessed.
- "physical" is a necessary distinction from the "normal" GC that we do
online in the pageserver, which operates at a higher level in terms of
LSNs+layers, whereas this type of GC is purely about S3 objects.
- "pageserver" makes clear that this command deals exclusively with
pageserver data, not safekeeper.
The keyspace utils like `is_rel_size_key` or `is_rel_fsm_block_key` and
many others are free functions and have to be either imported separately
or specified with the full path starting in `pageserver_api:🔑:`.
This is less convenient than if these functions were just inherent
impls.
Follow-up of #7890Fixes#6438
when running the regress tests locally without any environment variables
we use on CI, `test_pageserver_compaction_smoke` fails with division by
zero. fix it temporarily by allowing no vectored read happening. to be
cleaned when vectored get validation gets removed and the default value
can be changed.
Cc: https://github.com/neondatabase/neon/issues/7381
In issue #5590 it was proposed to implement metrics for Azure blob
storage. This PR implements them except for the part that performs the
rename, which is left for a followup.
Closes#5590
The general partial backup idea is that each safekeeper keeps only one
partial segment in remote storage at a time. Sometimes this is not true,
for example if we uploaded object to S3 but got an error when tried to
remove the previous upload. In this case we still keep a list of all
potentially uploaded objects in safekeeper state.
This commit prints a warning to logs if there is too many objects in
safekeeper state. This is not expected and we should try to fix this
state, we can do this by running gc.
I haven't seen this being an issue anywhere, but printing a warning is
something that I wanted to do and forgot in initial PR.
## Problem
- Because GC exposes all errors as an anyhow::Error, we have
intermittent issues with spurious log errors during shutdown, e.g. in
this failure of a performance test
https://neon-github-public-dev.s3.amazonaws.com/reports/main/9300804302/index.html#suites/07874de07c4a1c9effe0d92da7755ebf/214a2154f6f0217a/
```
Gc failed 1 times, retrying in 2s: shutting down
```
GC really doesn't do a lot of complicated IO: it doesn't benefit from
the backtrace capabilities of anyhow::Error, and can be expressed more
robustly as an enum.
## Summary of changes
- Add GcError type and use it instead of anyhow::Error in GC functions
- In `gc_iteration_internal`, return GcError::Cancelled on shutdown
rather than Ok(()) (we only used Ok before because we didn't have a
clear cancellation error variant to use).
- In `gc_iteration_internal`, skip past timelines that are shutting
down, to avoid having to go through another GC iteration if we happen to
see a deleting timeline during a GC run.
- In `refresh_gc_info_internal`, avoid an error case where a timeline
might not be found after being looked up, by carrying an Arc<Timeline>
instead of a TimelineId between the first loop and second loop in the
function.
- In HTTP request handler, handle Cancelled variants as 503 instead of
turning all GC errors into 500s.
## Problem
In all cases, AncestorStopping is equivalent to Cancelled.
This became more obvious in
https://github.com/neondatabase/neon/pull/7912#discussion_r1620582309
when updating these error types.
## Summary of changes
- Remove AncestorStopping, always use Cancelled instead
This pull request adds necessary interfaces to deterministically create
scenarios we want to test. Simplify some test cases to use this
interface to make it stable + reproducible.
Compaction test will be able to use this interface. Also the upcoming
delete tombstone tests will use this interface to make test
reproducible.
## Summary of changes
* `force_create_image_layer`
* `force_create_delta_layer`
* `force_advance_lsn`
* `create_test_timeline_with_states`
* `branch_timeline_test_with_states`
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
This is a preparation for
https://github.com/neondatabase/neon/issues/6337.
The idea is to add FullAccessTimeline, which will act as a guard for
tasks requiring access to WAL files. Eviction will be blocked on these
tasks and WAL won't be deleted from disk until there is at least one
active FullAccessTimeline.
To get FullAccessTimeline, tasks call `tli.full_access_guard().await?`.
After eviction is implemented, this function will be responsible for
downloading missing WAL file and waiting until the download finishes.
This commit also contains other small refactorings:
- Separate `get_tenant_dir` and `get_timeline_dir` functions for
building a local path. This is useful for looking at usages and finding
tasks requiring access to local filesystem.
- `timeline_manager` is now responsible for spawning all background
tasks
- WAL removal task is now spawned instantly after horizon is updated
## Problem
- Initial size calculations tend to fail with `Bad state (not active)`
Closes: https://github.com/neondatabase/neon/issues/7911
## Summary of changes
- In `wait_lsn`, return WaitLsnError::Cancelled rather than BadState
when the state is Stopping
- Replace PageReconstructError's `Other` variant with a specific
`BadState` variant
- Avoid returning anyhow::Error from get_ready_ancestor_timeline -- this
was only used for the case where there was no ancestor. All callers of
this function had implicitly checked that the ancestor timeline exists
before calling it, so they can pass in the ancestor instead of handling
an error.
epoch is a historical and potentially confusing name. It semantically means
lastLogTerm from the raft paper, so let's use it.
This commit changes only internal namings, not public interface (http).
## Problem
Looking at several noisy shutdown logs:
- In https://github.com/neondatabase/neon/issues/7861 we're hitting a
log error with `InternalServerError(timeline shutting down\n'` on the
checkpoint API handler.
- In the field, we see initial_logical_size_calculation errors on
shutdown, via DownloadError
- In the field, we see errors logged from layer download code
(independent of the error propagated) during shutdown
Closes: https://github.com/neondatabase/neon/issues/7861
## Summary of changes
The theme of these changes is to avoid propagating anyhow::Errors for
cases that aren't really unexpected error cases that we might want a
stacktrace for, and avoid "Other" error variants unless we really do
have unexpected error cases to propagate.
- On the flush_frozen_layers path, use the `FlushLayerError` type
throughout, rather than munging it into an anyhow::Error. Give
FlushLayerError an explicit from_anyhow helper that checks for timeline
cancellation, and uses it to give a Cancelled error instead of an Other
error when the timeline is shutting down.
- In logical size calculation, remove BackgroundCalculationError (this
type was just a Cancelled variant and an Other variant), and instead use
CalculateLogicalSizeError throughout. This can express a
PageReconstructError, and has a From impl that translates cancel-like
page reconstruct errors to Cancelled.
- Replace CalculateLogicalSizeError's Other(anyhow::Error) variant case
with a Decode(DeserializeError) variant, as this was the only kind of
error we actually used in the Other case.
- During layer download, drop out early if the timeline is shutting
down, so that we don't do an `error!()` log of the shutdown error in
this case.
Updates the `tokio-epoll-uring` dependency.
There is [only one change](342ddd197a...08ccfa94ff),
the adoption of linux-raw-sys for `statx` instead of using libc.
Part of #7889.
Consider the following sequence of migration:
```
1. user starts compute
2. force migrate to v2
3. user continues to write data
```
At the time of (3), the compute node is not aware that the page server
does not contain replication states any more, and might continue to
ingest neon-file records into the safekeeper. This will leave the
pageserver store a partial replication state and cause some errors. For
example, the compute could issue a deletion of some aux files in v1, but
this file does not exist in v2. Therefore, we should ignore all these
errors until everyone is migrated to v2.
Also note that if we see this warning in prod, it is likely because we
did not fully suspend users' compute when flipping the v1/v2 flag.
Signed-off-by: Alex Chi Z <chi@neon.tech>
Perf shows a significant amount of time is spent on `Keyspace::merge`.
This pull request postpones merging keyspace until retrieving the layer,
which contributes to a 30x improvement in aux keyspace basebackup time.
```
--- old
10000 files found in 0.580569459s
--- new
10000 files found in 0.02995075s
```
Signed-off-by: Alex Chi Z <chi@neon.tech>
field2 of metadata keys can be 0xFFFF because of the mapping. Allow
0xFFFF for `to_i128`. An alternative is to encode 0xFFFF as 0xFFFFFFFF
(which is allowed in the original `to_i128`). But checking the places
where field2 is referenced, the rest part of the system does not seem to
depend on this assertion.
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
#7371
## Summary of changes
* The VirtualFile::open, open_with_options, and create methods use
AsRef, similar to the standard library's std::fs APIs.
## Problem
proxy params being a `HashMap<String,String>` when it contains just
```
application_name: psql
database: neondb
user: neondb_owner
```
is quite wasteful allocation wise.
## Summary of changes
Keep the params in the wire protocol form, eg:
```
application_name\0psql\0database\0neondb\0user\0neondb_owner\0
```
Using a linear search for the map is fast enough at small sizes, which
is the normal case.
## Problem
We were rate limiting wake_compute in the wrong place
## Summary of changes
Move wake_compute rate limit to after the permit is acquired. Also makes
a slight refactor on normalize, as it caught my eye
## Problem
We use ubuntu-latest as a default OS for running jobs. It can cause
problems due to instability, so we should use the LTS version of Ubuntu.
## Summary of changes
The image ubuntu-latest was changed with ubuntu-22.04 in workflows.
## Checklist before requesting a review
- [x] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
## Problem
See https://github.com/neondatabase/cloud/issues/10845
## Summary of changes
Do not report error if GIN page is not restored
## Checklist before requesting a review
- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
---------
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
## Problem
Computes that are healthy can manage many connection attempts at a time.
Unhealthy computes cannot. We initially handled this with a fixed
concurrency limit, but it seems this inhibits pgbench.
## Summary of changes
Support AIMD for connect_to_compute lock to allow varying the
concurrency limit based on compute health
Get rid of postgres-native-tls and openssl in favour of rustls in our
dependency tree.
Do further steps to completely remove native-tls and openssl.
Among other advantages, this allows us to do static musl builds more
easily: #7889
## Problem
In 4ce6e2d2fc we added a warning when progress stats don't look right at
the end of a secondary download pass.
This `Correcting drift in progress stats` warning fired in staging on a
pageserver that had been doing some disk usage eviction.
The impact is low because in the same place we log the warning, we also
fix up the progress values.
## Summary of changes
- When we skip downloading a layer because it was recently evicted,
update the progress stats to ensure they still reach a clean complete
state at the end of a download pass.
- Also add a log for evicting secondary location layers, for symmetry
with attached locations, so that we can clearly see when eviction has
happened for a particular tenant's layers when investigating issues.
This is a point fix -- the code would also benefit from being refactored
so that there is some "download result" type with a Skip variant, to
ensure that we are updating the progress stats uniformly for those
cases.
## Problem
We want to regularly verify the performance of pgvector HNSW parallel
index builds and parallel similarity search using HNSW indexes.
The first release that considerably improved the index-build parallelism
was pgvector 0.7.0 and we want to make sure that we do not regress by
our neon compute VM settings (swap, memory over commit, pg conf etc.)
## Summary of changes
Prepare a Neon project with 1 million openAI vector embeddings (vector
size 1536).
Run HNSW indexing operations in the regression test for the various
distance metrics.
Run similarity queries using pgbench with 100 concurrent clients.
I have also added the relevant metrics to the grafana dashboards pgbench
and olape
---------
Co-authored-by: Alexander Bayandin <alexander@neon.tech>
## Problem
After [0e4f182680] which introduce async
connect
Neon is not able to connect to page server.
## Summary of changes
Perform sync commit at MacOS/X
## Checklist before requesting a review
- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
---------
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
Do pull_timeline while WAL is being removed. To this end
- extract pausable_failpoint to utils, sprinkle pull_timeline with it
- add 'checkpoint' sk http endpoint to force WAL removal.
After fixing checking for pull file status code test fails so far which is
expected.
## Problem
Seems the websocket buffering was broken for large query responses only
## Summary of changes
Move buffering until after the underlying stream is ready.
Tested locally confirms this fixes the bug.
Also fixes the pg-sni-router missing metrics bug
## Problem
A title for automatic proxy release PRs is `Proxy release`, and for
storage & compute, it's just `Release`
## Summary of changes
- Amend PR title for Storage & Compute releases to "Storage & Compute
release"
## Problem
Ongoing hunt for secondary location shutdown hang issues.
## Summary of changes
- Revert the functional changes from #7675
- Tweak a log in secondary downloads to make it more apparent when we
drop out on cancellation
- Modify DownloadStream's behavior to always return an Err after it has
been cancelled. This _should_ not impact anything, but it makes the
behavior simpler to reason about (e.g. even if the poll function somehow
got called again, it could never end up in an un-cancellable state)
Related #https://github.com/neondatabase/cloud/issues/13576
## Problem
- After a shard split of a large existing tenant, child tenants can end
up with oversized historic layers indefinitely, if those layers are
prevented from being GC'd by branchpoints.
This PR follows https://github.com/neondatabase/neon/pull/7531, and adds
rewriting of layers that contain a mixture of needed & un-needed
contents, in addition to dropping un-needed layers.
Closes: https://github.com/neondatabase/neon/issues/7504
## Summary of changes
- Add methods to ImageLayer for reading back existing layers
- Extend `compact_shard_ancestors` to rewrite layer files that contain a
mixture of keys that we want and keys we do not, if unwanted keys are
the majority of those in the file.
- Amend initialization code to handle multiple layers with the same
LayerName properly
- Get rid of of renaming bad layer files to `.old` since that's now
expected on restarts during rewrites.
## Problem
One database is too limiting. We have agreed to raise this limit to 10.
## Checklist before requesting a review
- [x] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
* Make PS connection startup use async APIs
This allows for improved query cancellation when we start connections
* Make PS connections have per-shard connection retry state.
Previously they shared global backoff state, which is bad for quickly
getting all connections started and/or back online.
* Make sure we clean up most connection state on failed connections.
Previously, we could technically leak some resources that we'd otherwise
clean up. Now, the resources are correctly cleaned up.
* pagestore_smgr.c now PANICs on unexpected response message types.
Unexpected responses are likely a symptom of having a desynchronized
view of the connection state. As a desynchronized connection state can
cause corruption, we PANIC, as we don't know what data may have been
written to buffers: the only solution is to fail fast & hope we didn't
write wrong data.
* Catch errors in sync pagestream request handling.
Previously, if a query was cancelled after a message was sent to
the pageserver, but before the data was received, the backend
could forget that it sent the synchronous request, and let others
deal with the repercussions. This could then lead to incorrect
responses, or errors such as "unexpected response from page
server with tag 0x68"
## Problem
## Summary of changes
## Checklist before requesting a review
- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
## Checklist before merging
- [ ] Do not forget to reformat commit message to not include the above
checklist
Once upon a time, we used to have duplicated types for runtime IndexPart
and whatever we stored. Because of the serde fixes in #5335 we have no
need for duplicated IndexPart type anymore, but the `IndexLayerMetadata`
stayed.
- remove the type
- remove LayerFileMetadata::file_size() in favor of direct field access
Split off from #7833. Cc: #3072.
* Reduce the logging level for create image layers of metadata keys.
(question: is it possible to adjust logging levels at runtime?)
* Do a info logging of image layers only after the layer is created. Now
there are a lot of cases where we create the image layer writer but then
discarding that image layer because it does not contain any key.
Therefore, I changed the new image layer logging to trace, and create
image layer logging to info.
Signed-off-by: Alex Chi Z <chi@neon.tech>
Reduces duplication between tiered and legacy compaction by using the
`Timeline::create_image_layer_for_rel_blocks` function. This way, we
also use vectored get in tiered compaction, so the change has two
benefits in one.
fixes#7659
---------
Co-authored-by: Alex Chi Z. <iskyzh@gmail.com>
The list timeline API gives something like
`"wal_source_connstr":"PgConnectionConfig { host:
Domain(\"safekeeper-5.us-east-2.aws.neon.build\"), port: 6500, password:
Some(REDACTED-STRING) }"`, which is weird. This pull request makes it
somehow like a connection string. This field is not used at least in the
neon database, so I assume no one is reading or parsing it.
Signed-off-by: Alex Chi Z <chi@neon.tech>
We'd like to get some bits reserved in the length field of image layers
for future usage (compression). This PR bases on the assumption that we
don't have any blobs that require more than 28 bits (3 bytes + 4 bits)
to store the length, but as a preparation, before erroring, we want to
first emit warnings as if the assumption is wrong, such warnings are less
disruptive than errors.
A metric would be even less disruptive (log messages are more slow, if
we have a LOT of such large blobs then it would take a lot of time to
print them). At the same time, likely such 256 MiB blobs will occupy an
entire layer file, as they are larger than our target size. For layer
files we already log something, so there shouldn't be a large increase
in overhead.
Part of #5431
## Problem
By default, pgvector compiles with `-march=native` on some platforms for
best performance. However, this can lead to `Illegal instruction` errors
if trying to run the compiled extension on a different machine.
I had this problem when trying to run the Neon compute docker image on
MacOS with Apple Silicon with Rosetta.
see
ff9b22977e/README.md (L1021)
## Summary of changes
Pass OPTFLAGS="" to make.
I looked at the metrics from
https://github.com/neondatabase/neon/pull/7768 on staging and it seems
that manager does too many iterations. This is probably caused by
background job `remove_wal.rs` which iterates over all timelines and
tries to remove WAL and persist control file. This causes shared state
updates and wakes up the manager. The fix is to skip notifying about the
updates if nothing was updated.
## Problem
We've seen some strange behaviors when doing lots of migrations
involving secondary locations. One of these was where a tenant was
apparently stuck in the `Scheduler::running` list, but didn't appear to
be making any progress. Another was a shutdown hang
(https://github.com/neondatabase/cloud/issues/13576).
## Summary of changes
- Fix one issue (probably not the only one) where a tenant in the
`pending` list could proceed to `spawn` even if the same tenant already
had a running task via `handle_command` (this could have resulted in a
weird value of SecondaryProgress)
- Add various extra logging:
- log before as well as after layer downloads so that it would be
obvious if we were stuck in remote storage code (we shouldn't be, it has
built in timeouts)
- log the number of running + pending jobs from the scheduler every time
it wakes up to do a scheduling iteration (~10s) -- this is quite chatty,
but not compared with the volume of logs on a busy pageserver. It should
give us confidence that the scheduler loop is still alive, and
visibility of how many tasks the scheduler thinks are running.
2024-05-23 09:13:55 +01:00
681 changed files with 63764 additions and 25176 deletions
if ${PG_BINARIES}/psql "${{ env.BENCHMARK_CONNSTR }}" -tAc "SELECT 1 FROM benchmark_restore_status WHERE databasename='${{ env.DATABASE_NAME }}' AND restore_done=true;" | grep -q 1; then
echo "Restore already done for database ${{ env.DATABASE_NAME }} on platform ${{ env.PLATFORM }}. Skipping this database."
skip=true
fi
echo "skip=${skip}" | tee -a $GITHUB_OUTPUT
- name:Check and create database if it does not exist
storage_broker={version="0.1",path="./storage_broker/"}# Note: main broker code is inside the binary crate, so linking with the library shouldn't be heavy.
# Use ARG as a build-time environment variable here to allow.
# It's not supposed to be set outside.
# Alternatively it can be obtained using the following command
# ```
# . /etc/os-release && echo "${VERSION_CODENAME}"
# ```
ARGDEBIAN_VERSION_CODENAME=bullseye
# Add nonroot user
RUN useradd -ms /bin/bash nonroot -b /home
SHELL["/bin/bash","-c"]
@@ -26,7 +34,6 @@ RUN set -e \
liblzma-dev \
libncurses5-dev \
libncursesw5-dev \
libpq-dev \
libreadline-dev \
libseccomp-dev \
libsqlite3-dev \
@@ -51,7 +58,7 @@ RUN set -e \
&& rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*
# protobuf-compiler (protoc)
ENV PROTOC_VERSION25.1
ENVPROTOC_VERSION=25.1
RUN curl -fsSL "https://github.com/protocolbuffers/protobuf/releases/download/v${PROTOC_VERSION}/protoc-${PROTOC_VERSION}-linux-$(uname -m | sed 's/aarch64/aarch_64/g').zip" -o "protoc.zip"\
&& unzip -q protoc.zip -d protoc \
&& mv protoc/bin/protoc /usr/local/bin/protoc \
@@ -67,19 +74,24 @@ RUN curl -sL "https://github.com/peak/s5cmd/releases/download/v${S5CMD_VERSION}/
# LLVM
ENVLLVM_VERSION=18
RUN curl -fsSL 'https://apt.llvm.org/llvm-snapshot.gpg.key'| apt-key add - \
# recommended approach from https://www.rust-lang.org/tools/install
@@ -126,7 +132,7 @@ make -j`sysctl -n hw.logicalcpu` -s
To run the `psql` client, install the `postgresql-client` package or modify `PATH` and `LD_LIBRARY_PATH` to include `pg_install/bin` and `pg_install/lib`, respectively.
To run the integration tests or Python scripts (not required to use the code), install
Python (3.9 or higher), and install the python3 packages using `./scripts/pysync` (requires [poetry>=1.3](https://python-poetry.org/)) in the project directory.
Python (3.9 or higher), and install the python3 packages using `./scripts/pysync` (requires [poetry>=1.8](https://python-poetry.org/)) in the project directory.
#### Running neon database
@@ -262,7 +268,7 @@ By default, this runs both debug and release modes, and all supported postgres v
testing locally, it is convenient to run just one set of permutations, like this:
.help("Use last Lsn of another timeline (and its data) as base when creating the new timeline. The timeline gets resolved by its branch name.").required(false))
@@ -1456,8 +1600,7 @@ fn cli() -> Command {
.about("Import timeline from basebackup directory")
assert_eq!(&PageServerConf::from(&conf),&self.conf,"during neon_local init, we derive the runtime state of ps conf (self.conf) from the --config flag fully");
// TODO(christian): instead of what we do here, create a pageserver_api::config::ConfigToml (PR #7656)
@@ -122,16 +127,19 @@ impl PageServerNode {
}
// Apply the user-provided overrides
overrides.push(
toml_edit::ser::to_string_pretty(&conf)
.expect("we deserialized this from toml earlier"),
);
overrides.push({
letmutdoc=
toml_edit::ser::to_document(&conf).expect("we deserialized this from toml earlier");
// `id` is written out to `identity.toml` instead of `pageserver.toml`
doc.remove("id").expect("it's part of the struct");
doc.to_string()
});
// Turn `overrides` into a toml document.
// TODO: above code is legacy code, it should be refactored to use toml_edit directly.
anyhow::bail!("This command is not a tenant deletion, and uncleanly drops all controller state for the tenant. If you know what you're doing, add `--unclean` to proceed.")
}
storcon_client
.dispatch::<(),()>(
Method::POST,
format!("debug/v1/tenant/{tenant_id}/drop"),
None,
)
.await?;
}
Command::NodeDrop{node_id,unclean}=>{
if!unclean{
anyhow::bail!("This command is not a clean node decommission, and uncleanly drops all controller state for the node, without checking if any tenants still refer to it. If you know what you're doing, add `--unclean` to proceed.")
@@ -11,15 +11,28 @@ page server. We currently use the same binary for both, with --wal-redo runtime
the WAL redo mode. Some PostgreSQL changes are needed in the compute node, while others are just for
the WAL redo process.
In addition to core PostgreSQL changes, there is a Neon extension in contrib/neon, to hook into the
smgr interface. Once all the core changes have been submitted to upstream or eliminated some other
way, the extension could live outside the postgres repository and build against vanilla PostgreSQL.
In addition to core PostgreSQL changes, there is a Neon extension in the pgxn/neon directory that
hooks into the smgr interface, and rmgr extension in pgxn/neon_rmgr. The extensions are loaded into
the Postgres processes with shared_preload_libraries. Most of the Neon-specific code is in the
extensions, and for any new features, that is preferred over modifying core PostgreSQL code.
Below is a list of all the PostgreSQL source code changes, categorized into changes needed for
compute, and changes needed for the WAL redo process:
# Changes for Compute node
## Prefetching
There are changes in many places to perform prefetching, for example for sequential scans. Neon
doesn't benefit from OS readahead, and the latency to pageservers is quite high compared to local
disk, so prefetching is critical for performance, also for sequential scans.
### How to get rid of the patch
Upcoming "streaming read" work in v17 might simplify this. And async I/O work in v18 will hopefully
do more.
## Add t_cid to heap WAL records
```
@@ -37,54 +50,11 @@ The problem is that the XLOG_HEAP_INSERT record does not include the command id
Bite the bullet and submit the patch to PostgreSQL, to add the t_cid to the WAL records. It makes the WAL records larger, which could make this unpopular in the PostgreSQL community. However, it might simplify some logical decoding code; Andres Freund briefly mentioned in PGCon 2022 discussion on Heikki's Neon presentation that logical decoding currently needs to jump through some hoops to reconstruct the same information.
Update from Heikki (2024-04-17): I tried to write an upstream patch for that, to use the t_cid field for logical decoding, but it was not as straightforward as it first sounded.
### Alternatives
Perhaps we could write an extra WAL record with the t_cid information, when a page is evicted that contains rows that were touched a transaction that's still running. However, that seems very complicated.
(errmsg("unexpected data beyond EOF in block %u of relation %s",
blockNum, relpath(smgr->smgr_rnode, forkNum)),
errhint("This has been seen to occur with buggy kernels; consider updating your system.")));
-
+ }
/*
* We *must* do smgrextend before succeeding, else the page will not
* be reserved by the kernel, and the next P_NEW call will decide to
```
PostgreSQL is a bit sloppy with extending relations. Usually, the relation is extended with zeros
first, then the page is filled, and finally the new page WAL-logged. But if multiple backends extend
a relation at the same time, the pages can be WAL-logged in different order.
I'm not sure what scenario exactly required this change in Neon, though.
### How to get rid of the patch
Submit patches to pgsql-hackers, to tighten up the WAL-logging around relation extension. It's a bit
confusing even in PostgreSQL. Maybe WAL log the intention to extend first, then extend the relation,
and finally WAL-log that the extension succeeded.
## Make smgr interface available to extensions
```
@@ -321,6 +186,8 @@ and finally WAL-log that the extension succeeded.
Submit to upstream. This could be useful for the Disk Encryption patches too, or for compression.
We have submitted this to upstream, but it's moving at glacial a speed.
https://commitfest.postgresql.org/47/4428/
## Added relpersistence argument to smgropen()
@@ -444,6 +311,148 @@ Ignore it. This is only needed for disaster recovery, so once we've eliminated a
patches, we can just keep it around as a patch or as separate branch in a repo.
## pg_waldump flags to ignore errors
After creating a new project or branch in Neon, the first timeline can begin in the middle of a WAL segment. pg_waldump chokes on that, so we added some flags to make it possible to ignore errors.
### How to get rid of the patch
Like previous one, ignore it.
## Backpressure if pageserver doesn't ingest WAL fast enough
```
@@ -3200,6 +3202,7 @@ ProcessInterrupts(void)
return;
InterruptPending = false;
+retry:
if (ProcDiePending)
{
ProcDiePending = false;
@@ -3447,6 +3450,13 @@ ProcessInterrupts(void)
if (ParallelApplyMessagePending)
HandleParallelApplyMessages();
+
+ /* Call registered callback if any */
+ if (ProcessInterruptsCallback)
+ {
+ if (ProcessInterruptsCallback())
+ goto retry;
+ }
}
```
### How to get rid of the patch
Submit a patch to upstream, for a hook in ProcessInterrupts. Could be useful for other extensions
Previously, SLRU files were included in the basebackup, but the total size of them can be large,
several GB, and downloading them all made the startup time too long.
### Alternatives
FUSE hook or LD_PRELOAD trick to intercept the reads on SLRU files
## WAL-log an all-zeros page as one large hole
- In XLogRecordAssemble()
### Problem we're trying to solve
This change was made in v16. Starting with v16, when PostgreSQL extends a relation, it first extends
it with zeros, and it can extend the relation more than one block at a time. The all-zeros page is WAL-ogged, but it's very wasteful to include 8 kB of zeros in the WAL for that. This hack was made so that we WAL logged a compact record with a whole-page "hole". However, PostgreSQL has assertions that prevent that such WAL records from being replayed, so this breaks compatibility such that unmodified PostreSQL cannot process Neon-generated WAL.
### How to get rid of the patch
Find another compact representation for a full-page image of an all-zeros page. A compressed image perhaps.
## Shut down walproposer after checkpointer
```
+ /* Neon: Also allow walproposer background worker to be treated like a WAL sender, so that it's shut down last */
+ if ((bp->bkend_type == BACKEND_TYPE_NORMAL || bp->bkend_type == BACKEND_TYPE_BGWORKER) &&
```
This changes was needed so that postmaster shuts down the walproposer process only after the shutdown checkpoint record is written. Otherwise, the shutdown record will never make it to the safekeepers.
### How to get rid of the patch
Do a bigger refactoring of the postmaster state machine, such that a background worker can specify
the shutdown ordering by itself. The postmaster state machine has grown pretty complicated, and
would benefit from a refactoring for the sake of readability anyway.
## EXPLAIN changes for prefetch and LFC
### How to get rid of the patch
Konstantin submitted a patch to -hackers already: https://commitfest.postgresql.org/47/4643/. Get that into a committable state.
## On-demand download of extensions
### How to get rid of the patch
FUSE or LD_PRELOAD trickery to intercept reads?
## Publication superuser checks
We have hacked CreatePublication so that also neon_superuser can create them.
### How to get rid of the patch
Create an upstream patch with more fine-grained privileges for publications CREATE/DROP that can be GRANTed to users.
## WAL log replication slots
### How to get rid of the patch
Utilize the upcoming v17 "slot sync worker", or a similar neon-specific background worker process, to periodically WAL-log the slots, or to export them somewhere else.
## WAL-log replication snapshots
### How to get rid of the patch
WAL-log them periodically, from a backgound worker.
## WAL-log relmapper files
Similarly to replications snapshot files, the CID mapping files generated during VACUUM FULL of a catalog table are WAL-logged
### How to get rid of the patch
WAL-log them periodically, from a backgound worker.
## XLogWaitForReplayOf()
??
# Not currently committed but proposed
## Disable ring buffer buffer manager strategies
@@ -472,23 +481,10 @@ hint bits are set. Wal logging hint bits updates requires FPI which significantl
Add special WAL record for setting page hints.
## Prefetching
### Why?
As far as pages in Neon are loaded on demand, to reduce node startup time
and also speedup some massive queries we need some mechanism for bulk loading to
reduce page request round-trip overhead.
Currently Postgres is supporting prefetching only for bitmap scan.
In Neon we should also use prefetch for sequential and index scans, because the OS is not doing it for us.
For sequential scan we could prefetch some number of following pages. For index scan we could prefetch pages
of heap relation addressed by TIDs.
## Prewarming
### Why?
Short downtime (or, in other words, fast compute node restart time) is one of the key feature of Zenith.
Short downtime (or, in other words, fast compute node restart time) is one of the key feature of Neon.
But overhead of request-response round-trip for loading pages on demand can make started node warm-up quite slow.
We can capture state of compute node buffer cache and send bulk request for this pages at startup.
- [neondatabase/neon](https://hub.docker.com/repository/docker/zenithdb/zenith) — image with pre-built `pageserver`, `safekeeper` and `proxy` binaries and all the required runtime dependencies. Built from [/Dockerfile](/Dockerfile).
- [neondatabase/compute-node](https://hub.docker.com/repository/docker/zenithdb/compute-node) — compute node image with pre-built Postgres binaries from [neondatabase/postgres](https://github.com/neondatabase/postgres).
- [neondatabase/neon](https://hub.docker.com/repository/docker/neondatabase/neon) — image with pre-built `pageserver`, `safekeeper` and `proxy` binaries and all the required runtime dependencies. Built from [/Dockerfile](/Dockerfile).
- [neondatabase/compute-node-v16](https://hub.docker.com/repository/docker/neondatabase/compute-node-v16) — compute node image with pre-built Postgres binaries from [neondatabase/postgres](https://github.com/neondatabase/postgres). Similar images exist for v15 and v14.
We build all images after a successful `release` tests run and push automatically to Docker Hub with two parallel CI jobs
1.`neondatabase/compute-tools` and `neondatabase/compute-node`
1.`neondatabase/compute-tools` and `neondatabase/compute-node-v16` (and -v15 and -v14)
2.`neondatabase/neon`
@@ -34,12 +34,12 @@ You can see a [docker compose](https://docs.docker.com/compose/) example to crea
1. create containers
You can specify version of neon cluster using following environment values.
- PG_VERSION: postgres version for compute (default is 14)
- TAG: the tag version of [docker image](https://registry.hub.docker.com/r/neondatabase/neon/tags) (default is latest), which is tagged in [CI test](/.github/workflows/build_and_test.yml)
- PG_VERSION: postgres version for compute (default is 16 as of this writing)
- TAG: the tag version of [docker image](https://registry.hub.docker.com/r/neondatabase/neon/tags), which is tagged in [CI test](/.github/workflows/build_and_test.yml). Default is 'latest'
```
$ cd docker-compose/
$ docker-compose down # remove the containers if exists
$ PG_VERSION=15 TAG=2937 docker-compose up --build -d # You can specify the postgres and image version
$ PG_VERSION=16 TAG=latest docker-compose up --build -d # You can specify the postgres and image version
Creating network "dockercompose_default" with the default driver
# Graceful Restarts of Storage Controller Managed Clusters
## Summary
This RFC describes new storage controller APIs for draining and filling tenant shards from/on pageserver nodes.
It also covers how these new APIs should be used by an orchestrator (e.g. Ansible) in order to implement
graceful cluster restarts.
## Motivation
Pageserver restarts cause read availablity downtime for tenants.
For example pageserver-3 @ us-east-1 was unavailable for a randomly
picked tenant (which requested on-demand activation) for around 30 seconds
during the restart at 2024-04-03 16:37 UTC.
Note that lots of shutdowns on loaded pageservers do not finish within the
[10 second systemd enforced timeout](https://github.com/neondatabase/infra/blob/0a5280b383e43c063d43cbf87fa026543f6d6ad4/.github/ansible/systemd/pageserver.service#L16). This means we are shutting down without flushing ephemeral layers
and have to reingest data in order to serve requests after restarting, potentially making first request latencies worse.
This problem is not yet very acutely felt in storage controller managed pageservers since
tenant density is much lower there. However, we are planning on eventually migrating all
pageservers to storage controller management, so it makes sense to solve the issue proactively.
## Requirements
- Pageserver re-deployments cause minimal downtime for tenants
- The storage controller exposes HTTP API hooks for draining and filling tenant shards
from a given pageserver. Said hooks can be used by an orchestrator proces or a human operator.
- The storage controller exposes some HTTP API to cancel draining and filling background operations.
- Failures to drain or fill the node should not be fatal. In such cases, cluster restarts should proceed
as usual (with downtime).
- Progress of draining/filling is visible through metrics
## Non Goals
- Integration with the control plane
- Graceful restarts for large non-HA tenants.
## Impacted Components
- storage controller
- deployment orchestrator (i.e. Ansible)
- pageserver (indirectly)
## Terminology
** Draining ** is the process through which all tenant shards that can be migrated from a given pageserver
are distributed across the rest of the cluster.
** Filling ** is the symmetric opposite of draining. In this process tenant shards are migrated onto a given
pageserver until the cluster reaches a resonable, quiescent distribution of tenant shards across pageservers.
** Node scheduling policies ** act as constraints to the scheduler. For instance, when a
node is set in the `Paused` policy, no further shards will be scheduled on it.
** Node ** is a pageserver. Term is used interchangeably in this RFC.
** Deployment orchestrator ** is a generic term for whatever drives our deployments.
Currently, it's an Ansible playbook.
## Background
### Storage Controller Basics (skip if already familiar)
Fundamentally, the storage controller is a reconciler which aims to move from the observed mapping between pageservers and tenant shards to an intended mapping. Pageserver nodes and tenant shards metadata is durably persisted in a database, but note that the mapping between the two entities is not durably persisted. Instead, this mapping (*observed state*) is constructed at startup by sending `GET location_config` requests to registered pageservers.
An internal scheduler maps tenant shards to pageservers while respecting certain constraints. The result of scheduling is the *intent state*. When the intent state changes, a *reconciliation* will inform pageservers about the new assigment via `PUT location_config` requests and will notify the compute via the configured hook.
### Background Optimizations
The storage controller performs scheduling optimizations in the background. It will
migrate attachments to warm secondaries and replace secondaries in order to balance
the cluster out.
### Reconciliations Concurrency Limiting
There's a hard limit on the number of reconciles that the storage controller
can have in flight at any given time. To get an idea of scales, the limit is
128 at the time of writing.
## Implementation
Note: this section focuses on the core functionality of the graceful restart process.
It doesn't neccesarily describe the most efficient approach. Optimizations are described
separately in a later section.
### Overall Flow
This section describes how to implement graceful restarts from the perspective
of Ansible, the deployment orchestrator. Pageservers are already restarted sequentially.
The orchestrator shall implement the following epilogue and prologue steps for each
pageserver restart:
#### Prologue
The orchestrator shall first fetch the pageserver node id from the control plane or
the pageserver it aims to restart directly. Next, it issues an HTTP request
to the storage controller in order to start the drain of said pageserver node.
All error responses are retried with a short back-off. When a 202 (Accepted)
HTTP code is returned, the drain has started. Now the orchestrator polls the
node status endpoint exposed by the storage controller in order to await the
end of the drain process. When the `policy` field of the node status response
becomes `PauseForRestart`, the drain has completed and the orchestrator can
proceed with restarting the pageserver.
The prologue is subject to an overall timeout. It will have a value in the ballpark
of minutes. As storage controller managed pageservers become more loaded this timeout
will likely have to increase.
#### Epilogue
After restarting the pageserver, the orchestrator issues an HTTP request
to the storage controller to kick off the filling process. This API call
may be retried for all error codes with a short backoff. This also serves
as a synchronization primitive as the fill will be refused if the pageserver
has not yet re-attached to the storage controller. When a 202(Accepted) HTTP
code is returned, the fill has started. Now the orchestrator polls the node
status endpoint exposed by the storage controller in order to await the end of
the filling process. When the `policy` field of the node status response becomes
`Active`, the fill has completed and the orchestrator may proceed to the next pageserver.
Again, the epilogue is subject to an overall timeout. We can start off with
using the same timeout as for the prologue, but can also consider relying on
the storage controller's background optimizations with a shorter timeout.
In the case that the deployment orchestrator times out, it attempts to cancel
the fill. This operation shall be retried with a short back-off. If it ultimately
fails it will require manual intervention to set the nodes scheduling policy to
`NodeSchedulingPolicy::Active`. Not doing that is not immediately problematic,
but it constrains the scheduler as mentioned previously.
### Node Scheduling Policy State Machine
The state machine below encodes the behaviours discussed above and
the various failover situations described in a later section.
Assuming no failures and/or timeouts the flow should be:
storing safekeepers per timeline instead of per tenant because we can't migrate
tenants atomically.
The important question is how updated configuration is delivered from
storage_controller to control plane to provide it to computes. As always, there
are two options, pull and push. Let's do it the same push as with pageserver
`/notify-attach` because 1) it keeps storage_controller out of critical compute
start path 2) provides easier upgrade: there won't be such a thing as 'timeline
managed by control plane / storcon', cplane just takes the value out of its db
when needed 3) uniformity. It makes storage_controller responsible for retrying notifying
control plane until it succeeds.
So, cplane `/notify-safekeepers` for the timeline accepts `Configuration` and
updates it in the db if the provided conf generation is higher (the cplane db
should also store generations for this). Similarly to [`/notify-attach`](https://www.notion.so/neondatabase/Storage-Controller-Control-Plane-interface-6de56dd310a043bfa5c2f5564fa98365), it
should update db which makes the call successful, and then try to schedule
`apply_config` if possible, it is ok if not. storage_controller
should rate limit calling the endpoint, but likely this won't be needed, as migration
throughput is limited by `pull_timeline`.
Timeline (branch) creation in cplane should call storage_controller POST
`tenant/:tenant_id/timeline` like it currently does for sharded tenants.
Response should be augmented with `safekeeper_conf: Configuration`. The call
should be retried until succeeds.
Timeline deletion and tenant deletion in cplane should call appropriate
storage_controller endpoints like it currently does for sharded tenants. The
calls should be retried until they succeed.
### storage_controller implementation
Current 'load everything on startup and keep in memory' easy design is fine.
Single timeline shouldn't take more than 100 bytes (it's 16 byte tenant_id, 16
byte timeline_id, int generation, vec of ~3 safekeeper ids plus some flags), so
10^6 of timelines shouldn't take more than 100MB.
Similar to pageserver attachment Intents storage_controller would have in-memory
`MigrationRequest` (or its absense) for each timeline and pool of tasks trying
to make these request reality; this ensures one instance of storage_controller
won't do several migrations on the same timeline concurrently. In the first
version it is simpler to have more manual control and no retries, i.e. migration
failure removes the request. Later we can build retries and automatic
scheduling/migration. `MigrationRequest` is
```
enum MigrationRequest {
To(Vec<NodeId>),
FinishPending,
}
```
`FinishPending` requests to run the procedure to ensure state is clean: current
configuration is not joint and majority of safekeepers are aware of it, but do
not attempt to migrate anywhere. If current configuration fetched on step 1 is
not joint it jumps to step 7. It should be run at startup for all timelines (but
similarly, in the first version it is ok to trigger it manually).
#### Schema
`safekeepers` table mirroring current `nodes` should be added, except that for
`scheduling_policy` field (seems like `status` is a better name for it): it is enough
to have at least in the beginning only 3 fields: 1) `active` 2) `offline` 3)
`decomissioned`.
`timelines` table:
```
table! {
// timeline_id is primary key
timelines (tenant_id, timeline_id) {
timeline_id -> Varchar,
tenant_id -> Varchar,
generation -> Int4,
sk_set -> Array<Int4>, // list of safekeeper ids
new_sk_set -> Nullable<Array<Int4>>, // list of safekeeper ids, null if not joint conf
cplane_notified_generation -> Int4,
}
}
```
#### API
Node management is similar to pageserver:
1) POST `/control/v1/safekeepers` upserts safekeeper.
2) GET `/control/v1/safekeepers` lists safekeepers.
3) GET `/control/v1/safekeepers/:node_id` gets safekeeper.
4) PUT `/control/v1/safekepers/:node_id/status` changes status to e.g.
`offline` or `decomissioned`. Initially it is simpler not to schedule any
migrations here.
Safekeeper deploy scripts should register safekeeper at storage_contorller as
they currently do with cplane, under the same id.
Timeline creation/deletion: already existing POST `tenant/:tenant_id/timeline`
would 1) choose initial set of safekeepers; 2) write to the db initial
`Configuration` with `INSERT ON CONFLICT DO NOTHING` returning existing row in
case of conflict; 3) create timeline on the majority of safekeepers (already
created is ok).
We don't want to block timeline creation when one safekeeper is down. Currently
this is solved by compute implicitly creating timeline on any safekeeper it is
connected to. This creates ugly timeline state on safekeeper when timeline is
created, but start LSN is not defined yet. It would be nice to remove this; to
do that, controller can in the background retry to create timeline on
safekeeper(s) which missed that during initial creation call. It can do that
through `pull_timeline` from majority so it doesn't need to remember
`parent_lsn` in its db.
Timeline deletion removes the row from the db and forwards deletion to the
current configuration members. Without additional actions deletions might leak,
see below on this; initially let's ignore these, reporting to cplane success if
at least one safekeeper deleted the timeline (this will remove s3 data).
Tenant deletion repeats timeline deletion for all timelines.
Migration API: the first version is the simplest and the most imperative:
1) PUT `/control/v1/safekeepers/migrate` schedules `MigrationRequest`s to move
all timelines from one safekeeper to another. It accepts json
```
{
"src_sk": u32,
"dst_sk": u32,
"limit": Optional<u32>,
}
```
Returns list of scheduled requests.
2) PUT `/control/v1/tenant/:tenant_id/timeline/:timeline_id/safekeeper_migrate` schedules `MigrationRequest`
to move single timeline to given set of safekeepers:
```
{
"desired_set": Vec<u32>,
}
```
Returns scheduled request.
Similar call should be added for the tenant.
It would be great to have some way of subscribing to the results (apart from
looking at logs/metrics).
Migration is executed as described above. One subtlety is that (local) deletion on
source safekeeper might fail, which is not a problem if we are going to
decomission the node but leaves garbage otherwise. I'd propose in the first version
1) Don't attempt deletion at all if node status is `offline`.
2) If it failed, just issue warning.
And add PUT `/control/v1/safekeepers/:node_id/scrub` endpoint which would find and
remove garbage timelines for manual use. It will 1) list all timelines on the
safekeeper 2) compare each one against configuration storage: if timeline
doesn't exist at all (had been deleted), it can be deleted. Otherwise, it can
be deleted under generation number if node is not member of current generation.
Automating this is untrivial; we'd need to register all potential missing
deletions <tenant_id,timeline_id,generation,node_id> in the same transaction
which switches configurations. Similarly when timeline is fully deleted to
prevent cplane operation from blocking when some safekeeper is not available
deletion should be also registered.
One more task pool should infinitely retry notifying control plane about changed
safekeeper sets.
3) GET `/control/v1/tenant/:tenant_id/timeline/:timeline_id/` should return
current in memory state of the timeline and pending `MigrationRequest`,
if any.
4) PUT `/control/v1/tenant/:tenant_id/timeline/:timeline_id/safekeeper_migrate_abort` tries to abort the
migration by switching configuration from the joint to the one with (previous) `sk_set` under CAS
(incrementing generation as always).
#### Dealing with multiple instances of storage_controller
Operations described above executed concurrently might create some errors but do
not prevent progress, so while we normally don't want to run multiple instances
of storage_controller it is fine to have it temporarily, e.g. during redeploy.
Any interactions with db update in-memory controller state, e.g. if migration
request failed because different one is in progress, controller remembers that
and tries to finish it.
## Testing
`neon_local` should be switched to use storage_controller, playing role of
control plane.
There should be following layers of tests:
1) Model checked TLA+ spec specifies the algorithm and verifies its basic safety.
2) To cover real code and at the same time test many schedules we should have
simulation tests. For that, configuration storage, storage_controller <->
safekeeper communication and pull_timeline need to be mocked and main switch
procedure wrapped to as a node (thread) in simulation tests, using these
mocks. Test would inject migrations like it currently injects
safekeeper/walproposer restars. Main assert is the same -- committed WAL must
not be lost.
3) Since simulation testing injects at relatively high level points (not
syscalls), it omits some code, in particular `pull_timeline`. Thus it is
better to have basic tests covering whole system as well. Extended version of
`test_restarts_under_load` would do: start background load and do migration
under it, then restart endpoint and check that no reported commits
had been lost. I'd also add one more creating classic network split scenario, with
one compute talking to AC and another to BD while migration from nodes ABC to ABD
happens.
4) Simple e2e test should ensure that full flow including cplane notification works.
## Order of implementation and rollout
Note that
- Control plane parts and integration with it is fully independent from everything else
(tests would use simulation and neon_local).
- There is a lot of infra work making storage_controller aware of timelines and safekeepers
and its impl/rollout should be separate from migration itself.
- Initially walproposer can just stop working while it observers joint configuration.
Such window would be typically very short anyway.
To rollout smoothly, both walproposer and safekeeper should have flag
`configurations_enabled`; when set to false, they would work as currently, i.e.
walproposer is able to commit on whatever safekeeper set it is provided. Until
all timelines are managed by storcon we'd need to use current script to migrate
and update/drop entries in the storage_controller database if it has any.
Safekeepers would need to be able to talk both current and new protocol version
with compute to reduce number of computes restarted in prod once v2 protocol is
deployed (though before completely switching we'd need to force this).
Let's have the following rollout order:
- storage_controller becomes aware of safekeepers;
- storage_controller gets timeline creation for new timelines and deletion requests, but
doesn't manage all timelines yet. Migration can be tested on these new timelines.
To keep control plane and storage_controller databases in sync while control
plane still chooses the safekeepers initially (until all timelines are imported
it can choose better), `TimelineCreateRequest` can get optional safekeepers
field with safekeepers chosen by cplane.
- Then we can import all existing timelines from control plane to
storage_controller and gradually enable configurations region by region.
Very rough implementation order:
- Add concept of configurations to safekeepers (including control file),
implement v3 protocol.
- Implement walproposer changes, including protocol.
- Implement storconn part. Use it in neon_local (and pytest).
- Make cplane store safekeepers per timeline instead of per tenant.
This RFC describes the issues around the current storage controller restart procedure
and describes an implementation which reduces downtime to a few milliseconds on the happy path.
## Motivation
Storage controller upgrades (restarts, more generally) can cause multi-second availability gaps.
While the storage controller does not sit on the main data path, it's generally not acceptable
to block management requests for extended periods of time (e.g. https://github.com/neondatabase/neon/issues/8034).
### Current Implementation
The storage controller runs in a Kubernetes Deployment configured for one replica and strategy set to [Recreate](https://kubernetes.io/docs/concepts/workloads/controllers/deployment/#recreate-deployment).
In non Kubernetes terms, during an upgrade, the currently running storage controller is stopped and, only after,
a new instance is created.
At start-up, the storage controller calls into all the pageservers it manages (retrieved from DB) to learn the
latest locations of all tenant shards present on them. This is usually fast, but can push into tens of seconds
under unfavourable circumstances: pageservers are heavily loaded or unavailable.
## Prior Art
There's probably as many ways of handling restarts gracefully as there are distributed systems. Some examples include:
* Active/Standby architectures: Two or more instance of the same service run, but traffic is only routed to one of them.
For fail-over, traffic is routed to one of the standbys (which becomes active).
* Consensus Algorithms (Raft, Paxos and friends): The part of consensus we care about here is leader election: peers communicate to each other
and use a voting scheme that ensures the existence of a single leader (e.g. Raft epochs).
## Requirements
* Reduce storage controller unavailability during upgrades to milliseconds
* Minimize the interval in which it's possible for more than one storage controller
to issue reconciles.
* Have one uniform implementation for restarts and upgrades
* Fit in with the current Kubernetes deployment scheme
## Non Goals
* Implement our own consensus algorithm from scratch
* Completely eliminate downtime storage controller downtime. Instead we aim to reduce it to the point where it looks
like a transient error to the control plane
## Impacted Components
* storage controller
* deployment orchestration (i.e. Ansible)
* helm charts
## Terminology
* Observed State: in-memory mapping between tenant shards and their current pageserver locations - currently built up
at start-up by quering pageservers
* Deployment: Kubernetes [primitive](https://kubernetes.io/docs/concepts/workloads/controllers/deployment/) that models
a set of replicas
## Implementation
### High Level Flow
At a very high level the proposed idea is to start a new storage controller instance while
the previous one is still running and cut-over to it when it becomes ready. The new instance,
should coordinate with the existing one and transition responsibility gracefully. While the controller
has built in safety against split-brain situations (via generation numbers), we'd like to avoid such
scenarios since they can lead to availability issues for tenants that underwent changes while two controllers
were operating at the same time and require operator intervention to remedy.
### Kubernetes Deployment Configuration
On the Kubernetes configuration side, the proposal is to update the storage controller `Deployment`
to use `spec.strategy.type = RollingUpdate`, `spec.strategy.rollingUpdate.maxSurge=1` and `spec.strategy.maxUnavailable=0`.
Under the hood, Kubernetes creates a new replica set and adds one pod to it (`maxSurge=1`). The old replica set does not
scale down until the new replica set has one replica in the ready state (`maxUnavailable=0`).
The various possible failure scenarios are investigated in the [Handling Failures](#handling-failures) section.
### Storage Controller Start-Up
This section describes the primitives required on the storage controller side and the flow of the happy path.
#### Database Table For Leader Synchronization
A new table should be added to the storage controller database for leader synchronization during startup.
This table will always contain at most one row. The proposed name for the table is `leader` and the schema
contains two elements:
*`hostname`: represents the hostname for the current storage controller leader - should be addressible
from other pods in the deployment
*`start_timestamp`: holds the start timestamp for the current storage controller leader (UTC timezone) - only required
for failure case handling: see [Previous Leader Crashes Before New Leader Readiness](#previous-leader-crashes-before-new-leader-readiness)
Storage controllers will read the leader row at start-up and then update it to mark themselves as the leader
at the end of the start-up sequence. We want compare-and-exchange semantics for the update: avoid the
situation where two concurrent updates succeed and overwrite each other. The default Postgres isolation
level is `READ COMMITTED`, which isn't strict enough here. This update transaction should use at least `REPEATABLE
READ` isolation level in order to [prevent lost updates](https://www.interdb.jp/pg/pgsql05/08.html). Currently,
the storage controller uses the stricter `SERIALIZABLE` isolation level for all transactions. This more than suits
our needs here.
```
START TRANSACTION ISOLATION LEVEL REPEATABLE READ
UPDATE leader SET hostname=<new_hostname>, start_timestamp=<new_start_ts>
WHERE hostname=<old_hostname>, start_timestampt=<old_start_ts>;
```
If the transaction fails or if no rows have been updated, then the compare-and-exchange is regarded as a failure.
#### Step Down API
A new HTTP endpoint should be added to the storage controller: `POST /control/v1/step_down`. Upon receiving this
request the leader cancels any pending reconciles and goes into a mode where it replies with 503 to all other APIs
and does not issue any location configurations to its pageservers. The successful HTTP response will return a serialized
snapshot of the observed state.
If other step down requests come in after the initial one, the request is handled and the observed state is returned (required
for failure scenario handling - see [Handling Failures](#handling-failures)).
#### Graceful Restart Happy Path
At start-up, the first thing the storage controller does is retrieve the sole row from the new
`leader` table. If such an entry exists, send a `/step_down` PUT API call to the current leader.
This should be retried a few times with a short backoff (see [1]). The aspiring leader loads the
observed state into memory and the start-up sequence proceeds as usual, but *without* querying the
pageservers in order to build up the observed state.
Before doing any reconciliations or persistence change, update the `leader` database table as described in the [Database Table For Leader Synchronization](database-table-for-leader-synchronization)
section. If this step fails, the storage controller process exits.
Note that no row will exist in the `leaders` table for the first graceful restart. In that case, force update the `leader` table
(without the WHERE clause) and perform with the pre-existing start-up procedure (i.e. build observed state by querying pageservers).
Summary of proposed new start-up sequence:
1. Call `/step_down`
2. Perform any pending database migrations
3. Load state from database
4. Load observed state returned in step (1) into memory
5. Do initial heartbeat round (may be moved after 5)
7. Mark self as leader by updating the database
8. Reschedule and reconcile everything
Some things to note from the steps above:
* The storage controller makes no changes to the cluster state before step (5) (i.e. no location config
calls to the pageserver and no compute notifications)
* Ask the current leader to step down before loading state from database so we don't get a lost update
if the transactions overlap.
* Before loading the observed state at step (3), cross-validate against the database. If validation fails,
fall back to asking the pageservers about their current locations.
* Database migrations should only run **after** the previous instance steps down (or the step down times out).
[1] The API call might fail because there's no storage controller running (i.e. [restart](#storage-controller-crash-or-restart)),
so we don't want to extend the unavailability period by much. We still want to retry since that's not the common case.
### Handling Failures
#### Storage Controller Crash Or Restart
The storage controller may crash or be restarted outside of roll-outs. When a new pod is created, its call to
`/step_down` will fail since the previous leader is no longer reachable. In this case perform the pre-existing
start-up procedure and update the leader table (with the WHERE clause). If the update fails, the storage controller
exists and consistency is maintained.
#### Previous Leader Crashes Before New Leader Readiness
When the previous leader (P1) crashes before the new leader (P2) passses the readiness check, Kubernetes will
reconcile the old replica set and create a new pod for it (P1'). The `/step_down` API call will fail for P1'
(see [2]).
Now we have two cases to consider:
* P2 updates the `leader` table first: The database update from P1' will fail and P1' will exit, or be terminated
by Kubernetes depending on timings.
* P1' updates the `leader` table first: The `hostname` field of the `leader` row stays the same, but the `start_timestamp` field changes.
The database update from P2 will fail (since `start_timestamp` does not match). P2 will exit and Kubernetes will
create a new replacement pod for it (P2'). Now the entire dance starts again, but with P1' as the leader and P2' as the incumbent.
[2] P1 and P1' may (more likely than not) be the same pod and have the same hostname. The implementation
should avoid this self reference and fail the API call at the client if the persisted hostname matches
the current one.
#### Previous Leader Crashes After New Leader Readiness
The deployment's replica sets already satisfy the deployment's replica count requirements and the
Kubernetes deployment rollout will just clean up the dead pod.
#### New Leader Crashes Before Pasing Readiness Check
The deployment controller scales up the new replica sets by creating a new pod. The entire procedure is repeated
with the new pod.
#### Network Partition Between New Pod and Previous Leader
This feels very unlikely, but should be considered in any case. P2 (the new aspiring leader) fails the `/step_down`
API call into P1 (the current leader). P2 proceeds with the pre-existing startup procedure and updates the `leader` table.
Kubernetes will terminate P1, but there may be a brief period where both storage controller can drive reconciles.
### Dealing With Split Brain Scenarios
As we've seen in the previous section, we can end up with two storage controller running at the same time. The split brain
duration is not bounded since the Kubernetes controller might become partitioned from the pods (unlikely though). While these
scenarios are not fatal, they can cause tenant unavailability, so we'd like to reduce the chances of this happening.
The rest of this section sketches some safety measure. It's likely overkill to implement all of them however.
### Ensure Leadership Before Producing Side Effects
The storage controller has two types of side effects: location config requests into pageservers and compute notifications into the control plane.
Before issuing either, the storage controller could check that it is indeed still the leader by querying the database. Side effects might still be
applied if they race with the database updatem, but the situation will eventually be detected. The storage controller process should terminate in these cases.
### Leadership Lease
Up until now, the leadership defined by this RFC is static. In order to bound the length of the split brain scenario, we could require the leadership
to be renewed periodically. Two new columns would be added to the leaders table:
1.`last_renewed` - timestamp indicating when the lease was last renewed
2.`lease_duration` - duration indicating the amount of time after which the lease expires
The leader periodically attempts to renew the lease by checking that it is in fact still the legitimate leader and updating `last_renewed` in the
same transaction. If the update fails, the process exits. New storage controller instances wishing to become leaders must wait for the current lease
to expire before acquiring leadership if they have not succesfully received a response to the `/step_down` request.
### Notify Pageserver Of Storage Controller Term
Each time that leadership changes, we can bump a `term` integer column in the `leader` table. This term uniquely identifies a leader.
Location config requests and re-attach responses can include this term. On the pageserver side, keep the latest term in memory and refuse
anything which contains a stale term (i.e. smaller than the current one).
### Observability
* The storage controller should expose a metric which describes it's state (`Active | WarmingUp | SteppedDown`).
Per region alerts should be added on this metric which triggers when:
+ no storage controller has been in the `Active` state for an extended period of time
+ more than one storage controllers are in the `Active` state
* An alert that periodically verifies that the `leader` table is in sync with the metric above would be very useful.
We'd have to expose the storage controller read only database to Grafana (perhaps it is already done).
## Alternatives
### Kubernetes Leases
Kubernetes has a [lease primitive](https://kubernetes.io/docs/concepts/architecture/leases/) which can be used to implement leader election.
Only one instance may hold a lease at any given time. This lease needs to be periodically renewed and has an expiration period.
In our case, it would work something like this:
*`/step_down` deletes the lease or stops it from renewing
* lease acquisition becomes part of the start-up procedure
The kubert crate implements a [lightweight lease API](https://docs.rs/kubert/latest/kubert/lease/struct.LeaseManager.html), but it's still
not exactly trivial to implement.
This approach has the benefit of baked in observability (`kubectl describe lease`), but:
* We offload the responsibility to Kubernetes which makes it harder to debug when things go wrong.
* More code surface than the simple "row in database" approach. Also, most of this code would be in
a dependency not subject to code review, etc.
* Hard to test. Our testing infra does not run the storage controller in Kubernetes and changing it do
so is not simple and complictes and the test set-up.
To my mind, the "row in database" approach is straightforward enough that we don't have to offload this
@@ -44,7 +44,7 @@ If you need to modify the database schema, here’s how to create a migration:
- Use `diesel migration generate <name>` to create a new migration
- Populate the SQL files in the `migrations/` subdirectory
- Use `DATABASE_URL=... diesel migration run` to apply the migration you just wrote: this will update the `[schema.rs](http://schema.rs)` file automatically.
- This requires a running database: the easiest way to do that is to just run `cargo neon init ; cargo neon start`, which will leave a database available at `postgresql://localhost:1235/attachment_service`
- This requires a running database: the easiest way to do that is to just run `cargo neon init ; cargo neon start`, which will leave a database available at `postgresql://localhost:1235/storage_controller`
- Commit the migration files and the changes to schema.rs
- If you need to iterate, you can rewind migrations with `diesel migration revert -a` and then `diesel migration run` again.
- The migrations are build into the storage controller binary, and automatically run at startup after it is deployed, so once you’ve committed a migration no further steps are needed.
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
