## Problem
- `test_basebackup_cache` fails in
https://github.com/neondatabase/neon/pull/11712 because after the
timelines on safekeepers are managed by storage controller, they do
contain proper start_lsn and the compute_ctl tool sends the first
basebackup request with this LSN.
- `Failed to prepare basebackup` log messages during timeline
initialization, because the timeline is not yet in the global timeline
map.
- Relates to https://github.com/neondatabase/cloud/issues/29353
## Summary of changes
- Account for `timeline_onto_safekeepers` storcon's option in the test.
- Do not trigger basebackup prepare during the timeline initialization.
## Problem
We support two ingest protocols on the pageserver: vanilla and
interpreted.
Interpreted has been the only protocol in use for a long time.
## Summary of changes
* Remove the ingest handling of the vanilla protocol
* Remove tenant and pageserver configuration for it
* Update all tests that tweaked the ingest protocol
## Compatibility
Backward compatibility:
* The new pageserver version can read the existing pageserver
configuration and it will ignore the unknown field.
* When the tenant config is read from the storcon db or from the
pageserver disk, the extra field will be ignored.
Forward compatiblity:
* Both the pageserver config and the tenant config map missing fields to
their default value.
I'm not aware of any tenant level override that was made for this knob.
## Problem
We want to repro an OOM situation, but large partial reads are required.
## Summary of Changes
Make the max partial read size configurable for import jobs.
## Problem
Setting `max_batch_size` to anything higher than
`Timeline::MAX_GET_VECTORED_KEYS` will cause runtime error. We should
rather fail fast at startup if this is the case.
## Summary of changes
* Create `max_get_vectored_keys` as a new configuration (default to 32);
* Validate `max_batch_size` against `max_get_vectored_keys` right at
config parsing and validation.
Closes https://github.com/neondatabase/neon/issues/11994
## Problem
We print a backtrace in an info level log every 10 seconds while waiting
for the import data to land in the bucket.
## Summary of changes
The backtrace is not useful. Remove it.
## Problem
We should expose the page service over gRPC.
Requires #12093.
Touches #11728.
## Summary of changes
This patch adds an initial page service implementation over gRPC. It
ties in with the existing `PageServerHandler` request logic, to avoid
the implementations drifting apart for the core read path.
This is just a bare-bones functional implementation. Several important
aspects have been omitted, and will be addressed in follow-up PRs:
* Limited observability: minimal tracing, no logging, limited metrics
and timing, etc.
* Rate limiting will currently block.
* No performance optimization.
* No cancellation handling.
* No tests.
I've only done rudimentary testing of this, but Pagebench passes at
least.
## Problem
If the wal receiver is cancelled, there's a 50% chance that it will
ingest yet more WAL.
## Summary of Changes
Always check cancellation first.
## Problem
There's a bunch of TODOs in the import code.
## Summary of changes
1. Bound max import byte range to 128MiB. This might still be too high,
given the default job concurrency, but it needs to be balanced with
going back and forth to S3.
2. Prevent unsigned overflow when determining key range splits for
concurrent jobs
3. Use sharded ranges to estimate task size when splitting jobs
4. Bubble up errors that we might hit due to invalid data in the bucket
back to the storage controller.
5. Tweak the import bucket S3 client configuration.
This patch is a fixup for
- https://github.com/neondatabase/neon/pull/6788
Background
----------
That PR 6788 added artificial advancement of `disk_consistent_lsn`
and `remote_consistent_lsn` for shards that weren't written to
while other shards _were_ written to.
See the PR description for more context.
At the time of that PR, Pageservers shards were doing WAL filtering.
Nowadays, the WAL filtering happens in Safekeepers.
Shards learn about the WAL gaps via
`InterpretedWalRecords::next_record_lsn`.
The Bug
-------
That artificial advancement code also runs if the flush failed.
So, we advance the disk_consistent_lsn / remote_consistent_lsn,
without having the corresponding L0 to the `index_part.json`.
The frozen layer remains in the layer map until detach,
so we continue to serve data correctly.
We're not advancing flush loop variable `flushed_to_lsn` either,
so, subsequent flush requests will retry the flush and repair the
situation if they succeed.
But if there aren't any successful retries, eventually the tenant
will be detached and when it is attached somewhere else, the
`index_part.json` and therefore layer map...
1. ... does not contain the frozen layer that failed to flush and
2. ... won't re-ingest that WAL either because walreceiver
starts up with the advanced disk_consistent_lsn/remote_consistent_lsn.
The result is that the read path will have a gap in the reconstruct
data for the keys whose modifications were lost, resulting in
a) either walredo failure
b) or an incorrect page@lsn image if walredo doesn't error.
The Fix
-------
The fix is to only do the artificial advancement if `result.is_ok()`.
Misc
----
As an aside, I took some time to re-review the flush loop and its
callers.
I found one more bug related to error handling that I filed here:
- https://github.com/neondatabase/neon/issues/12025
## Problem
## Summary of changes
## Problem
Disk usage eviction isn't sensitive to layers of imported timelines.
## Summary of changes
Hook importing timelines up into eviction and add a test for it.
I don't think we need any special eviction logic for this. These layers
will all be visible and
their access time will be their creation time. Hence, we'll remove
covered layers first
and get to the imported layers if there's still disk pressure.
## Problem
Importing timelines can't currently be deleted. This is problematic
because:
1. Cplane cannot delete failed imports and we leave the timeline behind.
2. The flow does not support user driven cancellation of the import
## Summary of changes
On the pageserver: I've taken the path of least resistance, extended
`TimelineOrOffloaded`
with a new variant and added handling in the right places. I'm open to
thoughts here,
but I think it turned out better than I was envisioning.
On the storage controller: Again, fairly simple business: when a DELETE
timeline request is
received, we remove the import from the DB and stop any finalization
tasks/futures. In order
to stop finalizations, we track them in-memory. For each finalizing
import, we associate a gate
and a cancellation token.
Note that we delete the entry from the database before cancelling any
finalizations. This is such
that a concurrent request can't progress the import into finalize state
and race with the deletion.
This concern about deleting an import with on-going finalization is
theoretical in the near future.
We are only going to delete importing timelines after the storage
controller reports the failure to
cplane. Alas, the design works for user driven cancellation too.
Closes https://github.com/neondatabase/neon/issues/11897
## Problem
Temporarily reduce the concurrency of gc-compaction to 1 job at a time.
We are going to roll out in the largest AWS region next week. Having one
job running at a time makes it easier to identify what tenant causes
problem if it's not running well and pause gc-compaction for that
specific tenant.
(We can make this configurable via pageserver config in the future!)
## Summary of changes
Reduce `CONCURRENT_GC_COMPACTION_TASKS` from 2 to 1.
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
Import planning takes a job size limit as its input. Previously, the job
size came from a pageserver config field. This field may change while
imports are in progress. If this happens, plans will no longer be
identical and the import would fail permanently.
## Summary of Changes
Bake the job size into the import progress reported to the storage
controller. For new imports, use the value from the pagesever config,
and, for existing imports, use the value present in the shard progress.
This value is identical for all shards, but we want it to be versioned
since future versions of the planner might split the jobs up
differently. Hence, it ends up in `ShardImportProgress`.
Closes https://github.com/neondatabase/neon/issues/11983
## Problem
Previous attempt https://github.com/neondatabase/neon/pull/10548 caused
some issues in staging and we reverted it. This is a re-attempt to
address https://github.com/neondatabase/neon/issues/11063.
Currently we create image layers at latest record LSN. We would create
"future image layers" (i.e., image layers with LSN larger than disk
consistent LSN) that need special handling at startup. We also waste a
lot of read operations to reconstruct from L0 layers while we could have
compacted all of the L0 layers and operate on a flat level of historic
layers.
## Summary of changes
* Run repartition at L0-L1 boundary.
* Roll out with feature flags.
* Piggyback a change that downgrades "image layer creating below
gc_cutoff" to debug level.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
Test coverage of timeline imports is lacking.
## Summary of changes
This PR adds a chaos import test. It runs an import while injecting
various chaos events
in the environment. All the commits that follow the test fix various
issues that were surfaced by it.
Closes https://github.com/neondatabase/neon/issues/10191
## Problem
Basebackup cache is on the hot path of compute startup and is generated
on every request (may be slow).
- Issue: https://github.com/neondatabase/cloud/issues/29353
## Summary of changes
- Add `BasebackupCache` which stores basebackups on local disk.
- Basebackup prepare requests are triggered by
`XLOG_CHECKPOINT_SHUTDOWN` records in the log.
- Limit the size of the cache by number of entries.
- Add `basebackup_cache_enabled` feature flag to TenantConfig.
- Write tests for the cache
## Not implemented yet
- Limit the size of the cache by total size in bytes
---------
Co-authored-by: Aleksandr Sarantsev <aleksandr@neon.tech>
## Problem
For billing, we'd like per-branch consumption metrics.
Requires https://github.com/neondatabase/neon/pull/11984.
Resolves https://github.com/neondatabase/cloud/issues/28155.
## Summary of changes
This patch adds two new consumption metrics:
* `written_size_since_parent`: `written_size - ancestor_lsn`
* `pitr_history_size_since_parent`: `written_size - max(pitr_cutoff,
ancestor_lsn)`
Note that `pitr_history_size_since_parent` will not be emitted until the
PITR cutoff has been computed, and may or may not increase ~immediately
when a user increases their PITR window (depending on how much history
we have available and whether the tenant is restarted/migrated).
## Problem
It is not currently possible to disambiguate a timeline with an
uninitialized PITR cutoff from one that was created within the PITR
window -- both of these have `GcCutoffs::time == Lsn(0)`. For billing
metrics, we need to disambiguate these to avoid accidentally billing the
entire history when a tenant is initially loaded.
Touches https://github.com/neondatabase/cloud/issues/28155.
## Summary of changes
Make `GcCutoffs::time` an `Option<Lsn>`, and only set it to `Some` when
initialized. A `pitr_interval` of 0 will yield `Some(last_record_lsn)`.
This PR takes a conservative approach, and mostly retains the old
behavior of consumers by using `unwrap_or_default()` to yield 0 when
uninitialized, to avoid accidentally introducing bugs -- except in cases
where there is high confidence that the change is beneficial (e.g. for
the `pageserver_pitr_history_size` Prometheus metric and to return early
during GC).
## Problem
See
Discussion:
https://neondb.slack.com/archives/C033RQ5SPDH/p1746645666075799
Issue: https://github.com/neondatabase/cloud/issues/28609
Relation size cache is not correctly updated at PS in case of replicas.
## Summary of changes
1. Have two caches for relation size in timeline:
`rel_size_primary_cache` and `rel_size_replica_cache`.
2. `rel_size_primary_cache` is actually what we have now. The only
difference is that it is not updated in `get_rel_size`, only by WAL
ingestion
3. `rel_size_replica_cache` has limited size (LruCache) and it's key is
`(Lsn,RelTag)` . It is updated in `get_rel_size`. Only strict LSN
matches are accepted as cache hit.
---------
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
## Problem
The gRPC page service API will require decoupling the `PageHandler` from
the libpq protocol implementation. As preparation for this, avoid
passing in the entire server config to `PageHandler`, and instead
explicitly pass in the relevant fields.
Touches https://github.com/neondatabase/neon/issues/11728.
## Summary of changes
* Change `PageHandler` to take a `GetVectoredConcurrentIo` instead of
the entire config.
* Change `IoConcurrency::spawn_from_conf` to take a
`GetVectoredConcurrentIo`.
There were some incompatible changes. Most churn was from switching from
the now-deprecated fcntl:flock() function to
fcntl::Flock::lock(). The new function returns a guard object, while
with the old function, the lock was associated directly with the file
descriptor.
It's good to stay up-to-date in general, but the impetus to do this now
is that in https://github.com/neondatabase/neon/pull/11929, I want to
use some functions that were added only in the latest version of 'nix',
and it's nice to not have to build multiple versions. (Although,
different versions of 'nix' are still pulled in as indirect dependencies
from other packages)
## Problem
Imports don't support schema evolution nicely. If we want to change the
stuff we keep in storcon,
we'd have to carry the old cruft around.
## Summary of changes
Version import progress. Note that the import progress version
determines the version of the import
job split and execution. This means that we can also use it as a
mechanism for deploying new import
implementations in the future.
## Problem
Timeline imports do not have progress checkpointing. Any time that the
tenant is shut-down, all progress is lost
and the import restarts from the beginning when the tenant is
re-attached.
## Summary of changes
This PR adds progress checkpointing.
### Preliminaries
The **unit of work** is a `ChunkProcessingJob`. Each
`ChunkProcessingJob` deals with the import for a set of key ranges. The
job split is done by using an estimation of how many pages each job will
produce.
The planning stage must be **pure**: given a fixed set of contents in
the import bucket, it will always yield the same plan. This property is
enforced by checking that the hash of the plan is identical when
resuming from a checkpoint.
The storage controller tracks the progress of each shard in the import
in the database in the form of the **latest
job** that has has completed.
### Flow
This is the high level flow for the happy path:
1. On the first run of the import task, the import task queries storcon
for the progress and sees that none is recorded.
2. Execute the preparatory stage of the import
3. Import jobs start running concurrently in a `FuturesOrdered`. Every
time the checkpointing threshold of jobs has been reached, notify the
storage controller.
4. Tenant is detached and re-attached
5. Import task starts up again and gets the latest progress checkpoint
from the storage controller in the form of a job index.
6. The plan is computed again and we check that the hash matches with
the original plan.
7. Jobs are spawned from where the previous import task left off. Note
that we will not report progress after the completion of each job, so
some jobs might run twice.
Closes https://github.com/neondatabase/neon/issues/11568
Closes https://github.com/neondatabase/neon/issues/11664
## Problem
Import up-calls did not enforce the usage of the latest generation. The
import might have finished in one previous generation, but not in the
latest one. Hence, the controller might try to activate a timeline
before it is ready. In theory, that would be fine, but it's tricky to
reason about.
## Summary of Changes
Pageserver provides the current generation in the upcall to the storage
controller and the later validates the generation. If the generation is
stale, we return an error which stops progress of the import job. Note
that the import job will retry the upcall until the stale location is
detached.
I'll add some proper tests for this as part of the [checkpointing
PR](https://github.com/neondatabase/neon/pull/11862).
Closes https://github.com/neondatabase/neon/issues/11884
## Problem
Hopefully resolves `test_gc_feedback` flakiness.
## Summary of changes
`accumulated_values` should not exceed 512MB to avoid OOM. Previously we
only use number of items, which is not a good estimation.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
Lifetime of imported timelines (and implicitly the import background
task) has some shortcomings:
1. Timeline activation upon import completion is tricky. Previously, a
timeline that finished importing
after a tenant detach would not get activated and there's concerns about
the safety of activating
concurrently with shut-down.
2. Import jobs can prevent tenant shut down since they hold the tenant
gate
## Summary of Changes
Track the import tasks in memory and abort them explicitly on tenant
shutdown.
Integrate more closely with the storage controller:
1. When an import task has finished all of its jobs, it notifies the
storage controller, but **does not** mark the import as done in the
index_part. When all shards have finished importing, the storage
controller will call the `/activate_post_import` idempotent endpoint for
all of them. The handler, marks the import complete in index part,
resets the tenant if required and checks if the timeline is active yet.
2. Not directly related, but the import job now gets the starting state
from the storage controller instead of the import bucket. This paves the
way for progress checkpointing.
Related: https://github.com/neondatabase/neon/issues/11568
## Problem
We observe image compaction errors after gc-compaction finishes
compacting below the gc_cutoff. This is because `repartition` returns an
LSN below the gc horizon as we (likely) determined that `distance <=
self.repartition_threshold`.
I think it's better to keep the current behavior of when to trigger
compaction but we should skip image compaction if the returned LSN is
below the gc horizon.
## Summary of changes
If the repartition returns an invalid LSN, skip image compaction.
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
The limitation we imposed last week
https://github.com/neondatabase/neon/pull/11709 is not enough to protect
excessive memory usage.
## Summary of changes
If a single key accumulated too much history, give up compaction. In the
future, we can make the `generate_key_retention` function take a stream
of keys instead of first accumulating them in memory, thus easily
support such long key history cases.
Signed-off-by: Alex Chi Z <chi@neon.tech>
# Problem
Before this PR, timeline shutdown would
- cancel the walreceiver cancellation token subtree (child token of
Timeline::cancel)
- call freeze_and_flush
- Timeline::cancel.cancel()
- ... bunch of waiting for things ...
- Timeline::gate.close()
As noted by the comment that is deleted by this PR, this left a window
where, after freeze_and_flush, walreceiver could still be running and
ingest data into a new InMemoryLayer.
This presents a potential source of log noise during Timeline shutdown
where the InMemoryLayer created after the freeze_and_flush observes
that Timeline::cancel is cancelled, failing the ingest with some
anyhow::Error wrapping (deeply) a `FlushTaskError::Cancelled` instance
(`flush task cancelled` error message).
# Solution
It turns out that it is quite easy to shut down, not just cancel,
walreceiver completely
because the only subtask spawned by walreceiver connection manager is
the `handle_walreceiver_connection` task, which is properly shut down
and waited upon when the manager task observes cancellation and exits
its retry loop.
The alternative is to replace all the usage of `anyhow` on the ingest
path
with differentiated error types. A lot of busywork for little gain to
fix
a potential logging noise nuisance, so, not doing that for now.
# Correctness / Risk
We do not risk leaking walreceiver child tasks because existing
discipline
is to hold a gate guard.
We will prolong `Timeline::shutdown` to the degree that we're no longer
making
progress with the rest of shutdown while the walreceiver task hasn't yet
observed cancellation. In practice, this should be negligible.
`Timeline::shutdown` could fail to complete if there is a hidden
dependency
of walreceiver shutdown on some subsystem. The code certainly suggests
there
isn't, and I'm not aware of any such dependency. Anyway, impact will be
low
because we only shut down Timeline instances that are obsolete, either
because
there is a newer attachment at a different location, or because the
timeline
got deleted by the user. We would learn about this through stuck cplane
operations or stuck storcon reconciliations. We would be able to
mitigate by
cancelling such stuck operations/reconciliations and/or by rolling back
pageserver.
# Refs
- identified this while investigating
https://github.com/neondatabase/neon/issues/11762
- PR that _does_ fix a bunch _real_ `flush task cancelled` noise on the
compaction path: https://github.com/neondatabase/neon/pull/11853
## Problem
Import code is one big block. Separating planning and execution will
help with reporting
progress of import to storcon (building block for resuming import).
## Summary of changes
Split up the import into planning and execution.
A concurrency limit driven by PS config is also added.
# Refs
- fixes https://github.com/neondatabase/neon/issues/11762
# Problem
PR #10993 introduced internal retries for BufferedWriter flushes.
PR #11052 added cancellation sensitivity to that retry loop.
That cancellation sensitivity is an error path that didn't exist before.
The result is that during timeline shutdown, after we
`Timeline::cancel`, compaction can now fail with error `flush task
cancelled`.
The problem with that:
1. We mis-classify this as an `error!`-worthy event.
2. This causes tests to become flaky because the error is not in global
`allowed_errors`.
Technically we also trip the `compaction_circuit_breaker` because the
resulting `CompactionError` is variant `::Other`.
But since this is Timeline shutdown, is doesn't matter practically
speaking.
# Solution / Changes
- Log the anyhow stack trace when classifying a compaction error as
`error!`.
This was helpful to identify sources of `flush task cancelled` errors.
We only log at `error!` level in exceptional circumstances, so, it's ok
to have bit verbose logs.
- Introduce typed errors along the `BufferedWriter::write_*`=>
`BlobWriter::write_blob`
=> `{Delta,Image}LayerWriter::put_*` =>
`Split{Delta,Image}LayerWriter::put_{value,image}` chain.
- Proper mapping to `CompactionError`/`CreateImageLayersError` via new
`From` impls.
I am usually opposed to any magic `From` impls, but, it's how most of
the compaction code
works today.
# Testing
The symptoms are most prevalent in
`test_runner/regress/test_branch_and_gc.py::test_branch_and_gc`.
Before this PR, I was able to reproduce locally 1 or 2 times per 400
runs using
`DEFAULT_PG_VERSION=15 BUILD_TYPE=release poetry run pytest --count 400
-n 8`.
After this PR, it doesn't reproduce anymore after 2000 runs.
# Future Work
Technically the ingest path is also exposed to this new source of errors
because `InMemoryLayer` is backed by `BufferedWriter`.
But we haven't seen it occur in flaky tests yet.
Details and a fix in
- https://github.com/neondatabase/neon/pull/11851
## Problem
There's a few rough edges around PS tracing.
## Summary of changes
* include compute request id in pageserver trace
* use the get page specific context for GET_REL_SIZE and GET_BATCH
* fix assertion in download layer trace
## Problem
Address comments in https://github.com/neondatabase/neon/pull/11709
## Summary of changes
- remove `iter` API, users always need to specify buffer size depending
on the expected memory usage.
- several doc improvements
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
Co-authored-by: Christian Schwarz <christian@neon.tech>
## Problem
Part of https://github.com/neondatabase/neon/issues/11615
## Summary of changes
We don't understand the root cause of why we get resident size surge
every now and then. This patch adds observability for that, and in the
next week, we might have a better understanding of what's going on.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
We didn't consider tombstones in replorigin read path in the past. This
was fine because tombstones are stored as LSN::Invalid before we
universally define what the tombstone is for sparse keyspaces.
Now we remove non-inherited keys during detach ancestor and write the
universal tombstone "empty image". So we need to consider it across all
the read paths.
related: https://github.com/neondatabase/neon/pull/11299
## Summary of changes
Empty value gets ignored for replorigin scans.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
We had retained the ability to run in a generation-less mode to support
test_generations_upgrade, which was replaced with a cleaner backward
compat test in https://github.com/neondatabase/neon/pull/10701
## Summary of changes
- Remove all the special cases for "if no generation" or "if no control
plane api"
- Make control_plane_api config mandatory
---------
Co-authored-by: Arpad Müller <arpad-m@users.noreply.github.com>
## Problem
close https://github.com/neondatabase/neon/issues/11694
We had the delta layer iterator and image layer iterator set to buffer
at most 8MB data. Note that 8MB is the compressed size, so it is
possible for those iterators contain more than 8MB data in memory.
For the recent OOM case, gc-compaction was running over 556 layers,
which means that we will have 556 active iterators. So in theory, it
could take up to 556*8=4448MB memory when the compaction is going on. If
images get compressed and the compression ratio is high (for that
tenant, we see 3x compression ratio across image layers), then that's
13344MB memory.
Also we have layer rewrites, which explains the memory taken by
gc-compaction itself (versus the iterators). We rewrite 424 out of 556
layers, and each of such rewrites need a pair of delta layer writer. So
we are buffering a lot of deltas in the memory.
The flamegraph shows that gc-compaction itself takes 6GB memory, delta
iterator 7GB, and image iterator 2GB, which can be explained by the
above theory.
## Summary of changes
- Reduce the buffer sizes.
- Estimate memory consumption and if it is too high.
- Also give up if the number of layers-to-rewrite is too high.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
# Problem
The Pageserver read path exclusively uses direct IO if
`virtual_file_io_mode=direct`.
The write path is half-finished. Here is what the various writing
components use:
|what|buffering|flags on <br/>`v_f_io_mode`<br/>=`buffered`|flags on
<br/>`virtual_file_io_mode`<br/>=`direct`|
|-|-|-|-|
|`DeltaLayerWriter`| BlobWriter<BUFFERED=true> | () | () |
|`ImageLayerWriter`| BlobWriter<BUFFERED=false> | () | () |
|`download_layer_file`|BufferedWriter|()|()|
|`InMemoryLayer`|BufferedWriter|()|O_DIRECT|
The vehicle towards direct IO support is `BufferedWriter` which
- largely takes care of O_DIRECT alignment & size-multiple requirements
- double-buffering to mask latency
`DeltaLayerWriter`, `ImageLayerWriter` use `blob_io::BlobWriter` , which
has neither of these.
# Changes
## High-Level
At a high-level this PR makes the following primary changes:
- switch the two layer writer types to use `BufferedWriter` & make
sensitive to `virtual_file_io_mode` (via open_with_options_**v2**)
- make `download_layer_file` sensitive to `virtual_file_io_mode` (also
via open_with_options_**v2**)
- add `virtual_file_io_mode=direct-rw` as a feature gate
- we're hackish-ly piggybacking on OpenOptions's ask for write access
here
- this means with just `=direct` InMemoryLayer reads and writes no
longer uses O_DIRECT
- this is transitory and we'll remove the `direct-rw` variant once the
rollout is complete
(The `_v2` APIs for opening / creating VirtualFile are those that are
sensitive to `virtual_file_io_mode`)
The result is:
|what|uses <br/>`BufferedWriter`|flags on
<br/>`v_f_io_mode`<br/>=`buffered`|flags on
<br/>`v_f_io_mode`<br/>=`direct`|flags on
<br/>`v_f_io_mode`<br/>=`direct-rw`|
|-|-|-|-|-|
|`DeltaLayerWriter`| ~~Blob~~BufferedWriter | () | () | O_DIRECT |
|`ImageLayerWriter`| ~~Blob~~BufferedWriter | () | () | O_DIRECT |
|`download_layer_file`|BufferedWriter|()|()|O_DIRECT|
|`InMemoryLayer`|BufferedWriter|()|~~O_DIRECT~~()|O_DIRECT|
## Code-Level
The main change is:
- Switch `blob_io::BlobWriter` away from its own buffering method to use
`BufferedWriter`.
Additional prep for upholding `O_DIRECT` requirements:
- Layer writer `finish()` methods switched to use IoBufferMut for
guaranteed buffer address alignment. The size of the buffers is PAGE_SZ
and thereby implicitly assumed to fulfill O_DIRECT requirements.
For the hacky feature-gating via `=direct-rw`:
- Track `OpenOptions::write(true|false)` in a field; bunch of mechanical
churn.
- Consolidate the APIs in which we "open" or "create" VirtualFile for
better overview over which parts of the code use the `_v2` APIs.
Necessary refactorings & infra work:
- Add doc comments explaining how BufferedWriter ensures that writes are
compliant with O_DIRECT alignment & size constraints. This isn't new,
but should be spelled out.
- Add the concept of shutdown modes to `BufferedWriter::shutdown` to
make writer shutdown adhere to these constraints.
- The `PadThenTruncate` mode might not be necessary in practice because
I believe all layer files ever written are sized in multiples `PAGE_SZ`
and since `PAGE_SZ` is larger than the current alignment requirements
(512/4k depending on platform), it won't be necesary to pad.
- Some test (I believe `round_trip_test_compressed`?) required it though
- [ ] TODO: decide if we want to accept that complexity; if we do then
address TODO in the code to separate alignment requirement from buffer
capacity
- Add `set_len` (=`ftruncate`) VirtualFile operation to support the
above.
- Allow `BufferedWriter` to start at a non-zero offset (to make room for
the summary block).
Cleanups unlocked by this change:
- Remove non-positional APIs from VirtualFile (e.g. seek, write_full,
read_full)
Drive-by fixes:
- PR https://github.com/neondatabase/neon/pull/11585 aimed to run unit
tests for all `virtual_file_io_mode` combinations but didn't because of
a missing `_` in the env var.
# Performance
This section assesses this PR's impact on deployments with current
production setting (`=direct`) and anticipated impact of switching to
(`=direct-rw`).
For `DeltaLayerWriter`, `=direct` should remain unchanged to slightly
improved on throughput because the `BlobWriter`'s buffer had the same
size as the `BufferedWriter`'s buffer, but it didn't have the
double-buffering that `BufferedWriter` has.
The `=direct-rw` enables direct IO; throughput should not be suffering
because of double-buffering; benchmarks will show if this is true.
The `ImageLayerWriter` was previously not doing any buffering
(`BUFFERED=false`).
It went straight to issuing the IO operation to the underlying
VirtualFile and the buffering was done by the kernel.
The switch to `BufferedWriter` under `=direct` adds an additional memcpy
into the BufferedWriter's buffer.
We will win back that memcpy when enabling direct IO via `=direct-rw`.
A nice win from the switch to `BufferedWriter` is that ImageLayerWriter
performs >=16x fewer write operations to VirtualFile (the BlobWriter
performs one write per len field and one write per image value).
This should save low tens of microseconds of CPU overhead from doing all
these syscalls/io_uring operations, regardless of `=direct` or
`=direct-rw`.
Aside from problems with alignment, this write frequency without
double-buffering is prohibitive if we actually have to wait for the
disk, which is what will happen when we enable direct IO via
(`=direct-rw`).
Throughput should not be suffering because of BufferedWrite's
double-buffering; benchmarks will show if this is true.
`InMemoryLayer` at `=direct` will flip back to using buffered IO but
remain on BufferedWriter.
The buffered IO adds back one memcpy of CPU overhead.
Throughput should not suffer and will might improve on
not-memory-pressured Pageservers but let's remember that we're doing the
whole direct IO thing to eliminate global memory pressure as a source of
perf variability.
## bench_ingest
I reran `bench_ingest` on `im4gn.2xlarge` and `Hetzner AX102`.
Use `git diff` with `--word-diff` or similar to see the change.
General guidance on interpretation:
- immediate production impact of this PR without production config
change can be gauged by comparing the same `io_mode=Direct`
- end state of production switched over to `io_mode=DirectRw` can be
gauged by comparing old results' `io_mode=Direct` to new results'
`io_mode=DirectRw`
Given above guidance, on `im4gn.2xlarge`
- immediate impact is a significant improvement in all cases
- end state after switching has same significant improvements in all
cases
- ... except `ingest/io_mode=DirectRw volume_mib=128 key_size_bytes=8192
key_layout=Sequential write_delta=Yes` which only achieves `238 MiB/s`
instead of `253.43 MiB/s`
- this is a 6% degradation
- this workload is typical for image layer creation
# Refs
- epic https://github.com/neondatabase/neon/issues/9868
- stacked atop
- preliminary refactor https://github.com/neondatabase/neon/pull/11549
- bench_ingest overhaul https://github.com/neondatabase/neon/pull/11667
- derived from https://github.com/neondatabase/neon/pull/10063
Co-authored-by: Yuchen Liang <yuchen@neon.tech>
Main change:
- `BufferedWriter` owns the `W`; no more `Arc<W>`
- We introduce auto-delete-on-drop wrappers for `VirtualFile`.
- `TempVirtualFile` for write-only users
- `TempVirtualFileCoOwnedByEphemeralFileAndBufferedWriter` for
EphemeralFile which requires read access to the immutable prefix of the
file (see doc comments for details)
- Users of `BufferedWriter` hand it such a wrapped `VirtualFile`.
- The wrapped `VirtualFile` moves to the background flush task.
- On `BufferedWriter` shutdown, ownership moves back.
- Callers remove the wrapper (`disarm_into_inner()`) after doing final
touches, e.g., flushing index blocks and summary for delta/image layer
writers.
If the BufferedWriter isn't shut down properly via
`BufferedWriter::shutdown`, or if there is an error during final
touches, the wrapper type ensures that the file gets unlinked.
We store a GateGuard inside the wrapper to ensure that the Timeline is
still alive when unlinking on drop.
Rust doesn't have async drop yet, so, the unlinking happens using a
synchronous syscall.
NB we don't fsync the surrounding directory.
This is how it's been before this PR; I believe it is correct because
all of these files are temporary paths that get cleaned up on timeline
load.
Again, timeline load does not need to fsync because the next timeline
load will unlink again if the file reappears.
The auto-delete-on-drop can happen after a higher-level mechanism
retries.
Therefore, we switch all users to monotonically increasing, never-reused
temp file disambiguators.
The aspects pointed out in the last two paragraphs will receive further
cleanup in follow-up task
- https://github.com/neondatabase/neon/issues/11692
Drive-by changes:
- It turns out we can remove the two-pronged code in the layer file
download code.
No need to make this a separate PR because all of production already
uses `tokio-epoll-uring` with the buffered writer for many weeks.
Refs
- epic https://github.com/neondatabase/neon/issues/9868
- alternative to https://github.com/neondatabase/neon/pull/11544
Adds a versioning API to remote_storage. We want to use it in the
scrubber, both for tenant snapshot as well as for metadata checks.
for #8830
and for #11588
## Problem
Pageservers notify control plane directly when a shard import has
completed.
Control plane has to download the status of each shard from S3 and
figure out if everything is truly done,
before proceeding with branch activation.
Issues with this approach are:
* We can't control shard split behaviour on the storage controller side.
It's unsafe to split
during import.
* Control plane needs to know about shards and implement logic to check
all timelines are indeed ready.
## Summary of changes
In short, storage controller coordinates imports, and, only when
everything is done, notifies control plane.
Big rocks:
1. Store timeline imports in the storage controller database. Each
import stores the status of its shards in the database.
We hook into the timeline creation call as our entry point for this.
2. Pageservers get a new upcall endpoint to notify the storage
controller of shard import updates.
3. Storage controller handles these updates by updating persisted state.
If an update finalizes the import,
then poll pageservers until timeline activation, and, then, notify the
control plane that the import is complete.
Cplane side change with new endpoint is in
https://github.com/neondatabase/cloud/pull/26166
Closes https://github.com/neondatabase/neon/issues/11566
## Problem
If all batched requests are excluded from the query by
`Timeine::get_rel_page_at_lsn_batched` (e.g. because they are past the
end of the relation), the read path would panic since it doesn't expect
empty queries. This is a change in behaviour that was introduced with
the scattered query implementation.
## Summary of Changes
Handle empty queries explicitly.
## Problem
Pageservers and safakeepers do not pass CA certificates to broker
client, so the client do not trust locally issued certificates.
- Part of https://github.com/neondatabase/cloud/issues/27492
## Summary of changes
- Change `ssl_ca_certs` type in PS/SK's config to `Pem` which may be
converted to both `reqwest` and `tonic` certificates.
- Pass CA certificates to storage broker client in PS and SK
## Problem
follow-up on https://github.com/neondatabase/neon/pull/11601
## Summary of changes
- serialize the start/end time using rfc3339 time string
- compute the size ratio of the compaction
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
`Tenant` isn't really a whole tenant: it's just one shard of a tenant.
## Summary of changes
- Automated rename of Tenant to TenantShard
- Followup commit to change references in comments
This is mostly a documentation update, but a few updates with regard to
neon_local, pageserver, and tests.
17 is our default for users in production, so dropping references to 16
makes sense.
Signed-off-by: Tristan Partin <tristan@neon.tech>
Signed-off-by: Tristan Partin <tristan@neon.tech>
## Problem
We currently only have gc-compaction statistics for each single
sub-compaction job.
## Summary of changes
Add meta statistics across all sub-compaction jobs scheduled.
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
During shard ancestor compaction, we currently recompress all page
images as we move them into a new layer file. This is expensive and
unnecessary.
Resolves#11562.
Requires #11607.
## Summary of changes
Pass through compressed page images in `ImageLayerInner::filter()`.
