## Problem
In the batching PR
- https://github.com/neondatabase/neon/pull/9870
I stopped deducting the time-spent-in-throttle fro latency metrics,
i.e.,
- smgr latency metrics (`SmgrOpTimer`)
- basebackup latency (+scan latency, which I think is part of
basebackup).
The reason for stopping the deduction was that with the introduction of
batching, the trick with tracking time-spent-in-throttle inside
RequestContext and swap-replacing it from the `impl Drop for
SmgrOpTimer` no longer worked with >1 requests in a batch.
However, deducting time-spent-in-throttle is desirable because our
internal latency SLO definition does not account for throttling.
## Summary of changes
- Redefine throttling to be a page_service pagestream request throttle
instead of a throttle for repository `Key` reads through `Timeline::get`
/ `Timeline::get_vectored`.
- This means reads done by `basebackup` are no longer subject to any
throttle.
- The throttle applies after batching, before handling of the request.
- Drive-by fix: make throttle sensitive to cancellation.
- Rename metric label `kind` from `timeline_get` to `pagestream` to
reflect the new scope of throttling.
To avoid config format breakage, we leave the config field named
`timeline_get_throttle` and ignore the `task_kinds` field.
This will be cleaned up in a future PR.
## Trade-Offs
Ideally, we would apply the throttle before reading a request off the
connection, so that we queue the minimal amount of work inside the
process.
However, that's not possible because we need to do shard routing.
The redefinition of the throttle to limit pagestream request rate
instead of repository `Key` rate comes with several downsides:
- We're no longer able to use the throttle mechanism for other other
tasks, e.g. image layer creation.
However, in practice, we never used that capability anyways.
- We no longer throttle basebackup.
## Problem
We don't have good observability for memory usage. This would be useful
e.g. to debug OOM incidents or optimize performance or resource usage.
We would also like to use continuous profiling with e.g. [Grafana Cloud
Profiles](https://grafana.com/products/cloud/profiles-for-continuous-profiling/)
(see https://github.com/neondatabase/cloud/issues/14888).
This PR is intended as a proof of concept, to try it out in staging and
drive further discussions about profiling more broadly.
Touches https://github.com/neondatabase/neon/issues/9534.
Touches https://github.com/neondatabase/cloud/issues/14888.
Depends on #9779.
Depends on #9780.
## Summary of changes
Adds a HTTP route `/profile/heap` that takes a heap profile and returns
it. Query parameters:
* `format`: output format (`jemalloc` or `pprof`; default `pprof`).
Unlike CPU profiles (see #9764), heap profiles are not symbolized and
require the original binary to translate addresses to function names. To
make this work with Grafana, we'll probably have to symbolize the
process server-side -- this is left as future work, as is other output
formats like SVG.
Heap profiles don't work on macOS due to limitations in jemalloc.
Fixes https://github.com/neondatabase/cloud/issues/20973.
This refactors `connect_raw` in order to return direct access to the
delayed notices.
I cannot find a way to test this with psycopg2 unfortunately, although
testing it with psql does return the expected results.
#8564
## Problem
The main and backup consumption metric pushes are completely
independent,
resulting in different event time windows and different idempotency
keys.
## Summary of changes
* Merge the push tasks, but keep chunks the same size.
# Problem
The timeout-based batching adds latency to unbatchable workloads.
We can choose a short batching timeout (e.g. 10us) but that requires
high-resolution timers, which tokio doesn't have.
I thoroughly explored options to use OS timers (see
[this](https://github.com/neondatabase/neon/pull/9822) abandoned PR).
In short, it's not an attractive option because any timer implementation
adds non-trivial overheads.
# Solution
The insight is that, in the steady state of a batchable workload, the
time we spend in `get_vectored` will be hundreds of microseconds anyway.
If we prepare the next batch concurrently to `get_vectored`, we will
have a sizeable batch ready once `get_vectored` of the current batch is
done and do not need an explicit timeout.
This can be reasonably described as **pipelining of the protocol
handler**.
# Implementation
We model the sub-protocol handler for pagestream requests
(`handle_pagrequests`) as two futures that form a pipeline:
2. Batching: read requests from the connection and fill the current
batch
3. Execution: `take` the current batch, execute it using `get_vectored`,
and send the response.
The Reading and Batching stage are connected through a new type of
channel called `spsc_fold`.
See the long comment in the `handle_pagerequests_pipelined` for details.
# Changes
- Refactor `handle_pagerequests`
- separate functions for
- reading one protocol message; produces a `BatchedFeMessage` with just
one page request in it
- batching; tried to merge an incoming `BatchedFeMessage` into an
existing `BatchedFeMessage`; returns `None` on success and returns back
the incoming message in case merging isn't possible
- execution of a batched message
- unify the timeline handle acquisition & request span construction; it
now happen in the function that reads the protocol message
- Implement serial and pipelined model
- serial: what we had before any of the batching changes
- read one protocol message
- execute protocol messages
- pipelined: the design described above
- optionality for execution of the pipeline: either via concurrent
futures vs tokio tasks
- Pageserver config
- remove batching timeout field
- add ability to configure pipelining mode
- add ability to limit max batch size for pipelined configurations
(required for the rollout, cf
https://github.com/neondatabase/cloud/issues/20620 )
- ability to configure execution mode
- Tests
- remove `batch_timeout` parametrization
- rename `test_getpage_merge_smoke` to `test_throughput`
- add parametrization to test different max batch sizes and execution
moes
- rename `test_timer_precision` to `test_latency`
- rename the test case file to `test_page_service_batching.py`
- better descriptions of what the tests actually do
## On the holding The `TimelineHandle` in the pending batch
While batching, we hold the `TimelineHandle` in the pending batch.
Therefore, the timeline will not finish shutting down while we're
batching.
This is not a problem in practice because the concurrently ongoing
`get_vectored` call will fail quickly with an error indicating that the
timeline is shutting down.
This results in the Execution stage returning a `QueryError::Shutdown`,
which causes the pipeline / entire page service connection to shut down.
This drops all references to the
`Arc<Mutex<Option<Box<BatchedFeMessage>>>>` object, thereby dropping the
contained `TimelineHandle`s.
- => fixes https://github.com/neondatabase/neon/issues/9850
# Performance
Local run of the benchmarks, results in [this empty
commit](1cf5b1463f)
in the PR branch.
Key take-aways:
* `concurrent-futures` and `tasks` deliver identical `batching_factor`
* tail latency impact unknown, cf
https://github.com/neondatabase/neon/issues/9837
* `concurrent-futures` has higher throughput than `tasks` in all
workloads (=lower `time` metric)
* In unbatchable workloads, `concurrent-futures` has 5% higher
`CPU-per-throughput` than that of `tasks`, and 15% higher than that of
`serial`.
* In batchable-32 workload, `concurrent-futures` has 8% lower
`CPU-per-throughput` than that of `tasks` (comparison to tput of
`serial` is irrelevant)
* in unbatchable workloads, mean and tail latencies of
`concurrent-futures` is practically identical to `serial`, whereas
`tasks` adds 20-30us of overhead
Overall, `concurrent-futures` seems like a slightly more attractive
choice.
# Rollout
This change is disabled-by-default.
Rollout plan:
- https://github.com/neondatabase/cloud/issues/20620
# Refs
- epic: https://github.com/neondatabase/neon/issues/9376
- this sub-task: https://github.com/neondatabase/neon/issues/9377
- the abandoned attempt to improve batching timeout resolution:
https://github.com/neondatabase/neon/pull/9820
- closes https://github.com/neondatabase/neon/issues/9850
- fixes https://github.com/neondatabase/neon/issues/9835
## Problem
It appears that the Azure storage API tends to hang TCP connections more
than S3 does.
Currently we use a 2 minute timeout for all downloads. This is large
because sometimes the objects we download are large. However, waiting 2
minutes when doing something like downloading a manifest on tenant
attach is problematic, because when someone is doing a "create tenant,
create timeline" workflow, that 2 minutes is long enough for them
reasonably to give up creating that timeline.
Rather than propagate oversized timeouts further up the stack, we should
use a different timeout for objects that we expect to be small.
Closes: https://github.com/neondatabase/neon/issues/9836
## Summary of changes
- Add a `small_timeout` configuration attribute to remote storage,
defaulting to 30 seconds (still a very generous period to do something
like download an index)
- Add a DownloadKind parameter to DownloadOpts, so that callers can
indicate whether they expect the object to be small or large.
- In the azure client, use small timeout for HEAD requests, and for GET
requests if DownloadKind::Small is used.
- Use DownloadKind::Small for manifests, indices, and heatmap downloads.
This PR intentionally does not make the equivalent change to the S3
client, to reduce blast radius in case this has unexpected consequences
(we could accomplish the same thing by editing lots of configs, but just
skipping the code is simpler for right now)
Our rust-postgres fork is getting messy. Mostly because proxy wants more
control over the raw protocol than tokio-postgres provides. As such,
it's diverging more and more. Storage and compute also make use of
rust-postgres, but in more normal usage, thus they don't need our crazy
changes.
Idea:
* proxy maintains their subset
* other teams use a minimal patch set against upstream rust-postgres
Reviewing this code will be difficult. To implement it, I
1. Copied tokio-postgres, postgres-protocol and postgres-types from
00940fcdb5
2. Updated their package names with the `2` suffix to make them compile
in the workspace.
3. Updated proxy to use those packages
4. Copied in the code from tokio-postgres-rustls 0.13 (with some patches
applied https://github.com/jbg/tokio-postgres-rustls/pull/32https://github.com/jbg/tokio-postgres-rustls/pull/33)
5. Removed as much dead code as I could find in the vendored libraries
6. Updated the tokio-postgres-rustls code to use our existing channel
binding implementation
## Problem
We currently see elevated levels of errors for GetBlob requests. This is
because 404 and 304 are counted as errors for metric reporting.
## Summary of Changes
Bring the implementation in line with the S3 client and treat 404 and
304 responses as ok for metric purposes.
Related: https://github.com/neondatabase/cloud/issues/20666
# Problem
VM (visibility map) pages are stored and managed as any regular relation
page, in the VM fork of the main relation. They are also sharded like
other pages. Regular WAL writes to the VM pages (typically performed by
vacuum) are routed to the correct shard as usual. However, VM pages are
also updated via `ClearVmBits` metadata records emitted when main
relation pages are updated. These metadata records were sent to all
shards, like other metadata records. This had the following effects:
* On shards responsible for VM pages, the `ClearVmBits` applies as
expected.
* On shard 0, which knows about the VM relation and its size but doesn't
necessarily have any VM pages, the `ClearVmBits` writes may have been
applied without also having applied the explicit WAL writes to VM pages.
* If VM pages are spread across multiple shards (unlikely with 256MB
stripe size), all shards may have applied `ClearVmBits` if the pages
fall within their local view of the relation size, even for pages they
do not own.
* On other shards, this caused a relation size cache miss and a DbDir
and RelDir lookup before dropping the `ClearVmBits`. With many
relations, this could cause significant CPU overhead.
This is not believed to be a correctness problem, but this will be
verified in #9914.
Resolves#9855.
# Changes
Route `ClearVmBits` metadata records only to the shards responsible for
the VM pages.
Verification of the current VM handling and cleanup of incomplete VM
pages on shard 0 (and potentially elsewhere) is left as follow-up work.
## Problem
https://github.com/neondatabase/neon/pull/9746 lifted decoding and
interpretation of WAL to the safekeeper.
This reduced the ingested amount on the pageservers by around 10x for a
tenant with 8 shards, but doubled
the ingested amount for single sharded tenants.
Also, https://github.com/neondatabase/neon/pull/9746 uses bincode which
doesn't support schema evolution.
Technically the schema can be evolved, but it's very cumbersome.
## Summary of changes
This patch set addresses both problems by adding protobuf support for
the interpreted wal records and adding compression support. Compressed
protobuf reduced the ingested amount by 100x on the 32 shards
`test_sharded_ingest` case (compared to non-interpreted proto). For the
1 shard case the reduction is 5x.
Sister change to `rust-postgres` is
[here](https://github.com/neondatabase/rust-postgres/pull/33).
## Links
Related: https://github.com/neondatabase/neon/issues/9336
Epic: https://github.com/neondatabase/neon/issues/9329
## Problem
For any given tenant shard, pageservers receive all of the tenant's WAL
from the safekeeper.
This soft-blocks us from using larger shard counts due to bandwidth
concerns and CPU overhead of filtering
out the records.
## Summary of changes
This PR lifts the decoding and interpretation of WAL from the pageserver
into the safekeeper.
A customised PG replication protocol is used where instead of sending
raw WAL, the safekeeper sends
filtered, interpreted records. The receiver drives the protocol
selection, so, on the pageserver side, usage
of the new protocol is gated by a new pageserver config:
`wal_receiver_protocol`.
More granularly the changes are:
1. Optionally inject the protocol and shard identity into the arguments
used for starting replication
2. On the safekeeper side, implement a new wal sending primitive which
decodes and interprets records
before sending them over
3. On the pageserver side, implement the ingestion of this new
replication message type. It's very similar
to what we already have for raw wal (minus decoding and interpreting).
## Notes
* This PR currently uses my [branch of
rust-postgres](https://github.com/neondatabase/rust-postgres/tree/vlad/interpreted-wal-record-replication-support)
which includes the deserialization logic for the new replication message
type. PR for that is open
[here](https://github.com/neondatabase/rust-postgres/pull/32).
* This PR contains changes for both pageservers and safekeepers. It's
safe to merge because the new protocol is disabled by default on the
pageserver side. We can gradually start enabling it in subsequent
releases.
* CI tests are running on https://github.com/neondatabase/neon/pull/9747
## Links
Related: https://github.com/neondatabase/neon/issues/9336
Epic: https://github.com/neondatabase/neon/issues/9329
Co-authored-by: Heikki Linnakangas <heikki@neon.tech>
Co-authored-by: Stas Kelvic <stas@neon.tech>
# Context
This PR contains PoC-level changes for a product feature that allows
onboarding large databases into Neon without going through the regular
data path.
# Changes
This internal RFC provides all the context
* https://github.com/neondatabase/cloud/pull/19799
In the language of the RFC, this PR covers
* the Importer code (`fast_import`)
* all the Pageserver changes (mgmt API changes, flow implementation,
etc)
* a basic test for the Pageserver changes
# Reviewing
As acknowledged in the RFC, the code added in this PR is not ready for
general availability.
Also, the **architecture is not to be discussed in this PR**, but in the
RFC and associated Slack channel instead.
Reviewers of this PR should take that into consideration.
The quality bar to apply during review depends on what area of the code
is being reviewed:
* Importer code (`fast_import`): practically anything goes
* Core flow (`flow.rs`):
* Malicious input data must be expected and the existing threat models
apply.
* The code must not be safe to execute on *dedicated* Pageserver
instances:
* This means in particular that tenants *on other* Pageserver instances
must not be affected negatively wrt data confidentiality, integrity or
availability.
* Other code: the usual quality bar
* Pay special attention to correct use of gate guards, timeline
cancellation in all places during shutdown & migration, etc.
* Consider the broader system impact; if you find potentially
problematic interactions with Storage features that were not covered in
the RFC, bring that up during the review.
I recommend submitting three separate reviews, for the three high-level
areas with different quality bars.
# References
(Internal-only)
* refs https://github.com/neondatabase/cloud/issues/17507
* refs https://github.com/neondatabase/company_projects/issues/293
* refs https://github.com/neondatabase/company_projects/issues/309
* refs https://github.com/neondatabase/cloud/issues/20646
---------
Co-authored-by: Stas Kelvich <stas.kelvich@gmail.com>
Co-authored-by: Heikki Linnakangas <heikki@neon.tech>
Co-authored-by: John Spray <john@neon.tech>
## Problem
Follow up of https://github.com/neondatabase/neon/pull/9682, that patch
didn't fully address the problem: what if shutdown fails due to whatever
reason and then we reattach the tenant? Then we will still remove the
future layer. The underlying problem is that the fix for #5878 gets
voided because of the generation optimizations.
Of course, we also need to ensure that delete happens after uploads, but
note that we only schedule deletes when there are no ongoing upload
tasks, so that's fine.
## Summary of changes
* Add a test case to reproduce the behavior (by changing the original
test case to attach the same generation).
* If layer upload happens after the deletion, drain the deletion queue
before uploading.
* If blocked_deletion is enabled, directly remove it from the
blocked_deletion queue.
* Local fs backend fix to avoid race between deletion and preload.
* test_emergency_mode does not need to wait for uploads (and it's
generally not possible to wait for uploads).
* ~~Optimize deletion executor to skip validation if there are no files
to delete.~~ this doesn't work
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
close https://github.com/neondatabase/neon/issues/9836
Looking at Azure SDK, the only related issue I can find is
https://github.com/azure/azure-sdk-for-rust/issues/1549. Azure uses
reqwest as the backend, so I assume there's some underlying magic
unknown to us that might have caused the stuck in #9836.
The observation is:
* We didn't get an explicit out of resource HTTP error from Azure.
* The connection simply gets stuck and times out.
* But when we retry after we reach the timeout, it succeeds.
This issue is hard to identify -- maybe something went wrong at the ABS
side, or something wrong with our side. But we know that a retry will
usually succeed if we give up the stuck connection.
Therefore, I propose the fix that we preempt stuck HTTP operation and
actively retry. This would mitigate the problem, while in the long run,
we need to keep an eye on ABS usage and see if we can fully resolve this
problem.
The reasoning of such timeout mechanism: we use a much smaller timeout
than before to preempt, while it is possible that a normal listing
operation would take a longer time than the initial timeout if it
contains a lot of keys. Therefore, after we terminate the connection, we
should double the timeout, so that such requests would eventually
succeed.
## Summary of changes
* Use exponential growth for ABS list timeout.
* Rather than using a fixed timeout, use a timeout that starts small and
grows
* Rather than exposing timeouts to the list_streaming caller as soon as
we see them, only do so after we have retried a few times
Signed-off-by: Alex Chi Z <chi@neon.tech>
The notifications need to be sent whenever the waiters heap changes, per
the comment in `update_status`. But if 'advance' is called when there
are no waiters, or the new LSN is lower than the waiters so that no one
needs to be woken up, there's no need to send notifications. This saves
some CPU cycles in the common case that there are no waiters.
## Problem
We don't have a convenient way to gather CPU profiles from a running
binary, e.g. during production incidents or end-to-end benchmarks, nor
during microbenchmarks (particularly on macOS).
We would also like to have continuous profiling in production, likely
using [Grafana Cloud
Profiles](https://grafana.com/products/cloud/profiles-for-continuous-profiling/).
We may choose to use either eBPF profiles or pprof profiles for this
(pending testing and discussion with SREs), but pprof profiles appear
useful regardless for the reasons listed above. See
https://github.com/neondatabase/cloud/issues/14888.
This PR is intended as a proof of concept, to try it out in staging and
drive further discussions about profiling more broadly.
Touches #9534.
Touches https://github.com/neondatabase/cloud/issues/14888.
## Summary of changes
Adds a HTTP route `/profile/cpu` that takes a CPU profile and returns
it. Defaults to a 5-second pprof Protobuf profile for use with e.g.
`pprof` or Grafana Alloy, but can also emit an SVG flamegraph. Query
parameters:
* `format`: output format (`pprof` or `svg`)
* `frequency`: sampling frequency in microseconds (default 100)
* `seconds`: number of seconds to profile (default 5)
Also integrates pprof profiles into Criterion benchmarks, such that
flamegraph reports can be taken with `cargo bench ... --profile-duration
<seconds>`. Output under `target/criterion/*/profile/flamegraph.svg`.
Example profiles:
* pprof profile (use [`pprof`](https://github.com/google/pprof)):
[profile.pb.gz](https://github.com/user-attachments/files/17756788/profile.pb.gz)
* Web interface: `pprof -http :6060 profile.pb.gz`
* Interactive flamegraph:
[profile.svg.gz](https://github.com/user-attachments/files/17756782/profile.svg.gz)
Fixes the masking for the CancelKeyData display format. Due to negative
i32 cast to u64, the top-bits all had `0xffffffff` prefix. On the
bitwise-or that followed, these took priority.
This PR also compresses 3 logs during sql-over-http into 1 log with
durations as label fields, as prior discussed.
Adds support to the `find_garbage` command to restrict itself to a
partial tenant ID prefix, say `a`, and then it only traverses tenants
with IDs starting with `a`. One can now pass the `--tenant-id-prefix`
parameter.
That way, one can shard the `find_garbage` command and make it run in
parallel.
The PR also does a change of how `remote_storage` first removes trailing
`/`s, only to then add them in the listing function. It turns out that
this isn't neccessary and it prevents the prefix functionality from
working. S3 doesn't do this either.
Follow up to #9803
See https://github.com/neondatabase/cloud/issues/14378
In collaboration with @cloneable and @awarus, we sifted through logs and
simply demoted some logs to debug. This is not at all finished and there
are more logs to review, but we ran out of time in the session we
organised. In any slightly more nuanced cases, we didn't touch the log,
instead leaving a TODO comment.
I've also slightly refactored the sql-over-http body read/length reject
code. I can split that into a separate PR. It just felt natural after I
switched to `read_body_with_limit` as we discussed during the meet.
## Problem
Two recently observed log errors indicate safekeeper tasks for a
timeline running after that timeline's deletion has started.
- https://github.com/neondatabase/neon/issues/8972
- https://github.com/neondatabase/neon/issues/8974
These code paths do not have a mechanism that coordinates task shutdown
with the overall shutdown of the timeline.
## Summary of changes
- Add a `Gate` to `Timeline`
- Take the gate as part of resident timeline guard: any code that holds
a guard over a timeline staying resident should also hold a guard over
the timeline's total lifetime.
- Take the gate from the wal removal task
- Respect Timeline::cancel in WAL send/recv code, so that we do not
block shutdown indefinitely.
- Add a test that deletes timelines with open pageserver+compute
connections, to check these get torn down as expected.
There is some risk to introducing gates: if there is code holding a gate
which does not properly respect a cancellation token, it can cause
shutdown hangs. The risk of this for safekeepers is lower in practice
than it is for other services, because in a healthy timeline deletion,
the compute is shutdown first, then the timeline is deleted on the
pageserver, and finally it is deleted on the safekeepers -- that makes
it much less likely that some protocol handler will still be running.
Closes: #8972Closes: #8974
part of https://github.com/neondatabase/neon/issues/9114, we want to be
able to run partial gc-compaction in tests. In the future, we can also
expand this functionality to legacy compaction, so that we can trigger
compaction for a specific key range.
## Summary of changes
* Support passing compaction key range through pageserver routes.
* Refactor input parameters of compact related function to take the new
`CompactOptions`.
* Add tests for partial compaction. Note that the test may or may not
trigger compaction based on GC horizon. We need to improve the test case
to ensure things always get below the gc_horizon and the gc-compaction
can be triggered.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
Earlier work (#7547) has made the scrubber internally generic, but one
could only configure it to use S3 storage.
This is the final piece to make (most of, snapshotting still requires
S3) the scrubber be able to be configured via GenericRemoteStorage.
I.e. you can now set an env var like:
```
REMOTE_STORAGE_CONFIG='remote_storage = { bucket_name = "neon-dev-safekeeper-us-east-2d", bucket_region = "us-east-2" }
```
and the scrubber will read it instead.
## Problem
We don't take advantage of queue depth generated by the compute
on the pageserver. We can process getpage requests more efficiently
by batching them.
## Summary of changes
Batch up incoming getpage requests that arrive within a configurable
time window (`server_side_batch_timeout`).
Then process the entire batch via one `get_vectored` timeline operation.
By default, no merging takes place.
## Testing
* **Functional**: https://github.com/neondatabase/neon/pull/9792
* **Performance**: will be done in staging/pre-prod
# Refs
* https://github.com/neondatabase/neon/issues/9377
* https://github.com/neondatabase/neon/issues/9376
Co-authored-by: Christian Schwarz <christian@neon.tech>
## Problem
See https://github.com/neondatabase/neon/issues/7750
test_wal_restore.sh is copying file to current working directory which
can cause interfere of test_wa_restore.py tests spawned of different
configurations.
## Summary of changes
Copy file to $DATA_DIR
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
## Problem
It turns out that `WalStreamDecoder::poll_decode` returns the start LSN
of the next record and not the end LSN of the current record. They are
not always equal. For example, they're not equal when the record in
question is an XLOG SWITCH record.
## Summary of changes
Rename things to reflect that.
## Problem
We want to serialize interpreted records to send them over the wire from
safekeeper to pageserver.
## Summary of changes
Make `InterpretedWalRecord` ser/de. This is a temporary change to get
the bulk of the lift merged in
https://github.com/neondatabase/neon/pull/9746. For going to prod, we
don't want to use bincode since we can't evolve the schema.
Questions on serialization will be tackled separately.
## Problem
https://github.com/neondatabase/neon/issues/9240
## Summary of changes
Correctly truncate VM page instead just replacing it with zero page.
## 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
Historically, if a control component passed a pageserver "generation: 1"
this could be a quick way to corrupt a tenant by loading a historic
index.
Follows https://github.com/neondatabase/neon/pull/9383Closes#6951
## Summary of changes
- Introduce a Fatal variant to DownloadError, to enable index downloads
to signal when they have encountered a scary enough situation that we
shouldn't proceed to load the tenant.
- Handle this variant by putting the tenant into a broken state (no
matter which timeline within the tenant reported it)
- Add a test for this case
In the event that this behavior fires when we don't want it to, we have
ways to intervene:
- "Touch" an affected index to update its mtime (download+upload S3
object)
- If this behavior is triggered, it indicates we're attaching in some
old generation, so we should be able to fix that by manually bumping
generation numbers in the storage controller database (this should never
happen, but it's an option if it does)
## Problem
WAL segment fsyncs significantly affect WAL ingestion throughput.
`durable_rename()` is used when initializing every 16 MB segment, and
issues 3 fsyncs of which 1 was unnecessary.
## Summary of changes
Remove an fsync in `durable_rename` which is unnecessary with Linux and
ext4 (which we currently use). This improves WAL ingestion throughput by
up to 23% with large appends on my MacBook.
I had an impression that gc-compaction didn't test the case where the
first record of the key history is will_init because of there are some
code path that will panic in this case. Luckily it got fixed in
https://github.com/neondatabase/neon/pull/9026 so we can now implement
such tests.
Part of https://github.com/neondatabase/neon/issues/9114
## Summary of changes
* Randomly changed some images into will_init neon wal record
* Split `test_simple_bottom_most_compaction_deltas` into two test cases,
one of them has the bottom layer as delta layer with will_init flags,
while the other is the original one with image layers.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
The control file is flushed on the WAL ingest path when the commit LSN
advances by one segment, to bound the amount of recovery work in case of
a crash. This involves 3 additional fsyncs, which can have a significant
impact on WAL ingest throughput. This is to some extent mitigated by
`AppendResponse` not being emitted on segment bound flushes, since this
will prevent commit LSN advancement, which will be addressed separately.
## Summary of changes
Don't flush the control file on the WAL ingest path at all. Instead,
leave that responsibility to the timeline manager, but ask it to flush
eagerly if the control file lags the in-memory commit LSN by more than
one segment. This should not cause more than `REFRESH_INTERVAL` (300 ms)
additional latency before flushing the control file, which is
negligible.
Removes some unnecessary initdb arguments, and fixes Neon for MacOS
since it doesn't seem to ship a C.UTF-8 locale.
Signed-off-by: Tristan Partin <tristan@neon.tech>
## Problem
We wish to stop using admin tokens in the infra repo, but step down
requests use the admin token.
## Summary of Changes
Introduce a new "ControllerPeer" scope and use it for step-down requests.
Right now, our environments create databases with the C locale, which is
really unfortunate for users who have data stored in other languages
that they want to analyze. For instance, show_trgm on Hebrew text
currently doesn't work in staging or production.
I don't envision this being the final solution. I think this is just a
way to set a known value so the pageserver doesn't use its parent
environment. The final solution to me is exposing initdb parameters to
users in the console. Then they could use a different locale or encoding
if they so chose.
Signed-off-by: Tristan Partin <tristan@neon.tech>
## Problem
In test environments, the `syncfs` that the pageserver does on startup
can take a long time, as other tests running concurrently might have
many gigabytes of dirty pages.
## Summary of changes
- Add a `no_sync` option to the pageserver's config.
- Skip syncfs on startup if this is set
- A subsequent PR (https://github.com/neondatabase/neon/pull/9678) will
enable this by default in tests. We need to wait until after the next
release to avoid breaking compat tests, which would fail if we set
no_sync & use an old pageserver binary.
Q: Why is this a different mechanism than safekeeper, which as a
--no-sync CLI?
A: Because the way we manage pageservers in neon_local depends on the
pageserver.toml containing the full configuration, whereas safekeepers
have a config file which is neon-local-specific and can drive a CLI
flag.
Q: Why is the option no_sync rather than sync?
A: For boolean configs with a dangerous value, it's preferable to make
"false" the safe option, so that any downstream future config tooling
that might have a "booleans are false by default" behavior (e.g. golang
structs) is safe by default.
Q: Why only skip the syncfs, and not all fsyncs?
A: Skipping all fsyncs would require more code changes, and the most
acute problem isn't fsyncs themselves (these just slow down a running
test), it's the syncfs (which makes a pageserver startup slow as a
result of _other_ tests)
## Problem
Benchmarks need more control over the WAL generated by `WalGenerator`.
In particular, they need to vary the size of logical messages.
## Summary of changes
* Make `WalGenerator` generic over `RecordGenerator`, which constructs
WAL records.
* Add `LogicalMessageGenerator` which emits logical messages, with a
configurable payload.
* Minor tweaks and code reorganization.
There are no changes to the core logic or emitted WAL.
## Problem
https://github.com/neondatabase/neon/pull/9524 split the decoding and
interpretation step from ingestion.
The output of the first phase is a `wal_decoder::models::InterpretedWalRecord`.
Before this patch set that struct contained a list of `Value` instances.
We wish to lift the decoding and interpretation step to the safekeeper,
but it would be nice if the safekeeper gave us a batch containing the raw data instead of actual values.
## Summary of changes
Main goal here is to make `InterpretedWalRecord` hold a raw buffer which
contains pre-serialized Values.
For this we do:
1. Add a `SerializedValueBatch` type. This is `inmemory_layer::SerializedBatch` with some
extra functionality for extension, observing values for shard 0 and tests.
2. Replace `inmemory_layer::SerializedBatch` with `SerializedValueBatch`
3. Make `DatadirModification` maintain a `SerializedValueBatch`.
### `DatadirModification` changes
`DatadirModification` now maintains a `SerializedValueBatch` and extends
it as new WAL records come in (to avoid flushing to disk on every
record).
In turn, this cascaded into a number of modifications to
`DatadirModification`:
1. Replace `pending_data_pages` and `pending_zero_data_pages` with `pending_data_batch`.
2. Removal of `pending_zero_data_pages` and its cousin `on_wal_record_end`
3. Rename `pending_bytes` to `pending_metadata_bytes` since this is what it tracks now.
4. Adapting of various utility methods like `len`, `approx_pending_bytes` and `has_dirty_data_pages`.
Removal of `pending_zero_data_pages` and the optimisation associated
with it ((1) and (2)) deserves more detail.
Previously all zero data pages went through `pending_zero_data_pages`.
We wrote zero data pages when filling gaps caused by relation extension
(case A) and when handling special wal records (case B). If it happened
that the same WAL record contained a non zero write for an entry in
`pending_zero_data_pages` we skipped the zero write.
Case A: We handle this differently now. When ingesting the
`SerialiezdValueBatch` associated with one PG WAL record, we identify the gaps and fill the
them in one go. Essentially, we move from a per key process (gaps were filled after each
new key), and replace it with a per record process. Hence, the optimisation is not
required anymore.
Case B: When the handling of a special record needs to zero out a key,
it just adds that to the current batch. I inspected the code, and I
don't think the optimisation kicked in here.
## Problem
We don't have any observability for Safekeeper WAL receiver queues.
## Summary of changes
Adds a few WAL receiver metrics:
* `safekeeper_wal_receivers`: gauge of currently connected WAL
receivers.
* `safekeeper_wal_receiver_queue_depth`: histogram of queue depths per
receiver, sampled every 5 seconds.
* `safekeeper_wal_receiver_queue_depth_total`: gauge of total queued
messages across all receivers.
* `safekeeper_wal_receiver_queue_size_total`: gauge of total queued
message sizes across all receivers.
There are already metrics for ingested WAL volume: `written_wal_bytes`
counter per timeline, and `safekeeper_write_wal_bytes` per-request
histogram.
It seems the ecosystem is not so keen on moving to aws-lc-rs as it's
build setup is more complicated than ring (requiring cmake).
Eventually I expect the ecosystem should pivot to
https://github.com/ctz/graviola/tree/main/rustls-graviola as it
stabilises (it has a very simply build step and license), but for now
let's try not have a headache of juggling two crypto libs.
I also noticed that tonic will just fail with tls without a default
provider, so I added some defensive code for that.
## Problem
Decoding and ingestion are still coupled in `pageserver::WalIngest`.
## Summary of changes
A new type is added to `wal_decoder::models`, InterpretedWalRecord. This
type contains everything that the pageserver requires in order to ingest
a WAL record. The highlights are the `metadata_record` which is an
optional special record type to be handled and `blocks` which stores
key, value pairs to be persisted to storage.
This type is produced by
`wal_decoder::models::InterpretedWalRecord::from_bytes` from a raw PG
wal record.
The rest of this commit separates decoding and interpretation of the PG
WAL record from its application in `WalIngest::ingest_record`.
Related: https://github.com/neondatabase/neon/issues/9335
Epic: https://github.com/neondatabase/neon/issues/9329
## Problem
We don't have a convenient way to generate WAL records for benchmarks
and tests.
## Summary of changes
Adds a WAL generator, exposed as an iterator. It currently only
generates logical messages (noops), but will be extended to write actual
table rows later.
Some existing code for WAL generation has been replaced with this
generator, to reduce duplication.
## Problem
We wish to have high level WAL decoding logic in `wal_decoder::decoder`
module.
## Summary of Changes
For this we need the `Value` and `NeonWalRecord` types accessible there, so:
1. Move `Value` and `NeonWalRecord` to `pageserver::value` and
`pageserver::record` respectively.
2. Get rid of `pageserver::repository` (follow up from (1))
3. Move PG specific WAL record types to `postgres_ffi::walrecord`. In
theory they could live in `wal_decoder`, but it would create a circular
dependency between `wal_decoder` and `postgres_ffi`. Long term it makes
sense for those types to be PG version specific, so that will work out nicely.
4. Move higher level WAL record types (to be ingested by pageserver)
into `wal_decoder::models`
Related: https://github.com/neondatabase/neon/issues/9335
Epic: https://github.com/neondatabase/neon/issues/9329
## Problem
Part of https://github.com/neondatabase/neon/issues/8623
## Summary of changes
Removed all aux-v1 config processing code. Note that we persisted it
into the index part file, so we cannot really remove the field from
index part. I also kept the config item within the tenant config, but we
will not read it any more.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
In complement to
https://github.com/neondatabase/tokio-epoll-uring/pull/56.
## Problem
We want to make tokio-epoll-uring slots waiters queue depth observable
via Prometheus.
## Summary of changes
- Add `pageserver_tokio_epoll_uring_slots_submission_queue_depth`
metrics as a `Histogram`.
- Each thread-local tokio-epoll-uring system is given a `LocalHistogram`
to observe the metrics.
- Keep a list of `Arc<ThreadLocalMetrics>` used on-demand to flush data
to the shared histogram.
- Extend `Collector::collect` to report
`pageserver_tokio_epoll_uring_slots_submission_queue_depth`.
Signed-off-by: Yuchen Liang <yuchen@neon.tech>
Co-authored-by: Christian Schwarz <christian@neon.tech>
