Vlad Lazar d03235bd68 pageserver: rate limit log for loads of layers visited (#8228) ...

## Problem
At high percentiles we see more than 800 layers being visited by the
read path. We need the tenant/timeline to investigate.

## Summary of changes
Add a rate limited log line when the average number of layers visited
per key is in the last specified histogram bucket.
I plan to use this to identify tenants in us-east-2 staging that exhibit
this behaviour. Will revert before next week's release.

2024-07-08 16:21:26 +01:00

..

bin

pageserver: refine tenant_id->shard lookup (#7762)

2024-05-16 08:26:34 +00:00

consumption_metrics

pageserver: improve readability of shard.rs (#7330)

2024-04-15 11:50:26 +01:00

context

feat(per-tenant throttling): exclude throttled time from page_service metrics + regression test (#6953)

2024-03-05 13:44:00 +00:00

deletion_queue

pageserver: local layer path followups (#7640)

2024-05-08 16:50:21 +00:00

http

pageserver: simpler, stricter config error handling (#8177)

2024-07-08 16:21:26 +01:00

tenant

pageserver: rate limit log for loads of layers visited (#8228)

2024-07-08 16:21:26 +01:00

virtual_file

virtual_file: take a Slice in the read APIs, eliminate read_exact_at_n, fix UB for engine std-fs (#8186)

2024-06-28 11:20:37 +02:00

walredo

test(pageserver): add test wal record for unit testing (#8015)

2024-06-13 09:44:37 -04:00

auth.rs

attachment_service: JWT auth enforcement (#6897)

2024-02-26 18:17:06 +00:00

aux_file.rs

feat(pageserver): compute aux file size on initial logical size calculation (#7958)

2024-06-04 13:47:48 -04:00

basebackup.rs

fix(pageserver): include aux file in basebackup only once (#8207)

2024-07-08 16:21:26 +01:00

config.rs

remote_storage config: move handling of empty inline table {} to callers (#8193)

2024-07-08 16:21:26 +01:00

consumption_metrics.rs

pageserver: improved synthetic size & find_gc_cutoff error handling (#8051)

2024-06-14 11:08:11 +01:00

context.rs

feat(per-tenant throttling): exclude throttled time from page_service metrics + regression test (#6953)

2024-03-05 13:44:00 +00:00

control_plane_client.rs

refactor: move NodeMetadata to pageserver_api; use it from neon_local (#7606)

2024-05-03 13:15:38 -04:00

deletion_queue.rs

pageserver: remove legacy tenant config code, clean up redundant generation none/broken usages (#7947)

2024-06-26 19:53:59 +00:00

disk_usage_eviction_task.rs

refactor(rtc): remove the duplicate IndexLayerMetadata (#7860)

2024-05-23 23:24:31 +03:00

import_datadir.rs

pageserver: apply shard filtering to blocks ingested during initdb (#7319)

2024-04-05 18:07:35 +01:00

lib.rs

pageserver: remove legacy tenant config code, clean up redundant generation none/broken usages (#7947)

2024-06-26 19:53:59 +00:00

metrics.rs

pageserver: remove tenant create API (#8135)

2024-06-28 09:14:19 +01:00

page_cache.rs

remove materialized page cache (#8105)

2024-06-20 11:56:14 +02:00

page_service.rs

Add counters for commands processed through the libpq page service API (#8089)

2024-06-26 19:53:03 +03:00

pgdatadir_mapping.rs

test(pageserver): add test keyspace into collect_keyspace (#8016)

2024-06-12 17:42:43 +00:00

repository.rs

feat(pageserver): use leases to temporarily block gc (#8084)

2024-06-18 17:37:06 +00:00

span.rs

debug_assert presence of shard_id tracing field (#6572)

2024-02-06 14:43:33 +00:00

statvfs.rs

Also allow unnecessary_fallible_conversions lint (#6294)

2024-01-09 04:22:36 +00:00

task_mgr.rs

feat: Timeline detach ancestor (#7456)

2024-05-07 13:47:57 +03:00

tenant.rs

pageserver: simpler, stricter config error handling (#8177)

2024-07-08 16:21:26 +01:00

trace.rs

fix: replace all std::PathBufs with camino::Utf8PathBuf (#5352)

2023-10-04 17:52:23 +03:00

utilization.rs

implement Serialize/Deserialize for SystemTime with RFC3339 format (#7203)

2024-04-08 15:53:07 +01:00

virtual_file.rs

virtual_file: take a Slice in the read APIs, eliminate read_exact_at_n, fix UB for engine std-fs (#8186)

2024-06-28 11:20:37 +02:00

walingest.rs

Restore running xacts from CLOG on replica startup (#7288)

2024-07-04 18:58:34 +03:00

walrecord.rs

test(pageserver): add test wal record for unit testing (#8015)

2024-06-13 09:44:37 -04:00

walredo.rs

fix: shutdown does not kill walredo processes (#8150)

2024-06-27 15:58:28 +02:00