Aliaksandr Valialkin 2024-09-22 21:56:27 +02:00
parent 8ed9978591
commit 0ada781cf2
No known key found for this signature in database
GPG key ID: 52C003EE2BCDB9EB

View file

@ -34,6 +34,7 @@ See also [LTS releases](https://docs.victoriametrics.com/lts-releases/).
* BUGFIX: [vmagent](https://docs.victoriametrics.com/vmagent/): properly consume messages [from kafka](https://docs.victoriametrics.com/vmagent/#kafka-integration). Previously vmagent could skip some messages during start-up.
* BUGFIX: [vmagent dashboard](https://github.com/VictoriaMetrics/VictoriaMetrics/blob/master/dashboards/vmagent.json): fix legend captions for stream aggregation related panels. Before they were displaying wrong label names.
* BUGFIX: [vmgateway](https://docs.victoriametrics.com/vmgateway/): add missing `datadog`, `newrelic`, `opentelemetry` and `pushgateway` routes to the `JWT` authorization routes. Allows prefixed (`promtheus/graphite`) routes for query requests.
* BUGFIX: [Single-node VictoriaMetrics](https://docs.victoriametrics.com/) and `vmstorage` in [VictoriaMetrics cluster](https://docs.victoriametrics.com/cluster-victoriametrics/): properly cache empty list of matching time series for the given [labels filter](https://docs.victoriametrics.com/keyconcepts/#filtering). This type of caching was broken since [v1.97.0](https://github.com/VictoriaMetrics/VictoriaMetrics/releases/tag/v1.97.0), which could result in the increased CPU usage when performing queries, which match zero time series. See [this issue](https://github.com/VictoriaMetrics/VictoriaMetrics/issues/7009).
* BUGFIX: [Single-node VictoriaMetrics](https://docs.victoriametrics.com/) and `vmstorage` in [VictoriaMetrics cluster](https://docs.victoriametrics.com/cluster-victoriametrics/): Fixes start-up crash on Windows OS. See this [issue](https://github.com/VictoriaMetrics/VictoriaMetrics/issues/6973) for details.
* BUGFIX: [Single-node VictoriaMetrics](https://docs.victoriametrics.com/) and `vmstorage` in [VictoriaMetrics cluster](https://docs.victoriametrics.com/cluster-victoriametrics/): fix metric `vm_object_references{type="indexdb"}`. Previously, it was overcounted.
* BUGFIX: [Single-node VictoriaMetrics](https://docs.victoriametrics.com/) and `vmstorage` in [VictoriaMetrics cluster](https://docs.victoriametrics.com/cluster-victoriametrics/): properly ingest stale NaN samples. Previously it could be dropped if series didn't exist at storage node. See this [issue](https://github.com/VictoriaMetrics/VictoriaMetrics/issues/5069) for details.