For a while already, we've been unable to update the Azure SDK crates
due to Azure adopting use of a non-tokio async runtime, see #7545.
The effort to upstream the fix got stalled, and I think it's better to
switch to a patched version of the SDK that is up to date.
Now we have a fork of the SDK under the neondatabase github org, to
which I have applied Conrad's rebased patches to:
https://github.com/neondatabase/azure-sdk-for-rust/tree/neon .
The existence of a fork will also help with shipping bulk delete support
before it's upstreamed (#7931).
Also, in related news, the Azure SDK has gotten a rift in development,
where the main branch pertains to a future, to-be-officially-blessed
release of the SDK, and the older versions, which we are currently
using, are on the `legacy` branch. Upstream doesn't really want patches
for the `legacy` branch any more, they want to focus on the `main`
efforts. However, even then, the `legacy` branch is still newer than
what we are having right now, so let's switch to `legacy` for now.
Depending on how long it takes, we can switch to the official version of
the SDK once it's released or switch to the upstream `main` branch if
there is changes we want before that.
As a nice side effect of this PR, we now use reqwest 0.12 everywhere,
dropping the dependency on version 0.11.
Fixes#7545
Result of running:
cargo update -p aws-types -p aws-sigv4 -p aws-credential-types -p
aws-smithy-types -p aws-smithy-async -p aws-sdk-kms -p aws-sdk-iam -p
aws-sdk-s3 -p aws-config
We want to keep the AWS SDK up to date as that way we benefit from new
developments and improvements.
Like #9931 but without rebasing upstream just yet, to try and minimise
the differences.
Removes all proxy-specific commits from the rust-postgres fork, now that
proxy no longer depends on them. Merging upstream changes to come later.
(stacked on #9990 and #9995)
Partially fixes#1287 with a custom option field to enable the fixed
behaviour. This allows us to gradually roll out the fix without silently
changing the observed behaviour for our customers.
related to https://github.com/neondatabase/cloud/issues/15284
Keeping the `mock` postgres cplane adaptor using "stock" tokio-postgres
allows us to remove a lot of dead weight from our actual postgres
connection logic.
## 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.
#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
To add Safekeeper heap profiling in #9778, we need to switch to an
allocator that supports it. Pageserver and proxy already use jemalloc.
Touches #9534.
## Summary of changes
Use jemalloc in Safekeeper.
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
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
* The futures-util crate we use was yanked. Bump it and its siblings to
new patch release.
https://github.com/rust-lang/futures-rs/releases/tag/0.3.31
* cargo-deny: Drop an unused license.
* cargo-deny: Don't warn about duplicate crate. Duplicate crates are
unavoidable and the noise just hides real warnings.
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
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)
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
We are pining our fork of rust-postgres to a commit hash and that
prevents us from making
further changes to it. The latest commit in rust-postgres requires
https://github.com/neondatabase/neon/pull/8747,
but that seems to have gone stale. I reverted rust-postgres `neon`
branch to the pinned commit in
https://github.com/neondatabase/rust-postgres/pull/31.
## Summary of changes
Switch back to using the `neon` branch of the rust-postgres fork.
and add /metrics endpoint to compute_ctl to expose such metrics
metric format example for extension pg_rag
with versions 1.2.3 and 1.4.2
installed in 3 and 1 databases respectively:
neon_extensions_installed{extension="pg_rag", version="1.2.3"} = 3
neon_extensions_installed{extension="pg_rag", version="1.4.2"} = 1
------
infra part: https://github.com/neondatabase/flux-fleet/pull/251
---------
Co-authored-by: Tristan Partin <tristan@neon.tech>
While setting up some tests, I noticed that we didn't support keycloak.
They make use of encryption JWKs as well as signature ones. Our current
jwks crate does not support parsing encryption keys which caused the
entire jwk set to fail to parse. Switching to lazy parsing fixes this.
Also while setting up tests, I couldn't use localhost jwks server as we
require HTTPS and we were using webpki so it was impossible to add a
custom CA. Enabling native roots addresses this possibility.
I saw some of our current e2e tests against our custom JWKS in s3 were
taking a while to fetch. I've added a timeout + retries to address this.
## Problem
We don't have any benchmarks for Safekeeper WAL ingestion.
## Summary of changes
Add some basic benchmarks for WAL ingestion, specifically for
`SafeKeeper::process_msg()` (single append) and `WalAcceptor` (pipelined
batch ingestion). Also add some baseline file write benchmarks.
The PROXY Protocol V2 offers a "command" concept. It can be of two
different values. "Local" and "Proxy". The spec suggests that "Local" be
used for health-checks. We can thus use this to silence logging for such
health checks such as those from NLB.
This additionally refactors the flow to be a bit more type-safe, self
documenting and using zerocopy deser.
AWS/azure private link shares extra information in the "TLV" values of
the proxy protocol v2 header. This code doesn't action on it, but it
parses it as appropriate.
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
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
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>
The forever ongoing effort of juggling multiple versions of rustls :3
now with new crypto library aws-lc.
Because of dependencies, it is currently impossible to not have both
ring and aws-lc in the dep tree, therefore our only options are not
updating rustls or having both crypto backends enabled...
According to benchmarks run by the rustls maintainer, aws-lc is faster
than ring in some cases too <https://jbp.io/graviola/>, so it's not
without its upsides,
part of https://github.com/neondatabase/neon/issues/9255
## Summary of changes
Upgrade remote_storage crate to use hyper1. Hyper0 is used when
providing the streaming HTTP body to the s3 SDK, and it is refactored to
use hyper1.
Signed-off-by: Alex Chi Z <chi@neon.tech>
Update hyper and tonic again in the storage broker, this time with a fix
for the issue that made us revert the update last time.
The first commit is a revert of #9268, the second a fix for the issue.
fixes#9231.
1. Adds local-proxy to compute image and vm spec
2. Updates local-proxy config processing, writing PID to a file eagerly
3. Updates compute-ctl to understand local proxy compute spec and to
send SIGHUP to local-proxy over that pid.
closes https://github.com/neondatabase/cloud/issues/16867
## Problem
The S3 tests couldn't use SSO authentication for local tests against S3.
## Summary of changes
Enable the `sso` feature of `aws-config`. Also run `cargo hakari
generate` which made some updates to `workspace_hack`.
Because:
- it's nice to be up-to-date,
- we already had axum 0.7 in our dependency tree, so this avoids having
to compile two versions, and
- removes one of the remaining dpendencies to hyper version 0
Also bumps the 'tokio-tungstenite' dependency, to avoid having two
versions in the dependency tree.
Fixes#9231 .
Upgrade hyper to 1.4.0 and use hyper 1.4 instead of 0.14 in the storage
broker, together with tonic 0.12. The two upgrades go hand in hand.
Thanks to the broker being independent from other components, we can
upgrade its hyper version without touching the other components, which
makes things easier.
* tracing-utils now returns a `Layer` impl. Removes the need for crates
to
import OTel crates.
* Drop the /v1/traces URI check. Verified that the code does the right
thing.
* Leave a TODO to hook in an error handler for OTel to log errors to
when it
assumes the regular pipeline cannot be used/is broken.
Part of #7497, closes#8817.
## Problem
See #8817.
## Summary of changes
**compute_ctl**
- Renew lsn lease as soon as `/configure` updates pageserver_connstr,
use `state_changed` Condvar for synchronization.
**pageserver**
As mentioned in
https://github.com/neondatabase/neon/issues/8817#issuecomment-2315768076,
we still want some permanent error reported if a lease cannot be
granted. By considering attachment mode and the added
`lsn_lease_deadline` when processing lease requests, we can also bound
the case of bad requests to a very short period after migration/restart.
- Refactor https://github.com/neondatabase/neon/pull/9024 and move
`lsn_lease_deadline` to `AttachedTenantConf` so timeline can easily
access it.
- Have separate HTTP `init_lsn_lease` and libpq `renew_lsn_lease` API.
- Always do LSN verification for the initial HTTP lease request.
- LSN verification for the renewal is **still done** when tenants are
not in `AttachedSingle` and we have pass the `lsn_lease_deadline`, which
give plenty of time for compute to renew the lease.
**neon_local**
- add and call `timeline_init_lsn_lease` mgmt_api at static endpoint
start. The initial lsn lease http request is sent when we run `cargo
neon endpoint start <static endpoint>`.
## Testing
- Extend `test_readonly_node_gc` to do pageserver restarts and
migration.
## Future Work
- The control plane should make the initial lease request through HTTP
when creating a static endpoint. This is currently only done in
`neon_local`.
Signed-off-by: Yuchen Liang <yuchen@neon.tech>
