## Problem
Read replicas cannot grant permissions for roles for Neon RLS. Usually
the permission is already granted, so we can optimistically check. See
INC-509
## Summary of changes
Perform a permission lookup prior to actually executing any grants.
Add `/lfc/(prewarm|offload)` routes to `compute_ctl` which interact with
endpoint storage.
Add `prewarm_lfc_on_startup` spec option which, if enabled, downloads
LFC prewarm data on compute startup.
Resolves: https://github.com/neondatabase/cloud/issues/26343
## Problem
Broker supports only HTTP, no HTTPS
- Closes: https://github.com/neondatabase/cloud/issues/27492
## Summary of changes
- Add `listen_https_addr`, `ssl_key_file`, `ssl_cert_file`,
`ssl_cert_reload_period` arguments to storage broker
- Make `listen_addr` argument optional
- Listen https in storage broker
- Support https for storage broker request in neon_local
- Add `use_https_storage_broker_api` option to NeonEnvBuilder
## Problem
It seems are production-ready cert-manager setup now includes a full
certificate chain. This was not accounted for and the decoder would
error.
## Summary of changes
Change the way we decode certificates to support cert-chains, ignoring
all but the first cert.
This also changes a log line to not use multi-line errors.
~~I have tested this code manually against real certificates/keys, I
didn't want to embed those in a test just yet, not until the cert
expires in 24 hours.~~
# 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>
* Replace yanked papaya version
* Remove unused allowed license: OpenSSL
* Remove Zlib license from general allow list since it's listed in the
exceptions section per crate
* Drop clarification for ring since they have separate LICENSE files now
* List the tower-otel repo as allowed source while we sort out the OTel
deps
Switches the tenant snapshot subcommand of the storage scrubber to
`remote_storage`. As this is the last piece of the storage scrubber
still using the S3 SDK, this finishes the project started in #7547.
This allows us to do tenant snapshots on Azure as well.
Builds on #11671Fixes#8830
Update the sentry crate to 0.37. This deduplicates the `webpki-roots`
crate in our crate graph, and brings another dependency onto newer
rustls `0.23.18`.
## 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
Finally figured out the right incantation. I had had this in my original
go, but due to some refactoring and apparently missed testing, I
committed a mistake. The reason this doesn't currently break anything is
that we bypass the authorization middleware when the "testing" cargo
feature is enabled.
Signed-off-by: Tristan Partin <tristan@neon.tech>
## Problem
We need to export some metrics about certs/connections to configure
alerts and make sure that all HTTP requests are gone before turning
https-only mode on.
- Closes: https://github.com/neondatabase/cloud/issues/25526
## Summary of changes
- Add started connection and connection error metrics to http/https
Server.
- Add certificate expiration time and reload metrics to
ReloadingCertificateResolver.
Service targeted for storing and retrieving LFC prewarm data.
Can be used for proxying S3 access for Postgres extensions like
pg_mooncake as well.
Requests must include a Bearer JWT token.
Token is validated using a pemfile (should be passed in infra/).
Note: app is not tolerant to extra trailing slashes, see app.rs
`delete_prefix` test for comments.
Resolves: https://github.com/neondatabase/cloud/issues/26342
Unrelated changes: gate a `rename_noreplace` feature and disable it in
`remote_storage` so as `object_storage` can be built with musl
Based on https://github.com/neondatabase/neon/pull/11139
## Problem
We want to export performance traces from the pageserver in OTEL format.
End goal is to see them in Grafana.
## Summary of changes
https://github.com/neondatabase/neon/pull/11139 introduces the
infrastructure required to run the otel collector alongside the
pageserver.
### Design
Requirements:
1. We'd like to avoid implementing our own performance tracing stack if
possible and use the `tracing` crate if possible.
2. Ideally, we'd like zero overhead of a sampling rate of zero and be a
be able to change the tracing config for a tenant on the fly.
3. We should leave the current span hierarchy intact. This includes
adding perf traces without modifying existing tracing.
To satisfy (3) (and (2) in part) a separate span hierarchy is used.
`RequestContext` gains an optional `perf_span` member
that's only set when the request was chosen by sampling. All perf span
related methods added to `RequestContext` are no-ops for requests that
are not sampled.
This on its own is not enough for (3), so performance spans use a
separate tracing subscriber. The `tracing` crate doesn't have great
support for this, so there's a fair amount of boilerplate to override
the subscriber at all points of the perf span lifecycle.
### Perf Impact
[Periodic
pagebench](https://neonprod.grafana.net/d/ddqtbfykfqfi8d/e904990?orgId=1&from=2025-02-08T14:15:59.362Z&to=2025-03-10T14:15:59.362Z&timezone=utc)
shows no statistically significant regression with a sample ratio of 0.
There's an annotation on the dashboard on 2025-03-06.
### Overview of changes:
1. Clean up the `RequestContext` API a bit. Namely, get rid of the
`RequestContext::extend` API and use the builder instead.
2. Add pageserver level configs for tracing: sampling ratio, otel
endpoint, etc.
3. Introduce some perf span tracking utilities and expose them via
`RequestContext`. We add a `tracing::Span` wrapper to be used for perf
spans and a `tracing::Instrumented` equivalent for it. See doc comments
for reason.
4. Set up OTEL tracing infra according to configuration. A separate
runtime is used for the collector.
5. Add perf traces to the read path.
## Refs
- epic https://github.com/neondatabase/neon/issues/9873
---------
Co-authored-by: Christian Schwarz <christian@neon.tech>
In sqlstate, we have a manual `phf` construction, which is not
explicitly guaranteed to be stable - you're intended to use a build.rs
or the macro to make sure it's constructed correctly each time. This was
inherited from tokio-postgres upstream, which has the same issue
(https://github.com/rust-phf/rust-phf/pull/321#issuecomment-2724521193).
We don't need this encoding of sqlstate, so I've switched it to simply
parse 5 bytes
(https://www.postgresql.org/docs/current/errcodes-appendix.html).
While here, I switched out log for tracing.
## Problem
SSL certs are loaded only during start up. It doesn't allow the rotation
of short-lived certificates without server restart.
- Closes: https://github.com/neondatabase/cloud/issues/25525
## Summary of changes
- Implement `ReloadingCertificateResolver` which reloads certificates
from disk periodically.
## Problem
Pageservers use unencrypted HTTP requests for storage controller API.
- Closes: https://github.com/neondatabase/cloud/issues/25524
## Summary of changes
- Replace hyper0::server::Server with http_utils::server::Server in
storage controller.
- Add HTTPS handler for storage controller API.
- Support `ssl_ca_file` in pageserver.
Both crates seem well maintained. x509-cert is part of the high quality
RustCrypto project that we already make heavy use of, and I think it
makes sense to reduce the dependencies where possible.
Closes: https://github.com/neondatabase/cloud/issues/22998
If control-plane reports that TLS should be used, load the certificates
(and watch for updates), make sure postgres use them, and detects
updates.
Procedure:
1. Load certificates
2. Reconfigure postgres/pgbouncer
3. Loop on a timer until certificates have loaded
4. Go to 1
Notes:
1. We only run this procedure if requested on startup by control plane.
2. We needed to compile pgbouncer with openssl enabled
3. Postgres doesn't allow tls keys to be globally accessible - must be
read only to the postgres user. I couldn't convince the autoscaling team
to let me put this logic into the VM settings, so instead compute_ctl
will copy the keys to be read-only by postgres.
4. To mitigate a race condition, we also verify that the key matches the
cert.
## Problem
`humantime` is unmaintained, we want to migrate to `jiff`, see
https://github.com/neondatabase/neon/issues/11179.
`env_logger` in older versions depend on `humantime`, and newer versions
depend on `jiff`, so we need to update it.
## Summary of changes
Update `env_logger` to the most recent release, which does not depend on
`humantime` anymore.
We want to export performance traces from the pageserver in OTEL format.
End goal is to see them in Grafana.
To this end, there are two changes here:
1. Update the `tracing-utils` crate to allow for explicitly specifying
the export configuration. Pageserver configuration is loaded from a file
on start-up. This allows us to use the same flow for export configs
there.
2. Update the `utils::logging::init` common entry point to set up OTEL
tracing infrastructure if requested. Note that an entirely different
tracing subscriber is used. This is to avoid interference with the
existing tracing set-up. For now, no service uses this functionality.
PR to plug this into the pageserver is
[here](https://github.com/neondatabase/neon/pull/11140).
Related https://github.com/neondatabase/neon/issues/9873
The upgrade to 8.0.0 caused severe performance regressions in the
start_postgres_ms metric, which measures the time it takes from execing
Postgres to the time Postgres marks itself as ready in the
postmaster.pid file. We use the notify crate to watch for changes in the
pgdata directory and the postmaster.pid file.
Signed-off-by: Tristan Partin <tristan@neon.tech>
This PR extends the storcon with basic safekeeper management of
timelines, mainly timeline creation and deletion. We want to make the
storcon manage safekeepers in the future. Timeline creation is
controlled by the `--timelines-onto-safekeepers` flag.
1. it adds the `timelines` and `safekeeper_timeline_pending_ops` tables
to the storcon db
2. extend code for the timeline creation and deletion
4. it adds per-safekeeper reconciler tasks
TODO:
* maybe not immediately schedule reconciliations for deletions but have
a prior manual step
* tenant deletions
* add exclude API definitions (probably separate PR)
* how to choose safekeeper to do exclude on vs deletion? this can be a
bit hairy because the safekeeper might go offline in the meantime.
* error/failure case handling
* tests (cc test_explicit_timeline_creation from #11002)
* single safekeeper mode: we often only have one SK (in tests for
example)
* `notify-safekeepers` hook:
https://github.com/neondatabase/neon/issues/11163
TODOs implemented:
* cancellations of enqueued reconciliations on a per-timeline basis,
helpful if there is an ongoing deletion
* implement pending ops overwrite behavior
* load pending operations from db
RFC section for important reading:
[link](https://github.com/neondatabase/neon/blob/main/docs/rfcs/035-safekeeper-dynamic-membership-change.md#storage_controller-implementation)
Implements the bulk of #9011
Successor of #10440.
---------
Co-authored-by: Arseny Sher <sher-ars@yandex.ru>
* pprof can also use `prost` as a backend, switch to it as `protobuf`
has no update available but a security issue.
* `paste` is a build time dependency, so add the unmaintained warning as
an exception.
## Problem
I like using `cargo stress` to hammer on a test, but it doesn't work out
of the box because it does parallel runs by default and tests always use
the same repo dir.
## Summary of changes
Add an uuid to the test repo dir when generating it.
## Problem
Grafana Loki's JSON handling is somewhat limited and the log message
should be structured in a way that it's easy to sift through logs and
filter.
## Summary of changes
* Drop span_id. It's too short lived to be of value and only bloats the
logs.
* Use the span's name as the object key, but append a unique numeric
value to prevent name collisions.
* Extract interesting span fields into a separate object at the root.
New format:
```json
{
"timestamp": "2025-03-04T18:54:44.134435Z",
"level": "INFO",
"message": "connected to compute node at 127.0.0.1 (127.0.0.1:5432) latency=client: 22.002292ms, cplane: 0ns, compute: 5.338875ms, retry: 0ns",
"fields": {
"cold_start_info": "unknown"
},
"process_id": 56675,
"thread_id": 9122892,
"task_id": "24",
"target": "proxy::compute",
"src": "proxy/src/compute.rs:288",
"trace_id": "5eb89b840ec63fee5fc56cebd633e197",
"spans": {
"connect_request#1": {
"ep": "endpoint",
"role": "proxy",
"session_id": "b8a41818-12bd-4c3f-8ef0-9a942cc99514",
"protocol": "tcp",
"conn_info": "127.0.0.1"
},
"connect_to_compute#6": {},
"connect_once#8": {
"compute_id": "compute",
"pid": "853"
}
},
"extract": {
"session_id": "b8a41818-12bd-4c3f-8ef0-9a942cc99514"
}
}
```
The compute should only act if requests come from the control plane.
Signed-off-by: Tristan Partin <tristan@neon.tech>
Signed-off-by: Tristan Partin <tristan@neon.tech>
## Problem
The code to generate symbolized pprof heap profiles and flamegraph SVGs
has been upstreamed to the `jemalloc_pprof` crate:
* https://github.com/polarsignals/rust-jemalloc-pprof/pull/22
* https://github.com/polarsignals/rust-jemalloc-pprof/pull/23
## Summary of changes
Use `jemalloc_pprof` to generate symbolized pprof heap profiles and
flamegraph SVGs.
This reintroduces a bunch of internal jemalloc stack frames that we'd
previously strip, e.g. each stack now always ends with
`prof_backtrace_impl` (where jemalloc takes a stack trace for heap
profiling), but that seems ok.
## Problem
Incoming requests often take the service lock, and sometimes even do
database transactions. That creates a risk that a rogue client can
starve the controller of the ability to do its primary job of
reconciling tenants to an available state.
## Summary of changes
* Use the `governor` crate to rate limit tenant requests at 10 requests
per second. This is ~10-100x lower than the worst "attack" we've seen
from a client bug. Admin APIs are not rate limited.
* Add a `storage_controller_http_request_rate_limited` histogram for
rate limited requests.
* Log a warning every 10 seconds for rate limited tenants.
The rate limiter is parametrized on TenantId, because the kinds of
client bug we're protecting against generally happen within tenant
scope, and the rates should be somewhat stable: we expect the global
rate of requests to increase as we do more work, but we do not expect
the rate of requests to one tenant to increase.
---------
Co-authored-by: John Spray <john@neon.tech>
Migrates the remaining crates to edition 2024. We like to stay on the
latest edition if possible. There is no functional changes, however some
code changes had to be done to accommodate the edition's breaking
changes.
Like the previous migration PRs, this is comprised of three commits:
* the first does the edition update and makes `cargo check`/`cargo
clippy` pass. we had to update bindgen to make its output [satisfy the
requirements of edition
2024](https://doc.rust-lang.org/edition-guide/rust-2024/unsafe-extern.html)
* the second commit does a `cargo fmt` for the new style edition.
* the third commit reorders imports as a one-off change. As before, it
is entirely optional.
Part of #10918
We lost this with the switch to axum for the HTTP server. Add it back.
In addition to just resurrecting the functionality we had before, pass
the tracing context of the /configure HTTP request to the start_postgres
operation that runs in the main thread. This way, the 'start_postgres'
and all its sub-spans like getting the basebackup become children of the
HTTP request span. This allows end-to-end tracing of a compute start,
all the way from the proxy to the SQL queries executed by compute_ctl as
part of compute startup.
## Problem
close https://github.com/neondatabase/cloud/issues/24485
## Summary of changes
This patch adds a new chaos injection mode for the storcon. The chaos
injector reads the crontab and exits immediately at the configured time.
---------
Signed-off-by: Alex Chi Z <chi@neon.tech>
The synchronous 'postgres' crate is just a wrapper around the async
'tokio_postgres' crate. Some places were unnecessarily using the
re-exported NoTls and Error from the synchronous 'postgres' crate, even
though they were otherwise using the 'tokio_postgres' crate. Tidy up by
using the tokio_postgres types directly.
## Problem
The storage controller treats durations in the tenant config as strings.
These are loaded from the db.
The pageserver maps these durations to a seconds only format and we
always get a mismatch compared
to what's in the db.
## Summary of changes
Treat durations as durations inside the storage controller and not as
strings.
Nothing changes in the cross service API's themselves or the way things
are stored in the db.
I also added some logging which I would have made the investigation a
10min job:
1. Reason for why the reconciliation was spawned
2. Location config diff between the observed and wanted states
## Problem
We don't have good observability for "stuck" getpage requests.
Resolves https://github.com/neondatabase/cloud/issues/23808.
## Summary of changes
Log a periodic warning (every 30 seconds) if GetPage request execution
is slow to complete, to aid in debugging stuck GetPage requests.
This does not cover response flushing (we have separate logging for
that), nor reading the request from the socket and batching it (expected
to be insignificant and not straightforward to handle with the current
protocol).
This costs 95 nanoseconds on the happy path when awaiting a
`tokio::task::yield_now()`:
```
warn_slow/enabled=false time: [45.716 ns 46.116 ns 46.687 ns]
warn_slow/enabled=true time: [141.53 ns 141.83 ns 142.18 ns]
```
## Problem
We'd like to enable CPU/heap profiling for storcon. This requires
jemalloc.
## Summary of changes
Use jemalloc as the global allocator, and enable heap sampling for
profiling.
This replaces the use of the awscli utility. awscli binary is massive,
it added about 200 MB to the docker image size, while the s3 client was
already a dependency so using that is essentially free, as far as binary
size is concerned.
I implemented a simple upload function that tries to keep 10 uploads
going in parallel. I believe that's the default behavior of the "aws s3
sync" command too.
Originally I wanted to switch back to the `neon` branch before merging
#10825, but I forgot to do it. Do it in a separate PR now.
No actual change of the source code, only changes the branch name (so
that maybe in a few weeks we can delete the temporary branch
`arpad/neon-rebase`).
