From d562d772a877f8920081cff8d5ba0b098f458a55 Mon Sep 17 00:00:00 2001 From: Roman Khavronenko Date: Mon, 15 Jan 2024 10:03:06 +0100 Subject: [PATCH] lib/storage: properly check for `storage/prefetchedMetricIDs` cache expiration deadline (#5607) Before, this cache was limited only by size. Cache invalidation by time happens with jitter to prevent thundering herd problem. Signed-off-by: hagen1778 --- docs/CHANGELOG.md | 1 + lib/storage/storage.go | 8 ++++++-- 2 files changed, 7 insertions(+), 2 deletions(-) diff --git a/docs/CHANGELOG.md b/docs/CHANGELOG.md index 2710228bd..77a465eea 100644 --- a/docs/CHANGELOG.md +++ b/docs/CHANGELOG.md @@ -47,6 +47,7 @@ The sandbox cluster installation is running under the constant load generated by * BUGFIX: [VictoriaMetrics cluster](https://docs.victoriametrics.com/Cluster-VictoriaMetrics.html): properly return full results when `-search.skipSlowReplicas` command-line flag is passed to `vmselect` and when [vmstorage groups](https://docs.victoriametrics.com/Cluster-VictoriaMetrics.html#vmstorage-groups-at-vmselect) are in use. Previously partial results could be returned in this case. * BUGFIX: `vminsert`: properly accept samples via [OpenTelemetry data ingestion protocol](https://docs.victoriametrics.com/#sending-data-via-opentelemetry) when these samples have no [resource attributes](https://opentelemetry.io/docs/instrumentation/go/resources/). Previously such samples were silently skipped. * BUGFIX: `vmstorage`: added missing `-inmemoryDataFlushInterval` command-line flag, which was missing in [VictoriaMetrics cluster](https://docs.victoriametrics.com/Cluster-VictoriaMetrics.html) after implementing [this feature](https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3337) in [v1.85.0](https://github.com/VictoriaMetrics/VictoriaMetrics/releases/tag/v1.85.0). +* BUGFIX: `vmstorage`: properly check for `storage/prefetchedMetricIDs` cache expiration deadline. Before, this cache was limited only by size. * BUGFIX: [vmalert](https://docs.victoriametrics.com/vmalert.html): check `-external.url` schema when starting vmalert, must be `http` or `https`. Before, alertmanager could reject alert notifications if `-external.url` contained no or wrong schema. * BUGFIX: [vmalert](https://docs.victoriametrics.com/vmalert.html): automatically add `exported_` prefix for original evaluation result label if it's conflicted with external or reserved one, previously it was overridden. See [this issue](https://github.com/VictoriaMetrics/VictoriaMetrics/issues/5161). * BUGFIX: [MetricsQL](https://docs.victoriametrics.com/MetricsQL.html): properly handle queries, which wrap [rollup functions](https://docs.victoriametrics.com/MetricsQL.html#rollup-functions) with multiple arguments without explicitly specified lookbehind window in square brackets into [aggregate functions](https://docs.victoriametrics.com/MetricsQL.html#aggregate-functions). For example, `sum(quantile_over_time(0.5, process_resident_memory_bytes))` was resulting to `expecting at least 2 args to ...; got 1 args` error. Thanks to @atykhyy for [the pull request](https://github.com/VictoriaMetrics/VictoriaMetrics/pull/5414). diff --git a/lib/storage/storage.go b/lib/storage/storage.go index fef79e6cc..96b6e2459 100644 --- a/lib/storage/storage.go +++ b/lib/storage/storage.go @@ -30,6 +30,7 @@ import ( "github.com/VictoriaMetrics/VictoriaMetrics/lib/workingsetcache" "github.com/VictoriaMetrics/fastcache" "github.com/VictoriaMetrics/metricsql" + "github.com/valyala/fastrand" ) const ( @@ -1268,10 +1269,13 @@ func (s *Storage) prefetchMetricNames(qt *querytracer.Tracer, accountID, project // Store the pre-fetched metricIDs, so they aren't pre-fetched next time. s.prefetchedMetricIDsLock.Lock() var prefetchedMetricIDsNew *uint64set.Set - if fasttime.UnixTimestamp() < atomic.LoadUint64(&s.prefetchedMetricIDsDeadline) { + if fasttime.UnixTimestamp() > atomic.LoadUint64(&s.prefetchedMetricIDsDeadline) { // Periodically reset the prefetchedMetricIDs in order to limit its size. prefetchedMetricIDsNew = &uint64set.Set{} - atomic.StoreUint64(&s.prefetchedMetricIDsDeadline, fasttime.UnixTimestamp()+73*60) + deadlineSec := 73 * 60 + jitterSec := fastrand.Uint32n(uint32(deadlineSec / 10)) + metricIDsDeadline := fasttime.UnixTimestamp() + uint64(deadlineSec) + uint64(jitterSec) + atomic.StoreUint64(&s.prefetchedMetricIDsDeadline, metricIDsDeadline) } else { prefetchedMetricIDsNew = prefetchedMetricIDs.Clone() }