We also don't need to return the 'rest' slice, we can just insert it into the vec
Lastly, we can simplify the hotloop by making write_char_escape cold and moving the vec write inside this fn
## Problem
We currently offload LFC state unconditionally, which can cause
problems. Imagine a situation:
1. Endpoint started with `autoprewarm: true`.
2. While prewarming is not completed, we upload the new incomplete
state.
3. Compute gets interrupted and restarts.
4. We start again and try to prewarm with the state from 2. instead of
the previous complete state.
During the orchestrated prewarming, it's probably not a big issue, but
it's still better to do not interfere with the prewarm process.
## Summary of changes
Do not offload LFC state if we are currently prewarming or any issue
occurred. While on it, also introduce `Skipped` LFC prewarm status,
which is used when the corresponding LFC state is not present in the
endpoint storage. It's primarily needed to distinguish the first compute
start for particular endpoint, as it's completely valid to do not have
LFC state yet.
## Problem
Previously, if a get page failure was cause by timeline shutdown, the
pageserver would attempt to tear down the connection gracefully:
`shutdown(SHUT_WR)` followed by `close()`.
This triggers a code path on the compute where it has to tell apart
between an idle connection and a closed one. That code is bug prone, so
we can just side-step the issue by shutting down the connection via a
libpq error message.
This surfaced as instability in test_shard_resolve_during_split_abort.
It's a new test, but the issue existed for ages.
## Summary of Changes
Send a libpq error message instead of doing graceful TCP connection
shutdown.
Closes LKB-648
Putting this in the neon codebase for now, to experiment. Can be lifted
into measured at a later date.
This metric family is like a MetricVec, but it only supports 1 label
being set at a time. It is useful for reporting info, rather than
reporting metrics.
https://www.robustperception.io/exposing-the-software-version-to-prometheus/
Second PR for the hashmap behind the updated LFC implementation ([see
first here](https://github.com/neondatabase/neon/pull/12595)). This only
adds the raw code for the hashmap/lock implementations and doesn't plug
it into the crate (that's dependent on the previous PR and should
probably be done when the full integration into the new communicator is
merged alongside `communicator-rewrite` changes?).
Some high level details: the communicator codebase expects to be able to
store references to entries within this hashmap for arbitrary periods of
time and so the hashmap cannot be allowed to move them during a rehash.
As a result, this implementation has a slightly unusual structure where
key-value pairs (and hash chains) are allocated in a separate region
with a freelist. The core hashmap structure is then an array of
"dictionary entries" that are just indexes into this region of key-value
pairs.
Concurrency support is very naive at the moment with the entire map
guarded by one big `RwLock` (which is implemented on top of a
`pthread_rwlock_t` since Rust doesn't guarantee that a
`std::sync::RwLock` is safe to use in shared memory). This (along with a
lot of other things) is being changed on the
`quantumish/lfc-resizable-map` branch.
## Problem
We were only resetting the limit in the wal proposer. If backends are
back pressured, it might take a while for the wal proposer to receive a
new WAL to reset the limit.
## Summary of changes
Backend also checks the time and resets the limit.
## How is this tested?
pgbench has more smooth tps
Signed-off-by: Tristan Partin <tristan.partin@databricks.com>
Co-authored-by: Haoyu Huang <haoyu.huang@databricks.com>
Initial PR for the hashmap behind the updated LFC implementation. This
refactors `neon-shmem` so that the actual shared memory utilities are in
a separate module within the crate. Beyond that, it slightly changes
some of the docstrings so that they play nicer with `cargo doc`.
We don't want to depend on postgres_ffi in an API crate. If there is no
such dependency, we can compile stuff like `storcon_cli` without needing
a full working postgres build. Fixes regression of #12548 (before we
could compile it).
# TLDR
This PR is a no-op.
## Problem
When a SK loses a disk, it must recover all WALs from the very
beginning. This may take days/weeks to catch up to the latest WALs for
all timelines it owns.
## Summary of changes
When SK starts up,
if it finds that it has 0 timelines,
- it will ask SC for the timeline it owns.
- Then, pulls the timeline from its peer safekeepers to restore the WAL
redundancy right away.
After pulling timeline is complete, it will become active and accepts
new WALs.
The current impl is a prototype. We can optimize the impl further, e.g.,
parallel pull timelines.
---------
Co-authored-by: Haoyu Huang <haoyu.huang@databricks.com>
## Problem
gRPC client retries currently include pool acquisition under the
per-attempt timeout. If pool acquisition is slow (e.g. full pool), this
will cause spurious timeout warnings, and the caller will lose its place
in the pool queue.
Touches #11735.
## Summary of changes
Makes several improvements to retries and related logic:
* Don't include pool acquisition time under request timeouts.
* Move attempt timeouts out of `Retry` and into the closure.
* Make `Retry` configurable, move constants into main module.
* Don't backoff on the first retry, and reduce initial/max backoffs to
5ms and 5s respectively.
* Add `with_retries` and `with_timeout` helpers.
* Add slow logging for pool acquisition, and a `warn_slow` counterpart
to `log_slow`.
* Add debug logging for requests and responses at the client boundary.
# TLDR
Problem-I is a bug fix. The rest are no-ops.
## Problem I
Page server checks image layer creation based on the elapsed time but
this check depends on the current logical size, which is only computed
on shard 0. Thus, for non-0 shards, the check will be ineffective and
image creation will never be done for idle tenants.
## Summary of changes I
This PR fixes the problem by simply removing the dependency on current
logical size.
## Summary of changes II
This PR adds a timeout when calling page server to split shard to make
sure SC does not wait for the API call forever. Currently the PR doesn't
adds any retry logic because it's not clear whether page server shard
split can be safely retried if the existing operation is still ongoing
or left the storage in a bad state. Thus it's better to abort the whole
operation and restart.
## Problem III
`test_remote_failures` requires PS to be compiled in the testing mode.
For PS in dev/staging, they are compiled without this mode.
## Summary of changes III
Remove the restriction and also increase the number of total failures
allowed.
## Summary of changes IV
remove test on PS getpage http route.
---------
Co-authored-by: Chen Luo <chen.luo@databricks.com>
Co-authored-by: Yecheng Yang <carlton.yang@databricks.com>
Co-authored-by: Vlad Lazar <vlad@neon.tech>
This is a no-op for the neon deployment
* Introduce the concept image consistent lsn: of the largest LSN below
which all pages have been redone successfully
* Use the image consistent LSN for forced image layer creations
* Optionally expose the image consistent LSN via the timeline describe
HTTP endpoint
* Add a sharded timeline describe endpoint to storcon
---------
Co-authored-by: Chen Luo <chen.luo@databricks.com>
## Problem
We have a `safekeeper_migrate` handler, but no subcommand in
`storcon_cli`. Same for `/:timeline_id/locate` for identifying current
set of safekeepers.
- Closes: https://github.com/neondatabase/neon/issues/12395
## Summary of changes
- Add `timeline-safekeeper-migrate` and `timeline-locate` subcommands to
`storcon_cli`
## Problem
One PG tenant may write too fast and overwhelm the PS. The other tenants
sharing the same PSs will get very little bandwidth.
We had one experiment that two tenants sharing the same PSs. One tenant
runs a large ingestion that delivers hundreds of MB/s while the other
only get < 10 MB/s.
## Summary of changes
Rate limit how fast PG can generate WALs. The default is -1. We may
scale the default value with the CPU count. Need to run some experiments
to verify.
## How is this tested?
CI.
PGBench. No limit first. Then set to 1 MB/s and you can see the tps
drop. Then reverted the change and tps increased again.
pgbench -i -s 10 -p 55432 -h 127.0.0.1 -U cloud_admin -d postgres
pgbench postgres -c 10 -j 10 -T 6000000 -P 1 -b tpcb-like -h 127.0.0.1
-U cloud_admin -p 55432
progress: 33.0 s, 986.0 tps, lat 10.142 ms stddev 3.856 progress: 34.0
s, 973.0 tps, lat 10.299 ms stddev 3.857 progress: 35.0 s, 1004.0 tps,
lat 9.939 ms stddev 3.604 progress: 36.0 s, 984.0 tps, lat 10.183 ms
stddev 3.713 progress: 37.0 s, 998.0 tps, lat 10.004 ms stddev 3.668
progress: 38.0 s, 648.9 tps, lat 12.947 ms stddev 24.970 progress: 39.0
s, 0.0 tps, lat 0.000 ms stddev 0.000 progress: 40.0 s, 0.0 tps, lat
0.000 ms stddev 0.000 progress: 41.0 s, 0.0 tps, lat 0.000 ms stddev
0.000 progress: 42.0 s, 0.0 tps, lat 0.000 ms stddev 0.000 progress:
43.0 s, 0.0 tps, lat 0.000 ms stddev 0.000 progress: 44.0 s, 0.0 tps,
lat 0.000 ms stddev 0.000 progress: 45.0 s, 0.0 tps, lat 0.000 ms stddev
0.000 progress: 46.0 s, 0.0 tps, lat 0.000 ms stddev 0.000 progress:
47.0 s, 0.0 tps, lat 0.000 ms stddev 0.000 progress: 48.0 s, 0.0 tps,
lat 0.000 ms stddev 0.000 progress: 49.0 s, 347.3 tps, lat 321.560 ms
stddev 1805.633 progress: 50.0 s, 346.8 tps, lat 9.898 ms stddev 3.809
progress: 51.0 s, 0.0 tps, lat 0.000 ms stddev 0.000 progress: 52.0 s,
0.0 tps, lat 0.000 ms stddev 0.000 progress: 53.0 s, 0.0 tps, lat 0.000
ms stddev 0.000 progress: 54.0 s, 0.0 tps, lat 0.000 ms stddev 0.000
progress: 55.0 s, 0.0 tps, lat 0.000 ms stddev 0.000 progress: 56.0 s,
0.0 tps, lat 0.000 ms stddev 0.000 progress: 57.0 s, 0.0 tps, lat 0.000
ms stddev 0.000 progress: 58.0 s, 0.0 tps, lat 0.000 ms stddev 0.000
progress: 59.0 s, 0.0 tps, lat 0.000 ms stddev 0.000 progress: 60.0 s,
0.0 tps, lat 0.000 ms stddev 0.000 progress: 61.0 s, 0.0 tps, lat 0.000
ms stddev 0.000 progress: 62.0 s, 0.0 tps, lat 0.000 ms stddev 0.000
progress: 63.0 s, 494.5 tps, lat 276.504 ms stddev 1853.689 progress:
64.0 s, 488.0 tps, lat 20.530 ms stddev 71.981 progress: 65.0 s, 407.8
tps, lat 9.502 ms stddev 3.329 progress: 66.0 s, 0.0 tps, lat 0.000 ms
stddev 0.000 progress: 67.0 s, 0.0 tps, lat 0.000 ms stddev 0.000
progress: 68.0 s, 504.5 tps, lat 71.627 ms stddev 397.733 progress: 69.0
s, 371.0 tps, lat 24.898 ms stddev 29.007 progress: 70.0 s, 541.0 tps,
lat 19.684 ms stddev 24.094 progress: 71.0 s, 342.0 tps, lat 29.542 ms
stddev 54.935
Co-authored-by: Haoyu Huang <haoyu.huang@databricks.com>
After https://github.com/neondatabase/neon/pull/12240 we observed
issues in our go code as `ComputeStatus` is not stateless, thus doesn't
deserialize as string.
```
could not check compute activity: json: cannot unmarshal object into Go struct field
ComputeState.status of type computeclient.ComputeStatus
```
- Fix this by splitting this status into two.
- Update compute OpenApi spec to reflect changes to `/terminate` in
previous PR
## Problem
Safekeeper and pageserver metrics collection might time out. We've seen
this in both hadron and neon.
## Summary of changes
This PR moves metrics collection in PS/SK to the background so that we
will always get some metrics, despite there may be some delays. Will
leave it to the future work to reduce metrics collection time.
---------
Co-authored-by: Chen Luo <chen.luo@databricks.com>
This PR introduces a `image_creation_timeout` to page servers so that we
can force the image creation after a certain period. This is set to 1
day on dev/staging for now, and will rollout to production 1/2 weeks
later.
Majority of the PR are boilerplate code to add the new knob. Specific
changes of the PR are:
1. During L0 compaction, check if we should force a compaction if
min(LSN) of all delta layers < force_image_creation LSN.
2. During image creation, check if we should force a compaction if the
image's LSN < force_image_creation LSN and there are newer deltas with
overlapping key ranges.
3. Also tweaked the check image creation interval to make sure we honor
image_creation_timeout.
Vlad's note: This should be a no-op. I added an extra PS config for the
large timeline
threshold to enable this.
---------
Co-authored-by: Chen Luo <chen.luo@databricks.com>
Change the unreliable storage wrapper to fail by probability when there
are more failure attempts left.
Co-authored-by: Yecheng Yang <carlton.yang@databricks.com>
The `--timelines-onto-safekeepers` flag is very consequential in the
sense that it controls every single timeline creation. However, we don't
have any automatic insight whether enabling the option will break things
or not.
The main way things can break is by misconfigured safekeepers, say they
are marked as paused in the storcon db. The best input so far we can
obtain via manually connecting via storcon_cli and listing safekeepers,
but this is cumbersome and manual so prone to human error.
So at storcon startup, do a simulated "test creation" in which we call
`timelines_onto_safekeepers` with the configuration provided to us, and
print whether it was successful or not. No actual timeline is created,
and nothing is written into the storcon db. The heartbeat info will not
have reached us at that point yet, but that's okay, because we still
fall back to safekeepers that don't have any heartbeat.
Also print some general scheduling policy stats on initial safekeeper
load.
Part of #11670.
## Problem
For the communicator, we need a rich Pageserver gRPC client.
Touches #11735.
Requires #12434.
## Summary of changes
This patch adds an initial rich Pageserver gRPC client. It supports:
* Sharded tenants across multiple Pageservers.
* Pooling of connections, clients, and streams for efficient resource
use.
* Concurrent use by many callers.
* Internal handling of GetPage bidirectional streams, with pipelining
and error handling.
* Automatic retries.
* Observability.
The client is still under development. In particular, it needs GetPage
batch splitting, shard map updates, and performance optimization. This
will be addressed in follow-up PRs.
## Problem
We lost capability to explicitly disable the global eviction task (for
testing).
## Summary of changes
Add an `enabled` flag to `DiskUsageEvictionTaskConfig` to indicate
whether we should run the eviction job or not.
# TLDR
All changes are no-op except
1. publishing additional metrics.
2. problem VI
## Problem I
It has come to my attention that the Neon Storage Controller doesn't
correctly update its "observed" state of tenants previously associated
with PSs that has come back up after a local data loss. It would still
think that the old tenants are still attached to page servers and won't
ask more questions. The pageserver has enough information from the
reattach request/response to tell that something is wrong, but it
doesn't do anything about it either. We need to detect this situation in
production while I work on a fix.
(I think there is just some misunderstanding about how Neon manages
their pageserver deployments which got me confused about all the
invariants.)
## Summary of changes I
Added a `pageserver_local_data_loss_suspected` gauge metric that will be
set to 1 if we detect a problematic situation from the reattch response.
The problematic situation is when the PS doesn't have any local tenants
but received a reattach response containing tenants.
We can set up an alert using this metric. The alert should be raised
whenever this metric reports non-zero number.
Also added a HTTP PUT
`http://pageserver/hadron-internal/reset_alert_gauges` API on the
pageserver that can be used to reset the gauge and the alert once we
manually rectify the situation (by restarting the HCC).
## Problem II
Azure upload is 3x slower than AWS. -> 3x slower ingestion.
The reason for the slower upload is that Azure upload in page server is
much slower => higher flush latency => higher disk consistent LSN =>
higher back pressure.
## Summary of changes II
Use Azure put_block API to uploads a 1 GB layer file in 8 blocks in
parallel.
I set the put_block block size to be 128 MB by default in azure config.
To minimize neon changes, upload function passes the layer file path to
the azure upload code through the storage metadata. This allows the
azure put block to use FileChunkStreamRead to stream read from one
partition in the file instead of loading all file data in memory and
split it into 8 128 MB chunks.
## How is this tested? II
1. rust test_real_azure tests the put_block change.
3. I deployed the change in azure dev and saw flush latency reduces from
~30 seconds to 10 seconds.
4. I also did a bunch of stress test using sqlsmith and 100 GB TPCDS
runs.
## Problem III
Currently Neon limits the compaction tasks as 3/4 * CPU cores. This
limits the overall compaction throughput and it can easily cause
head-of-the-line blocking problems when a few large tenants are
compacting.
## Summary of changes III
This PR increases the limit of compaction tasks as `BG_TASKS_PER_THREAD`
(default 4) * CPU cores. Note that `CONCURRENT_BACKGROUND_TASKS` also
limits some other tasks `logical_size_calculation` and `layer eviction`
. But compaction should be the most frequent and time-consuming task.
## Summary of changes IV
This PR adds the following PageServer metrics:
1. `pageserver_disk_usage_based_eviction_evicted_bytes_total`: captures
the total amount of bytes evicted. It's more straightforward to see the
bytes directly instead of layers.
2. `pageserver_active_storage_operations_count`: captures the active
storage operation, e.g., flush, L0 compaction, image creation etc. It's
useful to visualize these active operations to get a better idea of what
PageServers are spending cycles on in the background.
## Summary of changes V
When investigating data corruptions, it's useful to search the base
image and all WAL records of a page up to an LSN, i.e., a breakdown of
GetPage@LSN request. This PR implements this functionality with two
tools:
1. Extended `pagectl` with a new command to search the layer files for a
given key up to a given LSN from the `index_part.json` file. The output
can be used to download the files from S3 and then search the file
contents using the second tool.
Example usage:
```
cargo run --bin pagectl index-part search --tenant-id 09b99ea3239bbb3b2d883a59f087659d --timeline-id 7bedf4a6995baff7c0421ff9aebbcdab --path ~/Downloads/corruption/index_part.json-0000000c-formatted --key 000000067F000080140000802100000D61BD --lsn 70C/BF3D61D8
```
Example output:
```
tenants/09b99ea3239bbb3b2d883a59f087659d-0304/timelines/7bedf4a6995baff7c0421ff9aebbcdab/000000067F0000801400000B180000000002-000000067F0000801400008028000002FEFF__000007089F0B5381-0000070C7679EEB9-0000000c
tenants/09b99ea3239bbb3b2d883a59f087659d-0304/timelines/7bedf4a6995baff7c0421ff9aebbcdab/000000000000000000000000000000000000-000000067F0000801400008028000002F3F1__000006DD95B6F609-000006E2BA14C369-0000000c
tenants/09b99ea3239bbb3b2d883a59f087659d-0304/timelines/7bedf4a6995baff7c0421ff9aebbcdab/000000067F0000801400000B180000000002-000000067F000080140000802100001B0973__000006D33429F539-000006DD95B6F609-0000000c
tenants/09b99ea3239bbb3b2d883a59f087659d-0304/timelines/7bedf4a6995baff7c0421ff9aebbcdab/000000067F0000801400000B180000000002-000000067F00008014000080210000164D81__000006C6343B2D31-000006D33429F539-0000000b
tenants/09b99ea3239bbb3b2d883a59f087659d-0304/timelines/7bedf4a6995baff7c0421ff9aebbcdab/000000067F0000801400000B180000000002-000000067F0000801400008021000017687B__000006BA344FA7F1-000006C6343B2D31-0000000b
tenants/09b99ea3239bbb3b2d883a59f087659d-0304/timelines/7bedf4a6995baff7c0421ff9aebbcdab/000000067F0000801400000B180000000002-000000067F00008014000080210000165BAB__000006AD34613D19-000006BA344FA7F1-0000000b
tenants/09b99ea3239bbb3b2d883a59f087659d-0304/timelines/7bedf4a6995baff7c0421ff9aebbcdab/000000067F0000801400000B180000000002-000000067F00008014000080210000137A39__0000069F34773461-000006AD34613D19-0000000b
tenants/09b99ea3239bbb3b2d883a59f087659d-0304/timelines/7bedf4a6995baff7c0421ff9aebbcdab/000000067F000080140000802100000D4000-000000067F000080140000802100000F0000__0000069F34773460-0000000b
```
2. Added a unit test to search the layer file contents. It's not
implemented part of `pagectl` because it depends on some test harness
code, which can only be used by unit tests.
Example usage:
```
cargo test --package pageserver --lib -- tenant::debug::test_search_key --exact --nocapture -- --tenant-id 09b99ea3239bbb3b2d883a59f087659d --timeline-id 7bedf4a6995baff7c0421ff9aebbcdab --data-dir /Users/chen.luo/Downloads/corruption --key 000000067F000080140000802100000D61BD --lsn 70C/BF3D61D8
```
Example output:
```
# omitted image for brievity
delta: 69F/769D8180: will_init: false, "OgAAALGkuwXwYp12nwYAAECGAAASIqLHAAAAAH8GAAAUgAAAIYAAAL1hDQD/DLGkuwUDAAAAEAAWAA=="
delta: 69F/769CB6D8: will_init: false, "PQAAALGkuwXotZx2nwYAABAJAAAFk7tpACAGAH8GAAAUgAAAIYAAAL1hDQD/CQUAEAASALExuwUBAAAAAA=="
```
## Problem VI
Currently when page service resolves shards from page numbers, it
doesn't fully support the case that the shard could be split in the
middle. This will lead to query failures during the tenant split for
either commit or abort cases (it's mostly for abort).
## Summary of changes VI
This PR adds retry logic in `Cache::get()` to deal with shard resolution
errors more gracefully. Specifically, it'll clear the cache and retry,
instead of failing the query immediately. It also reduces the internal
timeout to make retries faster.
The PR also fixes a very obvious bug in
`TenantManager::resolve_attached_shard` where the code tries to cache
the computed the shard number, but forgot to recompute when the shard
count is different.
---------
Co-authored-by: William Huang <william.huang@databricks.com>
Co-authored-by: Haoyu Huang <haoyu.huang@databricks.com>
Co-authored-by: Chen Luo <chen.luo@databricks.com>
Co-authored-by: Vlad Lazar <vlad.lazar@databricks.com>
Co-authored-by: Vlad Lazar <vlad@neon.tech>
## Problem
Grafana Alloy in cluster mode seems to send duplicate "seconds" scrape
URL parameters
when one of its instances is disrupted.
## Summary of changes
Temporarily accept duplicate parameters as long as their value is
identical.
See #11992 and #11961 for some examples of usecases.
This introduces a JSON serialization lib, designed for more flexibility
than serde_json offers.
## Dynamic construction
Sometimes you have dynamic values you want to serialize, that are not
already in a serde-aware model like a struct or a Vec etc. To achieve
this with serde, you need to implement a lot of different traits on a
lot of different new-types. Because of this, it's often easier to
give-in and pull all the data into a serde-aware model
(serde_json::Value or some intermediate struct), but that is often not
very efficient.
This crate allows full control over the JSON encoding without needing to
implement any extra traits. Just call the relevant functions, and it
will guarantee a correctly encoded JSON value.
## Async construction
Similar to the above, sometimes the values arrive asynchronously. Often
collecting those values in memory is more expensive than writing them as
JSON, since the overheads of `Vec` and `String` is much higher, however
there are exceptions.
Serializing to JSON all in one go is also more CPU intensive and can
cause lag spikes, whereas serializing values incrementally spreads out
the CPU load and reduces lag.
## Problem
Names are not consistent between safekeeper migration RFC and the actual
implementation.
It's not used anywhere in production yet, so it's safe to rename. We
don't need to worry about backward compatibility.
- Follow up on https://github.com/neondatabase/neon/pull/12432
## Summary of changes
- rename term -> last_log_term in TimelineMembershipSwitchResponse
- add missing fields to TimelineMembershipSwitchResponse in python
- Add ComputeSpec flag `offload_lfc_interval_seconds` controlling
whether LFC should be offloaded to endpoint storage. Default value
(None) means "don't offload".
- Add glue code around it for `neon_local` and integration tests.
- Add `autoprewarm` mode for `test_lfc_prewarm` testing
`offload_lfc_interval_seconds` and `autoprewarm` flags in conjunction.
- Rename `compute_ctl_lfc_prewarm_requests_total` and
`compute_ctl_lfc_offload_requests_total` to
`compute_ctl_lfc_prewarms_total`
and `compute_ctl_lfc_offloads_total` to reflect we count prewarms and
offloads, not `compute_ctl` requests of those.
Don't count request in metrics if there is a prewarm/offload already
ongoing.
https://github.com/neondatabase/cloud/issues/19011
Resolves: https://github.com/neondatabase/cloud/issues/30770
## Problem
The current deletion operation is synchronous and blocking, which is
unsuitable for potentially long-running tasks like. In such cases, the
standard HTTP request-response pattern is not a good fit.
## Summary of Changes
- Added new `storcon_cli` commands: `NodeStartDelete` and
`NodeCancelDelete` to initiate and cancel deletion asynchronously.
- Added corresponding `storcon` HTTP handlers to support the new
start/cancel deletion flow.
- Introduced a new type of background operation: `Delete`, to track and
manage the deletion process outside the request lifecycle.
---------
Co-authored-by: Aleksandr Sarantsev <aleksandr.sarantsev@databricks.com>
When deploying new safekeepers, we don't immediately want to send
traffic to them. Maybe they are not ready yet by the time the deploy
script is registering them with the storage controller.
For pageservers, the storcon solves the problem by not scheduling stuff
to them unless there has been a positive heartbeat response. We can't do
the same for safekeepers though, otherwise a single down safekeeper
would mean we can't create new timelines in smaller regions where there
is only three safekeepers in total.
So far we have created safekeepers as `pause` but this adds a manual
step to safekeeper deployment which is prone to oversight. We want
things to be automatted. So we introduce a new state `activating` that
acts just like `pause`, except that we automatically transition the
policy to `active` once we get a positive heartbeat from the safekeeper.
For `pause`, we always keep the safekeeper paused.
## TLDR
This PR is a no-op. The changes are disabled by default.
## Problem
I. Currently we don't have a way to detect disk I/O failures from WAL
operations.
II.
We observe that the offloader fails to upload a segment due to race
conditions on XLOG SWITCH and PG start streaming WALs. wal_backup task
continously failing to upload a full segment while the segment remains
partial on the disk.
The consequence is that commit_lsn for all SKs move forward but
backup_lsn stays the same. Then, all SKs run out of disk space.
III.
We have discovered SK bugs where the WAL offload owner cannot keep up
with WAL backup/upload to S3, which results in an unbounded accumulation
of WAL segment files on the Safekeeper's disk until the disk becomes
full. This is a somewhat dangerous operation that is hard to recover
from because the Safekeeper cannot write its control files when it is
out of disk space. There are actually 2 problems here:
1. A single problematic timeline can take over the entire disk for the
SK
2. Once out of disk, it's difficult to recover SK
IV.
Neon reports certain storage errors as "critical" errors using a marco,
which will increment a counter/metric that can be used to raise alerts.
However, this metric isn't sliced by tenant and/or timeline today. We
need the tenant/timeline dimension to better respond to incidents and
for blast radius analysis.
## Summary of changes
I.
The PR adds a `safekeeper_wal_disk_io_errors ` which is incremented when
SK fails to create or flush WALs.
II.
To mitigate this issue, we will re-elect a new offloader if the current
offloader is lagging behind too much.
Each SK makes the decision locally but they are aware of each other's
commit and backup lsns.
The new algorithm is
- determine_offloader will pick a SK. say SK-1.
- Each SK checks
-- if commit_lsn - back_lsn > threshold,
-- -- remove SK-1 from the candidate and call determine_offloader again.
SK-1 will step down and all SKs will elect the same leader again.
After the backup is caught up, the leader will become SK-1 again.
This also helps when SK-1 is slow to backup.
I'll set the reelect backup lag to 4 GB later. Setting to 128 MB in dev
to trigger the code more frequently.
III.
This change addresses problem no. 1 by having the Safekeeper perform a
timeline disk utilization check check when processing WAL proposal
messages from Postgres/compute. The Safekeeper now rejects the WAL
proposal message, effectively stops writing more WAL for the timeline to
disk, if the existing WAL files for the timeline on the SK disk exceeds
a certain size (the default threshold is 100GB). The disk utilization is
calculated based on a `last_removed_segno` variable tracked by the
background task removing WAL files, which produces an accurate and
conservative estimate (>= than actual disk usage) of the actual disk
usage.
IV.
* Add a new metric `hadron_critical_storage_event_count` that has the
`tenant_shard_id` and `timeline_id` as dimensions.
* Modified the `crtitical!` marco to include tenant_id and timeline_id
as additional arguments and adapted existing call sites to populate the
tenant shard and timeline ID fields. The `critical!` marco invocation
now increments the `hadron_critical_storage_event_count` with the extra
dimensions. (In SK there isn't the notion of a tenant-shard, so just the
tenant ID is recorded in lieu of tenant shard ID.)
I considered adding a separate marco to avoid merge conflicts, but I
think in this case (detecting critical errors) conflicts are probably
more desirable so that we can be aware whenever Neon adds another
`critical!` invocation in their code.
---------
Co-authored-by: Chen Luo <chen.luo@databricks.com>
Co-authored-by: Haoyu Huang <haoyu.huang@databricks.com>
Co-authored-by: William Huang <william.huang@databricks.com>
## Problem
Previously, the background worker that collects the list of installed
extensions across DBs had a timeout set to 1 hour. This cause a problem
with computes that had a `suspend_timeout` > 1 hour as this collection
was treated as activity, preventing compute shutdown.
Issue: https://github.com/neondatabase/cloud/issues/30147
## Summary of changes
Passing the `suspend_timeout` as part of the `ComputeSpec` so that any
updates to this are taken into account by the background worker and
updates its collection interval.
## Problem
The gRPC API does not provide LSN leases.
## Summary of changes
* Add LSN lease support to the gRPC API.
* Use gRPC LSN leases for static computes with `grpc://` connstrings.
* Move `PageserverProtocol` into the `compute_api::spec` module and
reuse it.
## Problem
Location config changes can currently result in changes to the shard
identity. Such changes will cause data corruption, as seen with #12217.
Resolves#12227.
Requires #12377.
## Summary of changes
Assert that the shard identity does not change on location config
updates and on (re)attach.
This is currently asserted with `critical!`, in case it misfires in
production. Later, we should reject such requests with an error and turn
this into a proper assertion.
## Problem
Similarly to #12217, the following endpoints may result in a stripe size
mismatch between the storage controller and Pageserver if an unsharded
tenant has a different stripe size set than the default. This can lead
to data corruption if the tenant is later manually split without
specifying an explicit stripe size, since the storage controller and
Pageserver will apply different defaults. This commonly happens with
tenants that were created before the default stripe size was changed
from 32k to 2k.
* `PUT /v1/tenant/config`
* `PATCH /v1/tenant/config`
These endpoints are no longer in regular production use (they were used
when cplane still managed Pageserver directly), but can still be called
manually or by tests.
## Summary of changes
Retain the current shard parameters when updating the location config in
`PUT | PATCH /v1/tenant/config`.
Also opportunistically derive `Copy` for `ShardParameters`.
## Problem
The problem has been well described in already-commited PR #11853.
tl;dr: BufferedWriter is sensitive to cancellation, which the previous
approach was not.
The write path was most affected (ingest & compaction), which was mostly
fixed in #11853:
it introduced `PutError` and mapped instances of `PutError` that were
due to cancellation of underlying buffered writer into
`CreateImageLayersError::Cancelled`.
However, there is a long tail of remaining errors that weren't caught by
#11853 that result in `CompactionError::Other`s, which we log with great
noise.
## Solution
The stack trace logging for CompactionError::Other added in #11853
allows us to chop away at that long tail using the following pattern:
- look at the stack trace
- from leaf up, identify the place where we incorrectly map from the
distinguished variant X indicating cancellation to an `anyhow::Error`
- follow that anyhow further up, ensuring it stays the same anyhow all
the way up in the `CompactionError::Other`
- since it stayed one anyhow chain all the way up, root_cause() will
yield us X
- so, in `log_compaction_error`, add an additional `downcast_ref` check
for X
This PR specifically adds checks for
- the flush task cancelling (FlushTaskError, BlobWriterError)
- opening of the layer writer (GateError)
That should cover all the reports in issues
- https://github.com/neondatabase/cloud/issues/29434
- https://github.com/neondatabase/neon/issues/12162
## Refs
- follow-up to #11853
- fixup of / fixes https://github.com/neondatabase/neon/issues/11762
- fixes https://github.com/neondatabase/neon/issues/12162
- refs https://github.com/neondatabase/cloud/issues/29434
## Problem
Some pageservers hit `max_size_entries` limit in staging with only ~25
MiB storage used by basebackup cache. The limit is too strict. It should
be safe to relax it.
- Part of https://github.com/neondatabase/cloud/issues/29353
## Summary of changes
- Increase the default `max_size_entries` from 1000 to 10000
## Problem
In our infra config, we have to split server_api_key and other fields in
two files: the former one in the sops file, and the latter one in the
normal config. It creates the situation that we might misconfigure some
regions that it only has part of the fields available, causing
storcon/pageserver refuse to start.
## Summary of changes
Allow PostHog config to have part of the fields available. Parse it
later.
Signed-off-by: Alex Chi Z <chi@neon.tech>
## Problem
Basebackup cache now uses unbounded channel for prepare requests. In
theory it can grow large if the cache is hung and does not process the
requests.
- Part of https://github.com/neondatabase/cloud/issues/29353
## Summary of changes
- Replace an unbounded channel with a bounded one, the size is
configurable.
- Add `pageserver_basebackup_cache_prepare_queue_size` to observe the
size of the queue.
- Refactor a bit to move all metrics logic to `basebackup_cache.rs`
