VictoriaMetrics/docs/CHANGELOG.md
2022-07-21 20:48:05 +03:00

219 KiB

sort
16

CHANGELOG

The following tip changes can be tested by building VictoriaMetrics components from the latest commits according to the following docs:

tip

  • FEATURE: vmalert: allow configuring additional headers for datasource.url, remoteWrite.url and remoteRead.url URLs. Headers also can be set on group level via headers param. See this issue for details.

  • FEATURE: MetricsQL: execute left and right sides of certain operations in parallel. For example, q1 or q2, aggr_func(q1) <op> q2, q1 <op> aggr_func(q1). This may improve query performance if VictoriaMetrics has enough free resources for parallel processing of both sides of the operation. See this feature request.

  • FEATURE: vmauth: allow duplicate username records with different passwords at configuration file. It should allow password rotation without username change.

  • FEATURE: add ability to push internal metrics (e.g. metrics exposed at /metrics page) to the configured remote storage from all the VictoriaMetrics components. See these docs.

  • BUGFIX: vmagent: restart all the scrape jobs during config reload after global section is changed inside -promscrape.config. See this issue.

  • BUGFIX: vmagent: properly assume role with AWS ECS credentials. See this issue. Thanks to @transacid for the fix.

  • BUGFIX: MetricsQL: return series from q1 if q2 doesn't return matching time series in the query q1 ifnot q2. Previously series from q1 weren't returned in this case.

v1.79.0

Released at 14-07-2022

Update note 1: this release introduces backwards-incompatible changes to vm_partial_results_total metric by changing its labels to be consistent with vm_requests_total metric. If you use alerting rules or Grafana dashboards, which rely on this metric, then they must be updated. The official dashboards for VictoriaMetrics don't use this metric.

Update note 2: vmalert adds /vmalert/ prefix to web urls according to this issue. This may affect vmalert instances with non-empty -http.pathPrefix command-line flag. After the update, configuring this flag is no longer needed. Here's why.

Update note 3: this release introduces backwards-incompatible changes to communication protocol between vmselect and vmstorage nodes in cluster version of VictoriaMetrics because of added ability to query vmselect data from other vmselect nodes - see these docs, so read requests to vmselect will fail until the upgrade is complete. These errors will stop after all the vmselect and vmstorage nodes are updated to the new release. It is safe to downgrade to previous releases at any time.

Update note 4: this release removes support of deprecated in 1.70.0 param extra_filter_labels from vmalert's groups definition. This deprecated param was replaced with params.

Update note 5: this release changes naming for published linux binaries at releases. Now names for binaries for all the supported platforms match the following template - $(APP_NAME)-$(GOOS)-$(GOARCH)-$(VERSION).tar.gz. For example, victoria-metrics-linux-amd64-v1.79.0.tar.gz. Previously linux binaries didn't have $(GOOS) part, e.g. they had the name victoria-metrics-amd64-v1.79.0.tar.gz. Please update automation scripts for upgrading VictoriaMetrics releases according to this change.

  • FEATURE: vmagent: add azure_sd_configs service discovery mechanism. It allows discovering Virtual Machines at Azure Cloud. See this issue.
  • FEATURE: vmalert: deprecate alert's status link /api/v1/<groupID>/<alertID>/status in favour of api/v1/alert?group_id=<group_id>&alert_id=<alert_id>". The old alert's status link is still supported, but will be removed in future releases. See this issue.
  • FEATURE: cluster version of VictoriaMetrics: add support for querying lower-level vmselect nodes from upper-level vmselect nodes. This makes possible to build multi-level cluster setups for global querying view and HA purposes without the need to use Promxy. See these docs and this issue.
  • FEATURE: add -search.setLookbackToStep command-line flag, which enables InfluxDB-like gap filling during querying. See these docs for details.
  • FEATURE: vmui: add an UI for query tracing. It can be enabled by clicking trace query checkbox and re-running the query. See this feature request.
  • FEATURE: vmagent: add -remoteWrite.headers command-line option for specifying optional HTTP headers to send to the configured -remoteWrite.url. For example, -remoteWrite.headers='Foo:Bar^^Baz:x' would send Foo: Bar and Baz: x HTTP headers with every request to -remoteWrite.url. See this feature request.
  • FEATURE: vmagent: push per-target scrape_samples_limit metric to the cofigured -remoteWrite.url if sample_limit option is set for this target in scrape_config. See this feature request.
  • FEATURE: vmagent: attach node-level labels to kubernetes_sd_config targets if attach_metadata: {"node": true} is set for role: endpoints and role: endpointslice. This is a feature backport from Prometheus 2.37 - see this pull request.
  • FEATURE: vmagent: add ability to specify additional HTTP headers to send to scrape targets via headers section in scrape_configs. This can be used when the scrape target requires custom authorization and authentication like in this stackoverflow question. For example, the following config instructs sending My-Auth: top-secret and TenantID: FooBar headers with each request to http://host123:8080/metrics:
scrape_configs:
- job_name: foo
  headers:
  - "My-Auth: top-secret"
  - "TenantID: FooBar"
  static_configs:
  - targets: ["host123:8080"]
  • FEATURE: add ability to pass limit query arg to api/v1/series endpoint. This can be used if only a sample of up to limit series must be returned from the endpoint. See this feature request and these docs.

  • FEATURE: query tracing: show timestamps in query traces in human-readable format (aka RFC3339 in UTC timezone) instead of milliseconds since Unix epoch. For example, 2022-06-27T10:32:54.506Z instead of 1656325974506. This improves traces' readability.

  • FEATURE: improve performance of /api/v1/series requests, which return big number of time series.

  • FEATURE: VictoriaMetrics cluster: improve query performance when replication is enabled.

  • FEATURE: MetricsQL: properly handle partial counter resets in remove_resets function. Now remove_resets(sum(m)) should returns the expected increasing line when some time series matching m disappear on the selected time range. Previously such a query would return horizontal line after the disappeared series.

  • FEATURE: expose vm_next_retention_seconds metric at http://victoriametrics:8428/metrics, which shows the number of seconds left until the next indexdb rotation. Thanks to @guidao for the pull request.

  • FEATURE: expose additional histogram metrics at http://victoriametrics:8428/metrics, which may help understanding query workload:

    • vm_rows_read_per_query - the number of raw samples read per query.
    • vm_rows_scanned_per_query - the number of raw samples scanned per query. This number can exceed vm_rows_read_per_query if step query arg passed to /api/v1/query_range is smaller than the lookbehind window set in square brackets of rollup function. For example, if increase(some_metric[1h]) is executed with the step=5m, then the same raw samples on a hour time range are scanned 1h/5m=12 times. See this article for details.
    • vm_rows_read_per_series - the number of raw samples read per queried series.
    • vm_series_read_per_query - the number of series read per query.
  • FEATURE: publish binaries for FreeBSD and OpenBSD at releases page.

  • FEATURE: vmui: allow selecting the needed columns at table view. This functionaly may help when the selected time series contain many different labels. See this feature request and this pull request.

  • BUGFIX: consistently name binaries at releases page in the form $(APP_NAME)-$(GOOS)-$(GOARCH)-$(VERSION).tar.gz. For example, victoria-metrics-linux-amd64-v1.79.0.tar.gz. Previously the $(GOOS) part was missing in binaries for Linux.

  • BUGFIX: vmalert: allow using __name__ label (aka metric name) in alerting annotations. For example:

{% raw %}

{{ $labels.__name__ }}: Too high connection number for "{{ $labels.instance }}

{% endraw %}

  • BUGFIX: limit max memory occupied by the cache, which stores parsed regular expressions. Previously too long regular expressions passed in MetricsQL queries could result in big amounts of used memory (e.g. multiple of gigabytes). Now the max cache size for parsed regexps is limited to a a few megabytes.
  • BUGFIX: MetricsQL: properly handle partial counter resets when calculating rate, irate and increase functions. Previously these functions could return zero values after partial counter resets until the counter increases to the last value before partial counter reset. See this issue.
  • BUGFIX: MetricsQL: properly calculate histogram_quantile over Prometheus buckets with unexpected values. See this issue.
  • BUGFIX: MetricsQL: properly evaluate timezone_offset function over time range covering time zone offset switches. See this issue.
  • BUGFIX: vmagent: properly add service-level labels (__meta_kubernetes_service_*) to discovered targets for role: endpointslice in kubernetes_sd_config. Previously these labels were missing. See this issue.
  • BUGFIX: vmagent: make sure that stale markers are generated with the actual timestamp when unsuccessful scrape occurs. This should prevent from possible time series overlap on scrape target restart in dynmaic envirnoments such as Kubernetes.
  • BUGFIX: vmagent: properly reload changed -promscrape.config file when -promscrape.configCheckInterval option is set. The changed config file wasn't reloaded in this case since v1.69.0. See this pull request. Thanks to @ttyv for the fix.
  • BUGFIX: vmagent: properly set Host header during target scraping when proxy_url is set to http proxy. Previously the Host header was set to the proxy hostname instead of the target hostname. See this issue.
  • BUGFIX: VictoriaMetrics cluster: assume that the response is complete if -search.denyPartialResponse is enabled and up to -replicationFactor - 1 vmstorage nodes are unavailable. See this issue.
  • BUGFIX: vmselect: update vm_partial_results_total metric labels to be consistent with vm_requests_total labels.
  • BUGFIX: accept tags without values when reading data in DataDog format. Thanks to @PerGon for the pull request.
  • BUGFIX: vmui: properly pass the end of the selected time range to time query arg to /api/v1/query when displaying the requested data in JSON and table views. Previously the time query arg wasn't set, so /api/v1/query was always returning query results for the current time regardless of the selected time range. See this issue.
  • BUGFIX: vmui: allow clicking on the suggestion from autocomplete list. See this issue.
  • BUGFIX: vmui: apply the selected time range in date picker only after clicking the Apply button. See this issue.

v1.78.1

Released at 08-07-2022

Update notes: it is recommended clearing caches after the upgrade from v1.78.0 in order to immediately fix the issue for newly ingested data. Otherwise the issue may exist for newly ingested data for up to a day after the upgrade.

  • BUGFIX: properly register time series in per-day inverted index. Previously some series could miss registration in the per-day inverted index. This could result in missing time series during querying. The issue has been introduced in v1.78.0. See this and this issues.

v1.78.0

Released at 20-06-2022

Warning (03-07-2022): VictoriaMetrics v1.78.0 contains a bug, which may result in missing time series during queries. It is recommended upgrading to v1.78.1, which fixes the bug.

Update notes: this release introduces backwards-incompatible changes to communication protocol between vmselect and vmstorage nodes in cluster version of VictoriaMetrics because of added query tracing, so read requests to vmselect will fail until the upgrade is complete. These errors will stop after all the vmselect and vmstorage nodes are updated to the new release. It is safe to downgrade to previous releases.

  • SECURITY: add -flagsAuthKey command-line flag for protecting /flags endpoint from unauthorized access. Though this endpoint already hides values for command-line flags with key and password substrings in their names, other sensitive information could be exposed there. See This issue.

  • FEATURE: support query tracing, which allows determining bottlenecks during query processing. See these docs and this feature request.

  • FEATURE: vmui: add cardinality tab, which can help identifying the source of high cardinality and high churn rate issues. See this and this feature requests and these docs.

  • FEATURE: vmui: small UX enhancements according to this feature request.

  • FEATURE: allow overriding default limits for in-memory cache indexdb/tagFilters via flag -storage.cacheSizeIndexDBTagFilters. See this issue.

  • FEATURE: add support of lowercase and uppercase relabeling actions in the same way as Prometheus 2.36.0 does. See this issue.

  • FEATURE: add ability to change the indexdb rotation timezone offset via -retentionTimezoneOffset command-line flag. Previously it was performed at 4am UTC time. This could lead to performance degradation in the middle of the day when VictoriaMetrics runs in time zones located too far from UTC. Thanks to @cnych for the pull request.

  • FEATURE: limit the number of background merge threads on systems with big number of CPU cores by default. This increases the max size of parts, which can be created during background merge when -storageDataPath directory has limited free disk space. This may improve on-disk data compression efficiency and query performance. The limits can be tuned if needed with -smallMergeConcurrency and -bigMergeConcurrency command-line flags. See this pull request.

  • FEATURE: accept optional limit query arg at /api/v1/labels and /api/v1/label/.../values for limiting the numbef of sample entries returned from these endpoints. See these docs.

  • FEATURE: optimize performance for /api/v1/labels and /api/v1/label/.../values endpoints when match[], extra_label or extra_filters[] query args are passed to these endpoints. This should help with this issue.

  • FEATURE: vmalert: support limit param per-group for limiting number of produced samples per each rule. Thanks to @Howie59 for implementation.

  • FEATURE: vmalert: remove dependency on Internet access at web API pages. Previously the functionality and the layout of these pages was broken without Internet access. See shis issue.

  • FEATURE: vmalert: send alerts to the configured notifiers in parallel. Previously alerts were sent to notifiers sequentially. This could delay sending pending alerts when notifier blocks on the currently sent alert.

  • FEATURE: vmagent: implement the http://vmagent:8429/service-discovery page in the same way as Prometheus does. This page shows the original labels for all the discovered targets alongside the resulting labels after the relabeling. This simplifies service discovery debugging.

  • FEATURE: vmagent: remove dependency on Internet access at http://vmagent:8429/targets page. Previously the page layout was broken without Internet access. See shis issue.

  • FEATURE: vmagent: add support for kubeconfig_file option at kubernetes_sd_configs. It may be useful for Kubernetes monitoring by vmagent outside Kubernetes cluster. See this issue.

  • FEATURE: vmagent: expose /api/v1/status/config endpoint in the same way as Prometheus does. See these docs.

  • FEATURE: vmagent: add -promscrape.suppressScrapeErrorsDelay command-line flag, which can be used for delaying and aggregating the logging of per-target scrape errors. This may reduce the amounts of logs when vmagent scrapes many unreliable targets. See this feature request. Thanks to @jelmd for the initial implementation.

  • FEATURE: vmagent: add -promscrape.cluster.name command-line flag, which allows proper data de-duplication when the same target is scraped from multiple vmagent clusters. See this issue.

  • FEATURE: vmagent: add action: graphite relabeling rules optimized for extracting labels from Graphite-style metric names. See these docs and this feature request.

  • FEATURE: VictoriaMetrics enterprise: expose vm_downsampling_partitions_scheduled and vm_downsampling_partitions_scheduled_size_bytes metrics, which can be used for tracking the progress of initial downsampling for historical data. See this feature request.

  • FEATURE: VictoriaMetrics cluster: do not spend up to 5 seconds when trying to connect to unavailable vmstorage nodes. This should improve query latency when some of vmstorage nodes aren't available. Expose vm_tcpdialer_addr_available{addr="..."} metric at http://vmselect:8481/metrics for determining whether the given addr is available for establishing new connections. See this comment.

  • FEATURE: VictoriaMetrics cluster: add -vmstorageDialTimeout command-line flags to vmselect and vminsert for tuning the maximum duration for connection estabilishing to vmstorage nodes. This should help resolving this issue.

  • BUGFIX: support for data ingestion in DataDog format from legacy clients / agents. See this pull request. Thanks to @elProxy for the fix.

  • BUGFIX: vmagent: do not expose vm_promscrape_service_discovery_duration_seconds_bucket metric for unused service discovery types. This reduces the number of metrics exported at http://vmagent:8429/metrics. See this issue.

  • BUGFIX: vmalert: properly apply alert_relabel_configs relabeling rules to -notifier.config according to these docs. Thanks to @spectvtor for the bugfix.

  • BUGFIX: vmalert: properly add Content-Encoding: snappy, Content-Type: application/x-protobuf and X-Prometheus-Remote-Write-Version: 0.1.0 request headers when vmalert sends evaluated recording rules' data to -remoteWrite.url. These headers are needed by some remote storage systems in order to properly decode snappy-encoded request body. See this and this pull requests. Thanks to @manji-0 for th fix.

  • BUGFIX: deny background merge when the storage enters read-only mode, e.g. when free disk space becomes lower than -storage.minFreeDiskSpaceBytes. Background merge needs additional disk space, so it could result in no space left on device errors. See this issue.

  • BUGFIX: vmui: properly apply the selected time range when auto-refresh is enabled. See this issue.

  • BUGFIX: vmui: properly update the url with vmui state when new query is entered. See this issue.

  • BUGFIX: Graphite render API: properly calculate sample timestamps when moving*() functions such as movingAverage() are applied over summarize().

  • BUGFIX: limit the end query arg value to +2 days in the future at /api/v1/* endpoints, because VictoriaMetrics doesn't allow storing samples with timestamps bigger than +2 days in the future. This should help resolving this issue.

  • BUGFIX: properly register time series in per-day inverted index during the first hour after indexdb rotation. Previously this could lead to missing time series during querying if these time series stopped receiving new samples during the first hour after indexdb rotation. See this issue.

  • BUGFIX: do not register new series when -storage.maxHourlySeries or -storage.maxDailySeries limits were reached. Previously samples for new series weren't added to the database when the cardinality limit was reached, but series were still registered in the inverted index (aka indexdb). This could lead to unbound indexdb growth during high churn rate.

v1.77.2

Released at 21-05-2022

  • FEATURE: vmalert: support reusable templates for rules annotations. The path to the template files can be specified via -rule.templates flag. See more about this feature here. Thanks to @AndrewChubatiuk for the pull request. See this feature request.

  • FEATURE: vmalert: expose vmalert_iteration_interval_seconds metric at http://vmalert:8880/metrics. This metric shows the configured per-group evaluation interval. See this feature request.

  • FEATURE: vmctl: add influx-prometheus-mode command-line flag, which allows to restore the original time series written from Prometheus into InfluxDB during data migration from InfluxDB to VictoriaMetrics. See this feature request. Thanks to @mback2k for the pull request.

  • FEATURE: vmagent: add ability to specify AWS service name when issuing requests to AWS api. See this feature request. Thanks to @transacid for the pull request.

  • BUGFIX: vmagent: fix a bug, which could lead to incomplete discovery of scrape targets in Kubernetes (aka kubernetes_sd_config). the bug has been introduced in v1.77.0.

  • BUGFIX: vmalert: support scalar result type in response. See this issue.

  • BUGFIX: vmalert: support strings in humanize.* template function in the same way as Prometheus does. See this issue.

  • BUGFIX: vmalert: proxy /rules requests to vmalert from Grafana's alerting UI. This removes errors in Grafana's UI for Grafana versions older than 8.5.*. See this issue

  • BUGFIX: vmalert: do not add /api/v1/query suffix to -datasource.url if -remoteRead.disablePathAppend command-line flag is set. Previously this flag was applied only to -remoteRead.url, which could confuse users.

  • BUGFIX: vmalert: prevent from possible resource leak on config update, which could lead to the slowdown of vmalert over time. See this pull request.

  • BUGFIX: MetricsQL: do not return values from label_value() function if the original time series has no values at the selected timestamps.

  • BUGFIX: VictoriaMetrics cluster: limit the number of concurrently established connections from vmselect to vmstorage. This should prevent from potentially high spikes in the number of established connections after temporary slowdown in connection handshake procedure between vmselect and vmstorage because of spikes in workload. See this issue.

  • BUGFIX: vmctl: fix build for Solaris / SmartOS. See this issue.

v1.77.1

Released at 07-05-2022

  • FEATURE: vmagent: add ability to specify filters for Availability Zones in ec2_sd_config via az_filters section. This section can contain AZ-specific set of filters in the same way as the existing filters section, which is used for filtering EC2 instances. The list of supported AZ-specific filters is available here.

  • FEATURE: vmagent: expose vmagent_remotewrite_global_rows_pushed_before_relabel_total and vmagent_remotewrite_rows_pushed_after_relabel_total metrics at http://vmagent:8429/metrics, which can be used for monitoring the rate of rows (aka samples) pushed to remote storage before and after the relabeling via -remoteWrite.relabelConfig and -remoteWrite.urlRelabelConfig. See relabeling docs for details.

  • FEATURE: vmctl: add ability to skip db label during InfluxDB data import when influx-skip-database-label option is used. See this pull request. Thanks to @mback2k .

  • BUGFIX: vmagent: properly process passwords and secrets specified in the file pointed by -promscrape.config command-line flag. All the passwords and secrets were mistakenly replaced with <secret> string in v1.77.0. See this and this issue.

  • BUGFIX: vmagent: rename vmagent_remote_write_rate_limit_reached_total metric to vmagent_remotewrite_rate_limit_reached_total, so its name is consistent with the rest of vmagent_remotewrite_ metrics.

  • BUGFIX: vmagent: rename promscrape_stale_samples_created_total metric to vm_promscrape_stale_samples_created_total, so its name is consistent with the rest of vm_promscrape_ metrics.

  • BUGFIX: vmctl: properly import InfluxDB measurements if they contain db tag. Previously this could result in incomplete import of measurmenet tags. See this pull request. Thanks to @mback2k for the bugfix.

  • BUGFIX: vmui: do not reset the selected relative time range when entering new query. See this issue.

  • BUGFIX: vmbackup: disallow writing backups to -storageDataPath directory, since this directory is managed solely by VictoriaMetrics or vmstorage. Other apps shouldn't write into this directory. See this issue.

  • BUGFIX: do not allow setting -retentionPeriod smaller than one day, since VictoriaMetrics doesn't support properly such small retention periods. See this issue.

  • BUGFIX: VictoriaMetrics cluster: do not drop samples routed to readonly vmstorage nodes if -dropSamplesOnOverload command-line flag is set. Try re-routing them to healthy vmstorage nodes instead. See this issue.

v1.77.0

Released at 05-05-2022

  • FEATURE: vmagent: add support for sending data to remote storage with AWS sigv4 authorization. See this feature request.

  • FEATURE: vmagent: allow filtering targets by target url and by target labels with time series selector on http://vmagent:8429/targets page. This may be useful when vmagent scrapes big number of targets. See this feature request.

  • FEATURE: vmagent: reduce -promscrape.config reload duration when the config contains big number of jobs (aka scrape_config sections) and only a few of them are changed. Previously all the jobs were restarted. Now only the jobs with changed configs are restarted. This should reduce the probability of data miss because of slow config reload. See this issue.

  • FEATURE: vmagent: improve service discovery speed for big number of scrape targets. This should help when vmagent discovers big number of targets (e.g. thousands) in Kubernetes cluster. The service discovery speed now should scale with the number of CPU cores available to vmagent.

  • FEATURE: vmagent: add ability to attach node-level labels and annotations to discovered Kubernetes pod targets in the same way as Prometheus 2.35 does. See this feature request and this pull request.

  • FEATURE: vmagent: add support for tls_config and proxy_url options at oauth2 section in the same way as Prometheus does. See oauth2 docs.

  • FEATURE: vmagent: add support for min_version option at tls_config section in the same way as Prometheus does. See tls_config docs.

  • FEATURE: vmagent: expose vmagent_remotewrite_rate_limit metric at http://vmagent:8429/metrics, which can be used for alerting rules such as rate(vmagent_remotewrite_conn_bytes_written_total) / vmagent_remotewrite_rate_limit > 0.8 when -remoteWrite.rateLimit command-line flag is set. See this pull request.

  • FEATURE: vmalert: add support for DNS-based discovery for notifiers in the same way as Prometheus does (aka dns_sd_configs). See these docs and this feature request.

  • FEATURE: vmalert: add -replay.disableProgressBar command-line flag, which allows disabling progressbar in rules' backfilling mode. See this issue.

  • FEATURE: allow specifying TLS cipher suites for incoming https requests via -tlsCipherSuites command-line flag. See this feature request.

  • FEATURE: allow specifying TLS cipher suites for mTLS connections between cluster components via -cluster.tlsCipherSuites command-line flag. See these docs.

  • FEATURE: vmstorage: add -snapshotsMaxAge command-line flag for automatic removal of snapshots older than the given age.

  • FEATURE: vmui: show an empty graph on the selected time range when there is no data on it. Previously No data to show placeholder was shown instead of the graph in this case. This prevented from zooming and scrolling of such a graph.

  • FEATURE: vmui: show the selected last N minutes/hours/days in the top right corner. Previously the start - end duration was shown instead, which could be hard to interpret. See this feature request.

  • FEATURE: vmui: execute the query when enter button is pressed in the same way as Prometheus does. Multi-line query can be entered by pressing shift-enter in the query input field.

  • FEATURE: expose vm_indexdb_items_added_total and vm_indexdb_items_added_size_bytes_total counters at /metrics page, which can be used for monitoring the rate for addition of new entries in indexdb (aka inverted index) alongside the total size in bytes for the added entries. See this feature request.

  • FEATURE: vmctl: show data pocessing speed during data migration.

  • FEATURE: MetricsQL: add drop_common_labels() function, which drops common label="name" pairs from the passed time series. See these docs.

  • FEATURE: MetricsQL: add tlast_change_over_time(m[d]) function, which returns the timestamp of the last change of m on the given lookbehind window d. See these docs.

  • FEATURE: leave the last raw sample per each -dedup.minScrapeInterval discrete interval when the deduplication is enabled. This aligns better with the staleness rules in Prometheus comparing to the previous behaviour when the first sample per each -dedup.minScrapeInterval was left.

  • FEATURE: VictoriaMetrics cluster: add ability to disable peer TLS certificate verification with -cluster.tlsInsecureSkipVerify command-line flag. See mTLS docs for details. See this feature request.

  • FEATURE: add a handler for /api/v1/status/buildinfo endpoint, which is used by Grafana starting from v8.5.0 . See this pull request.

  • FEATURE: add ability to proxy alerting API requests from Grafana to vmalert by passing -vmalert.proxyURL command-line flag to single-node VictoriaMetrics or to vmselect at cluster version of VictoriaMetrics. See this issue.

  • BUGFIX: export staleness markers as null values from JSON export API. Previously they were exported as NaN values. This could break the exported JSON parsing, since NaN values aren't supported by JSON specification.

  • BUGFIX: VictoriaMetrics cluster: close vmselect->vmstorage connections if they were idle for more than 30 seconds. Expose vm_tcpdialer_conns_idle metric at http://vmselect:8481/metrics with the number of idle connections to vmstorage. See this issue.

  • BUGFIX: vmctl: return non-zero exit code on error. This allows handling vmctl errors in shell scripts. Previously vmctl was returning 0 exit code on error. See this issue.

  • BUGFIX: vmctl: prevent from indefinite hang on Ctrl+C. See this issue.

  • BUGFIX: vmagent: properly show scrape_timeout and scrape_interval options at http://vmagent:8429/config page. Previously these options weren't displayed even if they were set in -promscrape.config.

  • BUGFIX: vmagent: handle non-standard http redirect status codes, which may be returned by scrape targets, in the same way as Prometheus does. See this issue.

  • BUGFIX: vmalert: skip template execution during rules' validation. This should prevent from error evaluating annotation template errors when some template functions expect non-empty args. See this issue.

  • BUGFIX: vmalert: fixed truncating alerts expression in table, updated table cell layout. See this issue.

  • BUGFIX: MetricsQL: properly handle joins on time series filtered by values. For example, kube_pod_container_resource_requests{resource="cpu"} * on (namespace,pod) group_left() (kube_pod_status_phase{phase=~"Pending|Running"}==1). This query could result in duplicate time series on the right side error even if ==1 filter leaves only a single time series per (namespace,pod) labels. Now such query is properly executed.

  • BUGFIX: MetricsQL: properly handle scalar default vector, scalar if vector and scalar ifnot vector queries. Previously such queries could return unexpected results from the vector part.

  • BUGFIX: Official Grafana dashboards for VictoriaMetrics: take into account indexdb when calculating disk space usage. See this issue.

v1.76.1

Released at 12-04-2022

Update notes: this release introduces backwards-incompatible changes to communication protocol between vmselect and vmstorage nodes in cluster version of VictoriaMetrics, so read requests to vmselect will fail until the upgrade is complete. These errors will stop after all the vmselect and vmstorage nodes are updated to the new release. It is safe to downgrade to previous releases.

  • FEATURE: vmalert: add support for alert_relabel_configs option at -notifier.config. This option allows configuring relabeling rules for alerts before sending them to configured notifiers. See these docs for details.

  • FEATURE: vmagent: allow passing StatefulSet pod names to -promscrape.cluster.memberNum command-line flag. In this case the member number is automatically extracted from the pod name, which must end with the number in the range 0 ... promscrape.cluster.membersCount-1. For example, vmagent-0, vmagent-1, etc. See this feature request and these docs.

  • BUGFIX: VictoriaMetrics cluster: properly propagate limits at -search.max* command-line flags from vminsert to vmstorage. The limits are -search.maxUniqueTimeseries, -search.maxSeries, -search.maxFederateSeries, -search.maxExportSeries, -search.maxGraphiteSeries and -search.maxTSDBStatusSeries. They weren't propagated to vmstorage because of the bug. These limits were introduced in v1.76.0. See this bug.

  • BUGFIX: fix goroutine leak and possible deadlock when importing invalid data via native binary format. See this pull request.

  • BUGFIX: Graphite Render API: properly calculate hitCount function. Previously it could return empty results if there were no original samples in some parts of the selected time range.

  • BUGFIX: MetricsQL: allow overriding built-in function names inside WITH templates. For example, WITH (sum(a,b) = a + b + 1) sum(x,y) now expands into x + y + 1. Previously such a query would fail with cannot use reserved name error. See this bugreport.

  • BUGFIX: vmui: properly display values greater than 1000 on Y axis. See this issue.

v1.76.0

Released at 07-04-2022

Update notes: this release introduces backwards-incompatible changes to communication protocol between vmselect and vmstorage nodes in cluster version of VictoriaMetrics, so read requests to vmselect will fail until the upgrade is complete. These errors will stop after all the vmselect and vmstorage nodes are updated to the new release. It is safe to downgrade to previous releases.

  • FEATURE: vmctl: add ability to verify files obtained via native export. See these docs and this feature request.

  • FEATURE: vmui: add pre-defined dashboards for per-job CPU usage, memory usage and disk IO usage. See this pull request for details.

  • FEATURE: vmalert: improve compatibility with Prometheus Alert Generator specification. See this pull request.

  • FEATURE: vmalert: add -datasource.disableKeepAlive command-line flag, which can be used for disabling HTTP keep-alive connections to datasources. This option can be useful for distributing load among multiple datasources behind TCP proxy such as HAProxy.

  • FEATURE: Cluster version of VictoriaMetrics: reduce memory usage by up to 50% for vminsert and vmstorage under high ingestion rate.

  • FEATURE: vmgateway: Allow to read -ratelimit.config file from URL. Also add -ratelimit.configCheckInterval command-line option. See this issue.

  • FEATURE: add the following command-line flags, which can be used for fine-grained limiting of CPU and memory usage during various API calls:

    • -search.maxFederateSeries for limiting the number of time series, which can be returned from /federate.
    • -search.maxExportSeries for limiting the number of time series, which can be returned from /api/v1/export.
    • -search.maxSeries for limiting the number of time series, which can be returned from /api/v1/series.
    • -search.maxTSDBStatusSeries for limiting the number of time series, which can be scanned during the request to /api/v1/status/tsdb.
    • -search.maxGraphiteSeries for limiting the number of time series, which can be scanned during the request to Graphite Render API.

Previously the -search.maxUniqueTimeseries command-line flag was used as a global limit for all these APIs. Now the -search.maxUniqueTimeseries is used only for limiting the number of time series, which can be scanned during requests to /api/v1/query and /api/v1/query_range.

When using cluster version of VictoriaMetrics, these command-line flags (including -search.maxUniqueTimeseries) must be passed to vmselect instead of vmstorage.

  • BUGFIX: vmagent and vmauth: reduce the probability of TLS handshake error from XX.XX.XX.XX: EOF errors when -remoteWrite.url points to HTTPS url at vmauth. See this issue.
  • BUGFIX: return Content-Type: text/html response header when requesting / HTTP path at VictoriaMetrics components. Previously text/plain response header was returned, which could lead to broken page formatting. See this issue.
  • BUGFIX: Graphite Render API: accept floating-point values for maxDataPoints query arg, since some clients send floating-point values instead of integer values for this arg.

v1.75.1

Released at 28-03-2022

v1.75.0

Released at 18-03-2022

Update notes: release contains breaking change to vmalert's API introduced in ee396b5. It replaces the api/v1/groups API handler with api/v1/rules handler in order to become compatible with alerts generator specification. See other changes introduced to vmalert here.

  • FEATURE: VictoriaMetrics cluster: add support for mTLS communications between cluster components. See these docs and this feature request.

  • FEATURE: vmalert: add ability to use OAuth2 for -datasource.url, -notifier.url and -remoteRead.url. See the corresponding command-line flags containing oauth2 in their names here.

  • FEATURE: vmalert: add ability to use Bearer Token for -notifier.url via -notifier.bearerToken and -notifier.bearerTokenFile command-line flags. See this issue.

  • FEATURE: vmalert: add sortByLabel template function in order to be consistent with Prometheus. See these docs for more details.

  • FEATURE: vmalert: improve compliance with Prometheus Alert Generator Specification.

  • FEATURE: vmalert: add -rule.resendDelay command-line flag, which specifies the minumum amount of time to wait before resending an alert to Alertmanager (e.g. this is equivalent to -rules.alert.resend-delay option from Prometheus. See this feature request.

  • FEATURE: vmauth: transparently treat Authorization: Token ... request headers as Authorization: Bearer ... request headers. This allows sending requests to vmauth from InfluxDB clients. See this issue. Thanks to @dcircelli for the pull request.

  • FEATURE: do not log trivial network errors such as broken pipe and connection reset by peer. This error could occur when writing data to the client, which closes the connection to VictoriaMetrics due to request timeout or similar reason. See this issue.

  • BUGFIX: Graphite Render API: return an additional point after until timestamp in the same way as Graphite does. Previously VictoriaMetrics didn't return this point, which could result in missing last point on the graph.

  • BUGFIX: properly locate series with the given name and without the given label when using the name{label=~"foo|"} series selector. Previously such series could be skipped. See this issue. Thanks to @jduncan0000 for discovering and fixing the issue.

  • BUGFIX: properly free up memory occupied by deleted cache entries for the following caches: indexdb/dataBlocks, indexdb/indexBlocks, storage/indexBlocks. This should reduce the increased memory usage starting from v1.73.0. See this and this issue.

  • BUGFIX: reduce the interval for checking for free disk space from 30 seconds to 1 second. This should reduce the probability of no space left on device panics when -storage.minFreeDiskSpaceBytes is set to too low values. See this issue.

  • BUGFIX: vmagent: prevent from panic at vmagent when importing a time series with big number of samples. See this issue. Thanks to @bleedfish for discovering and fixing the issue.

v1.74.0

Released at 03-03-2022

Update notes: In this release VictoriaMetrics may use some extra memory due to issues #2242 and #2007. These issues were addressed in v1.75.0, so we recommend updating straight to it.

  • FEATURE: add support for conditional relabeling via if filter. The if filter can contain arbitrary series selector. For example, the following rule drops targets matching foo{bar="baz"} series selector:
- action: drop
  if: 'foo{bar="baz"}'

This rule is equivalent to less clear traditional one:

- action: drop
  source_labels: [__name__, bar]
  regex: 'foo;baz'

See relabeling docs and this issue for more details.

  • FEATURE: reduce memory usage for various caches under high churn rate.

  • FEATURE: vmagent: re-use Kafka client when pushing data from many tenants to Kafka. Previously a separate Kafka client was created per each tenant. This could lead to increased load on Kafka. See how to push data from vmagent to Kafka.

  • FEATURE: improve performance when registering new time series. See this issue. Thanks to @ahfuzhang .

  • BUGFIX: return the proper number of datapoints from moving*() functions such as movingAverage() in Graphite Render API. Previously these functions could return too big number of samples if maxDataPoints query arg is explicitly passed to /render API.

  • BUGFIX: properly handle series selector containing a filter for multiple metric names plus a negative filter. For example, {__name__=~"foo|bar",job!="baz"} . Previously VictoriaMetrics could return series with foo or bar names and with job="baz". See this issue.

  • BUGFIX: vmgateway: properly parse JWT tokens if they are encoded with URL-safe base64 encoding.

v1.73.1

Released at 22-02-2022

Update notes: In this release VictoriaMetrics may use some extra memory due to issues #2242 and #2007. These issues were addressed in v1.75.0, so we recommend updating straight to it.

  • FEATURE: allow overriding default limits for the following in-memory caches, which usually occupy the most memory:

    • storage/tsid - the cache speeds up lookups of internal metric ids by metric_name{labels...} during data ingestion. The size for this cache can be tuned with -storage.cacheSizeStorageTSID command-line flag.
    • indexdb/dataBlocks - the cache speeds up data lookups in <-storageDataPath>/indexdb files. The size for this cache can be tuned with -storage.cacheSizeIndexDBDataBlocks command-line flag.
    • indexdb/indexBlocks - the cache speeds up index lookups in <-storageDataPath>/indexdb files. The size for this cache can be tuned with -storage.cacheSizeIndexDBIndexBlocks command-line flag. See also cache tuning docs. See this issue.
  • FEATURE: add -influxDBLabel command-line flag for overriding db label name for the data imported into VictoriaMetrics via InfluxDB line protocol. Thanks to @johnatannvmd for the pull request.

  • FEATURE: return X-Influxdb-Version HTTP header in responses to InfluxDB write requests. This is needed for some InfluxDB clients. See this comment and this issue.

  • BUGFIX: reduce memory usage during the first three hours after the upgrade from versions older than v1.73.0. The memory usage spike was related to the need of in-memory caches' re-population after the upgrade because of the fix for this issue. Now cache size limits are reduced in order to occupy less memory during the upgrade.

  • BUGFIX: fix a bug, which could significantly slow down requests to /api/v1/labels and /api/v1/label/<label_name>/values. These APIs are used by Grafana for auto-completion of label names and label values. See this issue.

  • BUGFIX: vmalert: add support for $externalLabels and $externalURL template vars in the same way as Prometheus does. See this issue.

  • BUGFIX: vmalert: make sure notifiers are discovered during initialization if they are configured via consul_sd_configs. Previously they could be discovered in 30 seconds (the default value for -promscrape.consulSDCheckInterval command-line flag) after the initialization. See this pull request.

  • BUGFIX: update default value for -promscrape.fileSDCheckInterval, so it matches default duration used by Prometheus for checking for updates in file_sd_configs. See this issue. Thanks to @corporate-gadfly for the fix.

  • BUGFIX: VictoriaMetrics cluster: do not return partial responses from vmselect if at least a single vmstorage node was reachable and returned an app-level error. Such errors are usually related to cluster mis-configuration, so they must be returned to the caller instead of being masked by partial responses. Partial responses can be returned only if some of vmstorage nodes are unreachable during the query. This may help the following issues: one, two.

v1.73.0

Released at 14-02-2022

Update notes: In this release VictoriaMetrics may use some extra memory described in issues #2242 and #2007. These issues were addressed in v1.75.0, so we recommend updating straight to it.

  • FEATURE: publish VictoriaMetrics binaries for MacOS amd64 and MacOS arm64 (aka MacBook M1) at releases page. See this issue and this issue.

  • FEATURE: reduce CPU and disk IO usage during indexdb rotation once per -retentionPeriod. See this issue.

  • FEATURE: VictoriaMetrics cluster: add -dropSamplesOnOverload command-line flag for vminsert. If this flag is set, then vminsert drops incoming data if the destination vmstorage is temporarily unavailable or cannot keep up with the ingestion rate. The number of dropped rows can be monitored via vm_rpc_rows_dropped_on_overload_total metric at vminsert.

  • FEATURE: VictoriaMetrics cluster: improve re-routing logic, so it re-routes incoming data more evenly if some of vmstorage nodes are temporarily unavailable and/or accept data at slower rate than other vmstorage nodes. Also significantly reduce possible re-routing storm when vminsert runs with -disableRerouting=false command-line flag. This should help the following issues: one, two, three, four, five.

  • FEATURE: MetricsQL: cover more cases with the label filters' propagation optimization. This should improve the average performance for practical queries. The following cases are additionally covered:

    • Multi-level transform functions. For example, abs(round(foo{a="b"})) + bar{x="y"} is now optimized to abs(round(foo{a="b",x="y"})) + bar{a="b",x="y"}
    • Binary operations with on(), without(), group_left() and group_right() modifiers. For example, foo{a="b"} on (a) + bar is now optimized to foo{a="b"} on (a) + bar{a="b"}
    • Multi-level binary operations. For example, foo{a="b"} + bar{x="y"} + baz{z="q"} is now optimized to foo{a="b",x="y",z="q"} + bar{a="b",x="y",z="q"} + baz{a="b",x="y",z="q"}
    • Aggregate functions. For example, sum(foo{a="b"}) by (c) + bar{c="d"} is now optimized to sum(foo{a="b",c="d"}) by (c) + bar{c="d"}
  • FEATURE MetricsQL: optimize joining with *_info labels. For example: kube_pod_created{namespace="prod"} * on (uid) group_left(node) kube_pod_info now automatically adds the needed filters on uid label to kube_pod_info before selecting series for the right side of * operation. This may save CPU, RAM and disk IO resources. See this article for details on *_info labels. See this issue.

  • FEATURE: all: improve performance for arm64 builds of VictoriaMetrics components by up to 15%. See this pull request.

  • FEATURE: all: expose process_cpu_cores_available metric, which shows the number of CPU cores available to the app. The number can be fractional if the corresponding cgroup limit is set to a fractional value. This metric is useful for alerting on CPU saturation. For example, the following query alerts when the app uses more than 90% of CPU during the last 5 minutes: rate(process_cpu_seconds_total[5m]) / process_cpu_cores_available > 0.9 . See this issue.

  • FEATURE: vmalert: add ability to configure notifiers (e.g. alertmanager) via a file in the way similar to Prometheus. See these docs, this pull request.

  • FEATURE: vmalert: add support for Consul service discovery for notifiers. See this issue.

  • FEATURE: vmalert: add support for specifying Basic Auth password for notifiers via a file. See this issue.

  • FEATURE: vmagent: provide the ability to fetch target responses on behalf of vmagent by clicking the response link for the needed target at /targets page. This feature may be useful for debugging responses from targets located in isolated environments.

  • FEATURE: vmagent: show the total number of scrapes and the total number of scrape errors per target at /targets page. This information may be useful when debugging unreliable scrape targets.

  • FEATURE: vmagent and single-node VictoriaMetrics: disallow unknown fields at -promscrape.config file. Previously unknown fields were allowed. This could lead to long-living silent config errors. The previous behaviour can be returned by passing -promscrape.config.strictParse=false command-line flag.

  • FEATURE: vmagent: add __meta_kubernetes_endpointslice_label* and __meta_kubernetes_endpointslice_annotation* labels for role: endpointslice targets in kubernetes_sd_config to be consistent with other role values. See this issue.

  • FEATURE: vmagent: add collapse all and expand all buttons to http://vmagent:8429/targets page. See this issue.

  • FEATURE: vmagent: support Prometheus-like durations in -promscrape.config. See this comment.

  • FEATURE: automatically re-read -tlsCertFile and -tlsKeyFile files, so their contents can be updated without the need to restart VictoriaMetrics apps. See this issue.

  • BUGFIX: calculate absent_over_time() in the same way as Prometheus does. Previously it could return multiple time series instead of at most one time series like Prometheus does. See this issue.

  • BUGFIX: return proper results from highestMax() function at Graphite render API. Previously it was incorrectly returning timeseries with min peaks instead of max peaks.

  • BUGFIX: properly limit indexdb cache sizes. Previously they could exceed values set via -memory.allowedPercent and/or -memory.allowedBytes when indexdb contained many data parts. See this issue.

  • BUGFIX: vmui: fix a bug, which could break time range picker when editing From or To input fields. See this issue.

  • BUGFIX: vmui: fix a bug, which could break switching between graph, json and table views. See this issue.

  • BUGFIX: vmui: fix possible UI freeze after querying node_uname_info time series. See this issue.

  • BUGFIX: show the original location of the warning or error message when logging throttled messages. Previously the location inside lib/logger/throttler.go was shown. This could increase the complexity of debugging.

  • BUGFIX: vmalert: fix links at web UI. See this issue.

  • BUGFIX: vmagent: properly discover pods without exposed ports for the given service for role: endpoints and role: endpointslice in kubernetes_sd_config. See this issue.

  • BUGFIX: vmagent: properly display zone contents for gce_sd_configs section at http://vmagent:8429/config page. See this issue. Thanks to @artifactori for the bugfix.

  • BUGFIX: vmagent: properly handle all_tenants: true config option at openstack_sd_config. See this issue.

v1.72.0

Released at 18-01-2022

  • FEATURE: MetricsQL: add support for @ modifier, which is enabled by default in Prometheus starting from Prometheus v2.33.0. See these docs and this feature request. VictoriaMetrics extends @ modifier with the following additional features:
    • It can contain arbitrary expression. For example, foo @ (end() - 1h) would return foo value at end - 1 hour timestamp on the selected time range [start ... end]. Another example: foo @ (now() - 10m) would return foo value 10 minutes ago from the current time.
    • It can be put everywhere in the query. For example, sum(foo) @ start() would calculate sum(foo) at start timestamp on the selected time range [start ... end].
  • FEATURE: MetricsQL: add support for optional keep_metric_names modifier, which can be applied to all the rollup functions and transform functions. This modifier prevents from deleting metric names from function results. For example, rate({__name__=~"foo|bar"}[5m]) keep_metric_names leaves foo and bar metric names in rate() results. This feature provides an additional workaround for this issue.
  • FEATURE: vmagent: add support for Kubernetes service discovery in the current namespace in the same way as Prometheus does. For example, the following config limits pod discovery to the namespace where vmagent runs:
  scrape_configs:
  - job_name: 'kubernetes-pods'
    kubernetes_sd_configs:
    - role: pod
      namespaces:
        own_namespace: true
  • FEATURE: vmagent: add __meta_kubernetes_node_provider_id label for discovered Kubernetes nodes in the same way as Prometheus does.

  • FEATURE: vmagent: log error message when remote storage returns 400 or 409 http errors. This should simplify detection and debugging of this case. See this issue.

  • FEATURE: vmagent: expose promscrape_stale_samples_created_total metric for monitoring the total number of created stale samples when scraping Prometheus targets. See these docs for the information on when stale samples (aka staleness markers) can be created.

  • FEATURE: vmrestore: store restore-in-progress file in -dst directory while vmrestore is running. This file is automatically deleted when vmrestore is successfully finished. This helps detecting incompletely restored data on VictoriaMetrics start. See this issue.

  • FEATURE: vmctl: print the last sample timestamp when the data migration is interrupted either by user or by error. This helps continuing the data migration from the interruption moment. See this issue.

  • FEATURE: vmalert: expose vmalert_remotewrite_total metric at /metrics page. This makes possible calculating SLOs for error rate during writing recording rules and alert state to -remoteWrite.url with the query vmalert_remotewrite_errors_total / vmalert_remotewrite_total. See this issue. Thanks to @afoninsky .

  • FEATURE: vmalert: add stripPort template function in the same way as Prometheus does.

  • FEATURE: vmalert: add parseDuration template function in the same way as Prometheus does.

  • FEATURE: MetricsQL: add stale_samples_over_time(m[d]) function for calculating the number of staleness marks for time series m over the duration d. This function may be useful for detecting flapping metrics at scrape targets, which periodically disappear and then appear again.

  • FEATURE: vmgateway: add support for extra_filters option. See this issue.

  • FEATURE: vmui: improve UX according to this feature request. Thanks to @Loori-R .

  • FEATURE: vmui: limit the number of requests sent to VictoriaMetrics during zooming / scrolling. See this issue.

  • BUGFIX: vmagent: make sure that vmagent replicas scrape the same targets at different time offsets when replication is enabled in vmagent clustering mode. This guarantees that the deduplication consistently leaves samples from the same vmagent replica.

  • BUGFIX: return the proper response stub from /api/v1/query_exemplars handler, which is needed for Grafana v8+. See this issue.

  • BUGFIX: vmctl: fix a few edge cases and improve migration speed for OpenTSDB importer. See this pull request.

  • BUGFIX: fix possible data race when searching for time series matching {key=~"value|"} filter over time range covering multipe days. See this pull request. Thanks to @waldoweng for the provided fix.

  • BUGFIX: vmagent: do not send staleness markers on graceful shutdown. This follows Prometheus behavior. See this comment.

  • BUGFIX: vmagent: properly set __address__ label in dockerswarm_sd_config. See this issue. Thanks to @ashtuchkin for the fix.

  • BUGFIX: vmui: fix incorrect calculations for graph limits on y axis. This could result in incorrect graph rendering in some cases. See this issue.

  • BUGFIX: vmui: fix handling for multi-line queries. See this issue.

v1.71.0

Released at 20-12-2021

Update notes: deduplication logic was slightly changed on the release, which may cause extra background merges for already existing data parts for installations with -dedup.minScrapeInterval flag value greater than 0. This process is intentionally limited by one CPU core, but still can result into increase of CPU usage until merges are finished.

We recommend updating in "off-peak" time when load on the VictoriaMetrics is on its minimum.

  • FEATURE: VictoriaMetrics enterprise: add multi-level downsampling support. See these docs and this feature request.

  • FEATURE: vmui: add ability to analyze the correlation between two queries on a single graph. Just click +Query button, enter the second query in the newly appeared input field and press Ctrl+Enter. Results for both queries should be displayed simultaneously on the same graph. Every query has its own vertical scale, which is displayed on the left and the right side of the graph. Lines for the second query are dashed. See this pull request.

  • FEATURE: vmui: add ability to override the interval between returned datapoints. By default it is automatically calculated depending on the selected time range and horizontal resolution of the graph. Now it is possible to override it with custom values. This may be useful during data exploration and debugging.

  • FEATURE: accept optional extra_filters[]=series_selector query args at Prometheus query APIs additionally to extra_label query args. This allows enforcing additional filters for all the Prometheus query APIs by using vmgateway or vmauth. See this feature request.

  • FEATURE: vmauth: allow specifying http and https urls in -auth.config command-line flag. See this pull request. Thanks for @TFM93 .

  • FEATURE: vmagent: allow specifying http and https urls in the following command-line flags: -promscrape.config, -remoteWrite.relabelConfig and -remoteWrite.urlRelabelConfig.

  • FEATURE: vminsert: allow specifying http and https urls in -relabelConfig command-line flag.

  • FEATURE: vminsert: add -maxLabelValueLen command-line flag for the ability to configure the maximum length of label value. See this feature request.

  • FEATURE: preserve the order of time series passed to limit_offset function. This allows implementing series paging via limit_offset(limit, offset, sort_by_label(...)). See this and this issues.

  • FEATURE: automaticall convert (value1|...|valueN) into {value1,...,valueN} inside __graphite__ pseudo-label. This allows using Grafana multi-value template variables inside __graphite__ pseudo-label. For example, {__graphite__=~"foo.($bar)"} is expanded to {__graphite__=~"foo.{x,y}"} if both x and y are selected for $bar template variable. See these docs for details.

  • FEATURE: add timestamp_with_name function. It works the same as timestamp, but leaves the original time series names, so it can be used in queries, which match multiple time series names: timestamp_with_name({foo="bar"}[1h]). See this comment for more context.

  • FEATURE: add changes_prometheus, increase_prometheus and delta_prometheus functions, which don't take into account the previous sample before the given lookbehind window specified in square brackets. These functions may be used when the Prometheus behaviour for changes(), increase() and delta() functions is needed to be preserved. VictoriaMetrics uses slightly different behaviour for changes(), increase() and delta() functions by default - see this article for details. See this issue.

  • BUGFIX: fix unaligned 64-bit atomic operation panic on 32-bit architectures, which has been introduced in v1.70.0. See this issue.

  • BUGFIX: vmalert: restore the ability to use $labels.alertname in labels templating. See this issue.

  • BUGFIX: vmui: add missing query caption to the input field for the query. See this issue.

  • BUGFIX: vmui: fix navigation over query history with Ctrl+up/down and fix zoom relatively to the cursor position. See this pull request.

  • BUGFIX: deduplicate samples more thoroughly if deduplication is enabled. Previously some duplicate samples may be left on disk for time series with high churn rate. This may result in bigger storage space requirements.

  • BUGFIX: vmagent: follow up to 5 redirects when follow_redirects: true is set for a particular scrape config. Previously only a single redirect was performed in this case. It is expected these redirects are performed to the original hostname. See this issue.

  • BUGFIX: de-duplicate data exported via /api/v1/export/csv by default if deduplication is enabled. The de-duplication can be disabled by passing reduce_mem_usage=1 query arg to /api/v1/export/csv. See this issue.

  • BUGFIX: vmalert: properly store historical data to old Prometheus versions. See this issue.

v1.70.0

Released at 02-12-2021

  • FEATURE: vmalert: add ability to pass arbitrary query args to -datasource.url on a per-group basis via params option. See this pull request.

  • FEATURE: add now() function to MetricsQL. This function returns the current timestamp in seconds. See these docs.

  • FEATURE: vmauth: allow using optional name field in configs. This field is then used as username label value for vmauth_user_requests_total metric. See this feature request.

  • FEATURE: vmagent: export vm_persistentqueue_read_duration_seconds_total and vm_persistentqueue_write_duration_seconds_total metrics, which can be used for detecting persistent queue saturation with rate(vm_persistentqueue_write_duration_seconds_total) > 0.9 alerting rule.

  • FEATURE: export vm_filestream_read_duration_seconds_total and vm_filestream_write_duration_seconds_total metrics, which can be used for detecting persistent disk saturation with rate(vm_filestream_read_duration_seconds_total) > 0.9 alerting rule.

  • FEATURE: export vm_cache_size_max_bytes metrics, which show capacity for various caches. These metrics can be used for determining caches with reach its capacity with vm_cache_size_bytes / vm_cache_size_max_bytes > 0.9 query.

  • FEATURE: vmbackup, vmrestore: add -s3ForcePathStyle command-line flag, which can be used for making backups to Aliyun OSS. See this pull request.

  • FEATURE: vmctl: improve data migration from OpenTSDB. See this pull request. Thanks to @johnseekins .

  • FEATURE: suppress connection reset by peer errors when remote client resets TCP connection to VictoriaMetrics / vmagent while ingesting the data via InfluxDB line protocol, Graphite protocol or OpenTSDB protocol. This error is expected, so there is no need in logging it.

  • FEATURE: vmui: store the display type in URL, so it isn't lost when copy-pasting the URL. See this feature request.

  • FEATURE: vmalert: make -notifier.url command-line flag optional. This flag can be omitted if vmalert is used solely for recording rules and doesn't evaluate alerting rules. See this pull request.

  • FEATURE: vmbackup, vmrestore: export internal metrics at http://vmbackup:8420/metrics and http://vmrestore:8421/metrics for better visibility of the backup/restore process.

  • FEATURE: allow trailing whitespace after the timestamp when parsing Graphite plaintext lines. See this issue.

  • FEATURE: expose /-/healthy and /-/ready endpoints as Prometheus does. This is needed for improving integration with third-party solutions, which rely on these endpoints. See this issue.

  • BUGFIX: vmagent: prevent from scraping duplicate targets if -promscrape.dropOriginalLabels command-line flag is set. See this issue. Thanks to @guidao for the fix.

  • BUGFIX: vmstorage enterprise: added missing vm_tenant_used_tenant_bytes metric, which shows the approximate per-tenant disk usage. See these docs and this issue.

  • BUGFIX: vmauth: properly take into account the value passed to -maxIdleConnsPerBackend command-line flag. See this issue.

  • BUGFIX: vmagent: fix reading data from Kafka.

  • BUGFIX: vmalert: fix replay mode in enterprise version.

  • BUGFIX: consistently return zero from deriv() function applied to a constant time series. Previously it could return small non-zero values in this case.

  • BUGFIX: vmrestore: properly resume downloading for partially downloaded big files. Previously such files were re-downloaded from the beginning after the interrupt. Now only the remaining parts of the file are downloaded. This allows saving network bandwidth. See this issue.

  • BUGFIX: vmui: do not store the last query across vmui page reloads. See this issue.

  • BUGFIX: vmui: fix Cannot read properties of undefined error at table view. See this issue.

v1.69.0

Released at 08-11-2021

  • FEATURE: vmalert: allow groups with empty rules list like Prometheus does. See this pull request.

  • FEATURE: vmalert: allow groups with default tenant in -clusterMode. Default tenant values can be specified via -defaultTenant.prometheus and -defaultTenant.graphite. See these docs.

  • FEATURE: vmagent: add collapse and expand buttons per each group of targets with the same job_name at http://vmagent:8429/targets page.

  • FEATURE: automatically detect timestamp precision (ns, us, ms or s) for the data ingested into VictoriaMetrics via InfluxDB line protocol.

  • FEATURE: vmagent: add ability to protect /config page with auth key via -configAuthKey command-line flag. This page may contain sensitive config information, so it may be good to restrict access to this page. See this issue.

  • FEATURE: vmagent: hide passwords and auth tokens at /config page like Prometheus does. See this issue.

  • FEATURE: vmagent: add -promscrape.maxResponseHeadersSize command-line flag for tuning the maximum HTTP response headers size for Prometheus scrape targets.

  • FEATURE: vmagent: send data to multiple configured remote storage systems in parallel (e.g. when multiple -remoteWrite.url flag values are specified). This should improve data ingestion speed.

  • FEATURE: vmagent: add -remoteWrite.maxRowsPerBlock command-line flag for tuning the number of samples to send to remote storage per each block. Bigger values may improve data ingestion performance at the cost of higher memory usage.

  • FEATURE: vmagent: distribute Kafka messages among all the partitions when writing data to Kafka.

  • FEATURE: add label_graphite_group function for extracting the given groups from Graphite metric names.

  • FEATURE: add duration_over_time function for calculating the actual lifetime of the time series with possible gaps. See this feature request.

  • FEATURE: add limit_offset function, which can be used for implementing simple paging over big number of time series. See this feature request.

  • BUGFIX: vmagent: reduce the increased memory usage when scraping targets with big number of metrics which periodically change. The memory usage has been increased in v1.68.0 after vmagent started generating staleness markers in stream parse mode. See this issue.

  • BUGFIX: vmagent: properly display proxy_url config option at http://vmagent:8429/config page. See this issue.

  • BUGFIX: fix tests for Apple M1. See this issue.

v1.68.0

Released at 22-10-2021

  • FEATURE: vmagent: expose -promscrape.config contents at /config page as Prometheus does. See this feature request.

  • FEATURE: vmagent: add show original labels button per each scrape target displayed at http://vmagent:8429/targets page. This should improve debuggability for service discovery and relabeling issues similar to this one. See this feature request.

  • FEATURE: vmagent: shard targets among cluster nodes after the relabeling is applied. This should guarantee that targets with the same set of labels go to the same vmagent node in the cluster. See this issue.

  • FEATURE: vmagent: atomatically switch to stream parsing mode if the response from the given target exceeds the command-line flag value -promscrape.minResponseSizeForStreamParse. This should reduce memory usage when vmagent scrapes targets with non-uniform response sizes (this is the case in Kubernetes monitoring).

  • FEATURE: vmagent: send Prometheus-like staleness marks in stream parsing mode. Previously staleness marks wern't sent in stream parsing mode. See these docs for details.

  • FEATURE: vmagent: properly calculate scrape_series_added metric for targets in stream parsing mode. Previously it was set to 0 in stream parsing mode. See more details about this metric.

  • FEATURE: vmagent: expose promscrape_series_limit_max_series and promscrape_series_limit_current_series metrics at http://vmagent:8429/metrics for scrape targets with the enabled series limiter.

  • FEATURE: vmagent: return error if sample_limit or series_limit options are set when stream parsing mode is enabled, since these limits cannot be applied in stream parsing mode.

  • FEATURE: vmalert: add -remoteRead.disablePathAppend command-line flag, which allows specifying the full -remoteRead.url. If -remoteRead.disablePathAppend is set, then vmalert doesn't add /api/v1/query suffix to -remoteRead.url.

  • FEATURE: add trigonometric functions, which are going to be added in Prometheus 2.31: acosh, asinh, atan, atanh, cosh, deg, rad, sinh, tan, tanh. Also add atan2 binary operator. See this pull request.

  • FEATURE: consistently return the same set of time series from limitk function. This improves the usability of periodically refreshed graphs.

  • FEATURE: vmui: varios UX improvements. See this pull request and these docs.

  • FEATURE: vmauth: add ability to specify HTTP headers, which will be sent in requests to backends. See this feature request.

  • FEATURE: add /flags page to all the VictoriaMetrics components. This page contains command-line flags passed to the component.

  • FEATURE: allow using tab separators additionally to whitespace separators when ingesting data in Graphite plaintext protocol. Such separators are supported by Carbon-c-relay.

  • BUGFIX: vmstorage: fix unaligned 64-bit atomic operation panic on 32-bit architectures (arm and 386). The panic has been introduced in v1.67.0.

  • BUGFIX: vmalert, vmauth: prevent from frequent closing of TCP connections established to backends under high load. This should reduce the number of TCP sockets in TIME_WAIT state at vmalert and vmauth under high load. See this pull request.

  • BUGFIX: vmalert: correctly calculate alert ID including extra labels. Previously, ID for alert entity was generated without alertname or groupname. This led to collision, when multiple alerting rules within the same group producing same labelsets. E.g. expr: sum(metric1) by (job) > 0 and expr: sum(metric2) by (job) > 0 could result into same labelset job: "job". The bugfix adds all extra labels right after receiving response from the datasource. See this pull request.

  • BUGFIX: vmalert: fix links in Web UI. See this pull request.

  • BUGFIX: vmagent: set honor_timestamps: true by default in scrape configs if this options isn't set explicitly. This aligns the behaviour with Prometheus.

  • BUGFIX: vmagent: group scrape targets by the original job names at http://vmagent:8429/targets page like Prometheus does. Previously they were grouped by the job name after relabeling, which may result in unexpected empty target groups. See this issue.

  • BUGFIX: vmctl: fix importing boolean fields from InfluxDB line protocol. See this issue.

v1.67.0

Released at 08-10-2021

  • FEATURE: add ability to accept metrics from DataDog agent and DogStatsD. See these docs. This option simplifies the migration path from DataDog to VictoriaMetrics. See also this issue.

  • FEATURE: vmagent enterprise: add support for data reading and writing from/to Apache Kafka. See these docs.

  • FEATURE: vmui: switch to μPlot and add ability to naturally scroll and zoom graphs. See these docs. Thanks to @Loori-R.

  • FEATURE: vmstorage: stop accepting new data if -storageDataPath directory contains less than -storage.minFreeDiskSpaceBytes of free space. This should prevent from out of disk space crashes. See this feature request.

  • FEATURE: calculate quantiles in the same way as Prometheus does in such functions as quantile_over_time and quantile. Previously results from VictoriaMetrics could be slightly different than results from Prometheus. See this and this issues.

  • FEATURE: add rollup_scrape_interval(m[d]) function to MetricsQL, which returns min, max and avg values for the interval between samples for m on the given lookbehind window d.

  • FEATURE: add topk_last(k, q) and bottomk_last(k, q) functions to MetricsQL, which return up to k time series from q with the maximum / minimum last value on the graph.

  • BUGFIX: align behavior of the queries a or on (labels) b, a and on (labels) b and a unless on (labels) b where b has multiple time series with the given labels to Prometheus behavior. See this pull request.

  • BUGFIX: vmagent: fix openstack_sd_config service discovery when both domain_name and project_id config options are set. See this issue.

  • BUGFIX: return proper values (zeroes) from stddev_over_time and stdvar_over_time functions when the lookbehind window in square brackets contains only a single sample. Previously the sample value was incorrectly returned in this case.

  • BUGFIX: vminsert: fix uneven distribution of time series among storage nodes in multi-level cluster setup. See this issue.

v1.66.2

Released at 23-09-2021

  • FEATURE: vmagent: add vm_promscrape_max_scrape_size_exceeded_errors_total metric for counting of the failed scrapes due to the exceeded response size (the response size limit can be configured via -promscrape.maxScrapeSize command-line flag). See this issue.

  • BUGFIX: vmalert: properly reload rule groups if only the interval config option is changed. See this issue.

  • BUGFIX: properly handle {__name__=~"prefix(suffix1|suffix2)",other_label="..."} queries. They may return unexpected empty responses since v1.66.0. See this issue.

v1.66.1

Released at 22-09-2021

  • FEATURE: add -cluster and/or -enterprise suffixes to short_version label at vm_app_version metric exposed at /metrics page of every VictoriaMetrics component. See this issue.

  • BUGFIX: vmselect: fix accessing Graphite APIs. The access has been broken in v1.66.0, because /graphite/* path prefix accidentally clashed with /graph* path prefix used for VictoriaMetrics UI (aka vmui).

  • BUGFIX: fix parsing regex: <bool_or_number> in relabeling rules (for example, regex: true or regex: 123). The bug has been introduced in v1.66.0.

v1.66.0

Released at 20-09-2021

  • FEATURE: vmalert: add web UI with the list of alerting groups, alerts and alert statuses. See this pull request.

  • FEATURE: vmalert: add -rule.maxResolveDuration command-line flag, which could be used for limiting the auto-resolve duration for the alerting rule. By default it is limited to 3x evaluation interval. This could be too high for big evaluation intervals. See this issue.

  • FEATURE: vmalert: add support for Bearer token authorization for -datasource.url, -remoteRead.url and -remoteWrite.url. See this issue.

  • FEATURE: vmagent: send stale markers for disappeared metrics like Prometheus does. Previously stale markers were sent only when the scrape target disappears or when it becomes temporarily unavailable. See these docs for details.

  • FEATURE: vmagent: add ability to set series_limit option for a particular scrape target via __series_limit__ label. This allows setting the limit on the number of time series on a per-target basis. See these docs for details.

  • FEATURE: vmagent: add ability to set stream_parse option for a particular scrape target via __stream_parse__ label. This allows managing the stream parsing mode on a per-target basis. See these docs for details.

  • FEATURE: vmagent: add ability to set scrape_interval and scrape_timeout options for a particular target via __scrape_interval__ and __scrape_timeout__ labels in the same way as Prometheus 2.30 does. See this pull request.

  • FEATURE: vmagent: generate scrape_timeout_seconds metric per each scrape target, so the target saturation could be calculated with scrape_duration_seconds / scrape_timeout_seconds. See the corresponding pull request from Prometheus 2.30.

  • FEATURE: vmagent: reduce CPU usage when calculating the number of newly added series per scrape (this number is sent to remote storage in scrape_series_added metric).

  • FEATURE: vmagent: reduce CPU usage when applying series_limit to scrape targets with constant set of metrics. See more information about series_limit here.

  • FEATURE: vminsert: disable rerouting by default when a few of vmstorage nodes start accepting data at lower speed than the rest of vmstorage nodes. This should improve VictoriaMetrics cluster stability during rolling restarts and during spikes in time series churn rate. The rerouting can be enabled by passing -disableRerouting=false command-line flag to vminsert.

  • FEATURE: vmauth: do not put invalid auth tokens into log by default due to security reasons. The logging can be returned back by passing -logInvalidAuthTokens command-line flag to vmauth. Requests with invalid auth tokens are counted at vmagent_http_request_errors_total{reason="invalid_auth_token"} metric exposed by vmauth at /metrics page.

  • FEATURE: add new relabeling actions: keep_metrics and drop_metrics. This simplifies metrics filtering by metric names. See these docs for more details.

  • FEATURE: allow splitting long regex in relabeling filters into an array of shorter regexps, which can be put into multiple lines for better readability and maintainability. See these docs for more details.

  • FEATURE: optimize performance for queries with regexp filters on metric name like {__name__=~"metric1|...|metricN"}. See this pull request from @faceair.

  • FEATURE: vmui: use Prometheus-compatible query args, so vmui could be accessed from graph editor in Grafana. See this pull request. Thanks to @Loori-R.

  • FEATURE: vmselect: automatically add missing port to -storageNode hostnames. For example, -storageNode=vmstorage1,vmstorage2 is automatically translated to -storageNode=vmstorage1:8401,vmstorage2:8401. This simplifies manual setup of VictoriaMetrics cluster.

  • FEATURE: vminsert: automatically add missing port to -storageNode hostnames. For example, -storageNode=vmstorage1,vmstorage2 is automatically translated to -storageNode=vmstorage1:8400,vmstorage2:8400. This simplifies manual setup of VictoriaMetrics cluster.

  • FEATURE: add mad(q) function to MetricsQL. It calculates Median absolute deviation for groups of points with identical timestamps across multiple time series.

  • FEATURE: add outliers_mad(tolerance, q) function to MetricsQL. It returns time series with peaks outside the Median absolute deviation multiplied by tolerance.

  • FEATURE: add histogram_quantiles("phiLabel", phi1, ..., phiN, buckets) function to MetricsQL. It calculates the given phi*-quantiles over the given buckets and returns time series per each quantile with the corresponding {phiLabel="phi*"} label.

  • FEATURE: add quantiles_over_time("phiLabel", phi1, ..., phiN, series_selector[d]) function to MetricsQL. It calculates the given phi*-quantiles over raw samples selected by series_selector on the given lookbehind window d. It returns time series per each quantile with the corresponding {phiLabel="phi*"} label.

  • FEATURE: enterprise: do not ask for -eula flag if -version flag is passed to enteprise app. See this issue.

  • BUGFIX: properly handle queries with multiple filters matching empty labels such as metric{label1=~"foo|",label2="bar|"}. This filter must match the following series: metric, metric{label1="foo"}, metric{label2="bar"} and metric{label1="foo",label2="bar"}. Previously it was matching only metric{label1="foo",label2="bar"}. See this issue.

  • BUGFIX: vmselect: reset connection timeouts after each request to vmstorage. This should prevent from cannot read data in 0.000 seconds: unexpected EOF warning in logs. See this issue. Thanks to @mxlxm .

  • BUGFIX: keep metric name for time series returned from rollup_candlestick function, since the returned series don't change the meaning of the original series. See this issue.

  • BUGFIX: use Prometheus-compatible label value formatting for count_values function. Previously big values could be improperly formatted, which could break query results, which rely on label value such as ... on(label) count_values("label", ...).

  • BUGFIX: vmagent: properly use https scheme for wildcard TLS certificates for role: ingress targets in Kubernetes service discovery. See this issue.

  • BUGFIX: vmagent: support host networking mode for docker_sd_config. See this issue.

  • BUGFIX: fix non-repeatable results from quantile_over_time() function when the number of input samples exceeds 1000. See this issue.

  • BUGFIX: vmagent: fix EC2 zone discovery when filters are specified in ec2_sc_config. See this issue.

v1.65.0

Released at 01-09-2021

  • FEATURE: vmagent: add ability to read scrape configs from multiple files specified in scrape_config_files section. See this issue.

  • FEATURE: vmagent: reduce memory usage and CPU usage when Prometheus staleness tracking is enabled for metrics exported from the deleted or disappeared scrape targets.

  • FEATURE: vmagent: add the ability to limit the number of unique time series scraped per each target. This can be done either globally via -promscrape.seriesLimitPerTarget command-line option or on per-target basis via series_limit option at scrape_config section. See the updated docs on cardinality limiter and this issue.

  • FEATURE: vmagent: discover role: ingress and role: endpointslice in kubernetes_sd_config via v1 API instead of v1beta1 API if Kubernetes supports it. This fixes service discovery in Kubernetes v1.22 and newer versions. See these docs.

  • FEATURE: take into account failed queries in vm_request_duration_seconds summary at /metrics. Previously only successful queries were taken into account. This could result in skewed summary. See this pull request.

  • FEATURE: vmalert: add an official dashboard for vmalert. See these docs.

  • FEATURE: vmalert: add ability to set additional labels per group via labels config section. See this issue.

  • FEATURE: vmalert: add -disableAlertgroupLabel command-line flag for disabling the label with alert group name. This may be needed for proper deduplication in Alertmanager. See this issue.

  • FEATURE: update Go builder from v1.16.7 to v1.17.0. This improves data ingestion and query performance by up to 5% according to benchmarks. See the release post for Go1.17.

  • FEATURE: vmagent: expose promscrape_discovery_http_errors_total metric, which can be used for monitoring the number of failed discovery attempts per each http_sd config.

  • FEATURE: do not reset response cache when a sample with old timestamp is ingested into VictoriaMetrics if -search.disableAutoCacheReset command-line option is set. See this feature request.

  • FEATURE: add quantiles("phiLabel", phi1, ..., phiN, q) aggregate function to MetricsQL, which calculates the given phi* quantiles over time series returned by q. See this feature request.

  • BUGFIX: rename sign function to sgn in order to be consistent with PromQL. See this pull request from Prometheus.

  • BUGFIX: vmagent: add role: endpointslice in kubernetes_sd_config in order to be consistent with Prometheus. Previously this role was supported with incorrect name: role: endpointslices. Now both endpointslice and endpointslices are supported. See the corresponding code in Prometheus.

  • BUGFIX: improve the detection of the needed free space for background merge operation. This should prevent from possible out of disk space crashes during big merges. See this issue.

  • BUGFIX: vmauth: remove trailing slash from the full url before requesting it from the backend. See this pull request.

  • BUGFIX: vmbackupmanager: fix timeout error when snapshot takes longer than 10 seconds. See this issue.

  • BUGFIX: properly parse OpenTSDB put messages with multiple spaces between message elements. See this issue. Thanks to @envzhu for the fix.

v1.64.1

Released at 19-08-2021

  • FEATURE: add bitmap_and(q, mask), bitmap_or(q, mask) and bitmak_xor(q, mask) functions to MetricsQL. These functions allow performing bitwise operations over data points in time series. See this issue.

  • FEATURE: vmalert: add -remoteWrite.disablePathAppend command-line flag, which can be used when custom -remoteWrite.url must be specified. For example, ./vmalert -disablePathAppend -remoteWrite.url='http://foo.bar/a/b/c?d=e' would write data to http://foo.bar/a/b/c?d=e instead of http://foo.bar/a/b/c?d=e/api/v1/write. See this pull request.

  • FEATURE: vmagent: add -promscrape.noStaleMarkers command-line flag for disabling sending Prometheus stale markers for metrics from disappeared scrape targets. This option may be used for reducing memory usage when scraping big number of metrics with big number of labels and when stale markers aren't needed.

  • FEATURE: vmselect: add -search.noStaleMarkers command-line flag for disabling stale markers handling in queries. This may save some CPU time when the queried data doesn't contain stale markers.

  • BUGFIX: vmagent: stop scrapers for deleted targets before starting scrapers for added targets. This should prevent from possible time series overlap when old targets are substituted by new targets (for example, during new deployment in Kubernetes). The overlap could lead to incorrect query results. See this issue.

  • BUGFIX: vmagent: send Prometheus stale markers for the previously scraped metrics on failed scrapes like Prometheus does. See this article.

  • BUGFIX: upgrade base Docker image from Alpine 3.14.0 to Alpine 3.14.1 . This fixes potential security issues - see Alpine 3.14.1 release notes.

  • BUGFIX: disable overriding the lookbehind window [d] at last_over_time(m[d]) if d is smaller than the interval between samples, since users don't expect implicit overriding of explicitly set [d] in last_over_time(m[d]).

v1.64.0

Released at 15-08-2021

  • FEATURE: add support for Prometheus staleness markers. See this issue.

  • FEATURE: vmagent: automatically generate Prometheus staleness markers for the scraped metrics when scrape targets disappear in the same way as Prometheus does. See this issue.

  • FEATURE: add present_over_time(m[d]) function, which returns 1 if m has a least a single sample over the previous duration d. This function has been added also to Prometheus 2.29.

  • FEATURE: vmagent: support multitenant writes according to these docs. This allows using a single vmagent instance in front of VictoriaMetrics cluster for all the tenants. Thanks to @omarghader for the pull request. See this issue.

  • FEATURE: vmagent: add __meta_ec2_availability_zone_id label to discovered Amazon EC2 targets. This label is available in Prometheus starting from v2.29.

  • FAETURE: vmagent: add __meta_gce_interface_ipv4_<name> labels to discovered GCE targets. These labels are available in Prometheus starting from v2.29.

  • FEATURE: add -search.maxSamplesPerSeries command-line flag for limiting the number of raw samples a single query can process per each time series. This option can protect from out of memory errors when a query processes tens of millions of raw samples per series. See this issue.

  • FEATURE: add -search.maxSamplesPerQuery command-line flag for limiting the number of raw samples a single query can process across all the time series. This option can protect from heavy queries, which select too big number of raw samples. Thanks to @jiangxinlingdu for the initial pull request.

  • FEATURE: improve performance for queries that process big number of time series and/or samples on systems with big number of CPU cores.

  • FEATURE: vmalert: expose vmalert_alerting_rules_last_evaluation_samples and vmalert_recording_rules_last_evaluation_samples metrics. See this issue.

  • FEATURE: vminsert: expose vm_rpc_send_duration_seconds_total counter, which can be used for determining high saturation of every vminsert -> vmstorage link with an alerting query rate(vm_rpc_send_duration_seconds_total) > 0.9s. This query triggers when the link is saturated by more than 90%. This usually means that more vminsert or vmstorage nodes must be added to the cluster in order to increase the total number of vminsert -> vmstorage links.

  • FEATURE: vmagent: expose vmagent_remotewrite_send_duration_seconds_total counter, which can be used for determining high saturation of every connection to remote storage with an alerting query rate(vmagent_remotewrite_send_duration_seconds_total) > 0.9s. This query triggers when a connection is saturated by more than 90%. This usually means that -remoteWrite.queues command-line flag must be increased in order to increase the number of connections per each remote storage.

  • FEATURE: vmui: automatically fill Server URL field. See this issue.

  • BUGFIX: fix corner cases for queries on time ranges exceeding 40 days. Previously some series can be missing in query results. See this issue.

  • BUGFIX: vmselect: return dummy response at /rules page in the same way as for /api/v1/rules page. The /rules page is requested by Grafana 8. See this issue for details.

  • BUGFIX: vmbackup: automatically set default us-east-1 S3 region if it is missing. This should simplify using S3-compatible services such as MinIO for backups. See this issue.

  • BUGFIX: vmselect: prevent from possible deadlock when multiple target query args are passed to Graphite Render API.

  • BUGFIX: return series with a op b labels and N values for (a op b) default N if (a op b) returns series with all NaN values. Previously such series were removed.

  • BUGFIX: vmui: fix layout when the query selects more than 27 time series. See this issue.

  • BUGFIX: vmagent: restore highlighting in red for DOWN targets at /targets page. See this issue.

v1.63.0

Released at 15-07-2021

  • FEATURE: reduce memory usage by up to 30% on production workloads.

  • FEATURE: vmselect: embed vmui into a single-node VictoriaMetrics and into vmselect component of cluster version. See this feature request. The web interface is available at the following paths:

    • /vmui/ for a single-node VictoriaMetrics
    • /select/<accountID>/vmui/ for vmselect at cluster version of VictoriaMetrics
  • FEATURE: support durations anywhere in MetricsQL queries. For example, sum_over_time(m[1h]) / 1h is a valid query, which is equivalent to sum_over_time(m[1h]) / 3600.

  • FEATURE: support durations without suffixes in MetricsQL queries. For example, rate(m[3600]) is a valid query, which is equivalent to rate(m[1h]).

  • FEATURE: export vmselect_request_duration_seconds and vminsert_request_duration_seconds VictoriaMetrics histograms at /metrics page. These histograms can be used for determining latency distribution and SLI/SLO for the served requests. For example, the following query would return the percent of queries that took less than 500ms during the last hour: histogram_share(500ms, sum(rate(vmselect_request_duration_seconds_bucket[1h])) by (vmrange)).

  • FEATURE: vmagent: dynamically reload client TLS certificates from disk on every mTLS connection. This should allow using vmagent with Istio service mesh. See this feature request.

  • FEATURE: log http request path plus all the query args on errors during request processing. Previously only http request path was logged without query args, so it could be hard debugging such errors.

  • FEATURE: add is_set label to flag metrics. This allows determining explicitly set command-line flags with the query flag{is_set="true"}.

  • FEATURE: add ability to remove caches stored inside <-storageDataPath>/cache on startup if reset_cache_on_startup file is present there. See this feature request.

  • BUGFIX: vmagent: remove { %space %} typo in /targets output. The typo has been introduced in v1.62.0. See this issue.

  • BUGFIX: vmagent: fix CSS styles on /targets page. See this issue.

  • BUGFIX: vmalert: accept Prometheus-like durations in interval config option inside group section. See this issue.

  • BUGFIX: properly update vm_merge_need_free_disk_space metric at /metrics page when there is no enough free disk space for performing optimal merges. See this issue.

v1.62.0

Released at 25-06-2021

  • FEATURE: vmagent: add service discovery for Docker (aka docker_sd_config). See this pull request.

  • FEATURE: vmagent: add service discovery for DigitalOcean (aka digitalocean_sd_config). See this feature request.

  • FEATURE: vmagent: change the default value for -remoteWrite.queues from 4 to 2 * numCPUs. This should reduce scrape duration for highly loaded vmagent, which scrapes tens of thousands of targets. See this pull request.

  • FEATURE: vmagent: show the number of samples the target returns during the last scrape on /targets and /api/v1/targets pages. This should simplify debugging targets, which may return too big or too low number of samples. See this feature request.

  • FEATURE: vmagent: show jobs with zero discovered targets on /targets page. This should help debugging improperly configured scrape configs.

  • FEATURE: vmagent: support for http-based service discovery (aka http_sd_config), which has been added since Prometheus 2.28. See this feature request.

  • FEATURE: vmagent: support namespace in Consul serive discovery in the same way as Prometheus 2.28 does. See this issue for details.

  • FEATURE: vmagent: support generic auth configs in consul_sd_configs in the same way as Prometheus 2.28 does. See this issue for details.

  • FEATURE: vmctl: limit the number of samples per each imported JSON line. This should limit the memory usage at VictoriaMetrics side when importing time series with big number of samples.

  • FEATURE: vmselect: log slow queries across all the /api/v1/* handlers (aka Prometheus query API) if their execution duration exceeds -search.logSlowQueryDuration. This should simplify debugging slow requests to such handlers as /api/v1/labels or /api/v1/series additionally to /api/v1/query and /api/v1/query_range, which were logged in the previous releases.

  • FEATURE: vminsert: sort the -storageNode list in order to guarantee the identical series -> vmstorage mapping across all the vminsert nodes. This should reduce resource usage (RAM, CPU and disk IO) at vmstorage nodes if vmstorage addresses are passed in random order to vminsert nodes.

  • FEATURE: vmstorage: reduce memory usage on a system with many CPU cores under high ingestion rate.

  • BUGFIX: prevent from adding new samples to deleted time series after the rotation of the inverted index (the rotation is performed once per -retentionPeriod). See this comment for details.

  • BUGFIX: vmstorage: reduce high disk write IO usage on systems with big number of CPU cores. The issue has been introduced in the release v1.59.0. See this commit and this comment for details.

  • BUGFIX: vmstorage: prevent from incorrect stats collection when multiple concurrent queries execute the same tag filter. This may help reducing CPU usage under certain workloads. See this issue.

  • BUGFIX: vmselect: return the last timestamp for the max / min value from tmax_over_time(m[d]) and tmin_over_time(m[d]) MetricsQL functions as most users expect. See also this issue.

  • BUGFIX: vmselect: return the expected value for increase_pure() MetricsQL function after a gap in a time series. Previously incorrect too big value could be returned after the gap from increase_pure().

v1.61.1

Released at 11-06-2021

  • BUGFIX: vmalert: fix recording rules, which were broken in v1.61.0. See this issue.
  • BUGFIX: reset the on-disk cache for mapping from the full metric name to an internal metric id (e.g. metric_name{labels} -> internal_metric_id) after deleting metrics via delete API. This should prevent from possible inconsistent state after unclean shutdown. This this issue.

v1.61.0

Released at 09-06-2021

  • FEATURE: vmalert: add support for backfilling (aka replay) of recording and alerting rules. See these docs and this feature request.

  • FEATURE: vmalert: add a command-line flag -rule.configCheckInterval for automatic re-reading of -rule files without the need to send SIGHUP signal. See this issue.

  • FEATURE: vmagent: respect the sample_limit and -promscrape.maxScrapeSize values when scraping targets in stream parsing mode. See this pull request.

  • FEATURE: vmauth: add ability to specify mutliple url_prefix entries for balancing the load among multiple vmselect and/or vminsert nodes in a cluster. See these docs.

  • FEATURE: vminsert: add -disableRerouting command-line flag for forcibly disabling the rerouting. This should help resolving this and this issues.

  • FEATURE: vminsert: reduce the probability of global re-routing storm if all the vmstorage nodes cannot keep up with the given ingestion rate for some time. This should improve cluster stability in such cases. See this and this issues.

  • FEATURE: allow building VictoriaMetrics components for Solaris / SmartOS. See this issue.

  • FEATURE: vmagent: add ability to debug relabeling rules. See these docs and this issue.

  • BUGFIX: reduce CPU usage by up to 2x during querying a database with big number of active daily time series. The issue has been introduced in v1.59.0.

  • BUGFIX: vmagent: properly apply auth and tls configs in eureka_sd_configs. See this pull request.

  • BUGFIX: vmauth: do not panic on aborted http requests. See this issue.

  • BUGFIX: properly generate target property for *Series(foo.*.bar) responses returned from Graphite Render API. Previously the target contained the expanded list of series for foo.*.bar, e.g. sumSeries(foo.a.bar,foo.b.bar,...foo.z.bar). Now VictoriaMetrics returns sumSeries(foo.*.bar) as a target in the same way as Graphite does.

v1.60.0

Released at 24-05-2021

  • FEATURE: add ability to limit the number of unique time series, which can be added to storage per hour and per day. This can help dealing with high cardinality and high churn rate issues. See these docs.

  • FEATURE: vmagent: add ability to limit the number of unique time series, which can be sent to remote storage systems per hour and per day. This can help dealing with high cardinality and high churn rate issues. See these docs.

  • FEATURE: vmalert: add ability to run alerting and recording rules for multiple tenants. See this issue and these docs.

  • FEATURE: vminsert: add support for data ingestion via other vminsert nodes. This allows building multi-level data ingestion paths in VictoriaMetrics cluster by writing data from one level of vminsert nodes to another level of vminsert nodes. See these docs and this comment for details.

  • FEATURE: vmagent: reload bearer_token_file, credentials_file and password_file contents every second. This allows dynamically changing the contents of these files during target scraping and service discovery without the need to restart vmagent. See this issue.

  • FEATURE: vmalert: add a flag to control behaviour on startup for state restore errors. Such errors were returned and logged before as well. Now user can specify whether to just log these errors (-remoteRead.ignoreRestoreErrors=true) or to stop the process (-remoteRead.ignoreRestoreErrors=false). The latter is important when VM isn't ready yet to serve queries from vmalert and it needs to wait. See this issue.

  • FEATURE: vmalert: add ability to pass round_digits query arg to datasource via -datasource.roundDigits command-line flag. This can be used for limiting the number of decimal digits after the point in recording rule results. See this issue.

  • FEATURE: return X-Server-Hostname header in http responses of all the VictoriaMetrics components. This should simplify tracing the origin server behind a load balancer or behind auth proxy during troubleshooting.

  • FEATURE: vmselect: allow to use 2x more memory for query processing at vmselect nodes in VictoriaMetrics cluster. This should allow processing heavy queries without the need to increase RAM size at vmselect nodes.

  • FEATURE: add ability to filter /api/v1/status/tsdb output with arbitrary time series selectors passed via match[] query args. See these docs and this issue for details.

  • FEATURE: automatically detect memory and cpu limits for VictoriaMetrics components running under cgroup v2 environments such as HashiCorp Nomad. See this issue.

  • FEATURE: vmauth: allow -auth.config reloading via /-/reload http endpoint. See this issue.

  • FEATURE: add timezone_offset(tz) function. It returns offset in seconds for the given timezone tz relative to UTC. This can be useful when combining with datetime-related functions. For example, day_of_week(time()+timezone_offset("America/Los_Angeles")) would return weekdays for America/Los_Angeles time zone. Special Local time zone can be used for returning an offset for the time zone set on the host where VictoriaMetrics runs. See this issue and MetricsQL docs for more details.

  • FEATURE: vmagent: add support for OAuth2 authorization for scrape targets and service discovery in the same way as Prometheus does. See these docs.

  • FEATURE: vmagent: add support for OAuth2 authorization when writing data to -remoteWrite.url. See -remoteWrite.oauth2.* config params in /path/to/vmagent -help output.

  • FEATURE: vmalert: add ability to set extra_filter_labels at alerting and recording group configs. See these docs.

  • FEATURE: vmstorage: reduce memory usage by up to 30% when ingesting big number of active time series.

  • BUGFIX: vmagent: do not retry scraping targets, which don't support HTTP. This should reduce CPU load and network usage at vmagent and at scrape target. See this issue.

  • BUGFIX: vmagent: fix possible race when refreshing role: endpoints and role: endpointslices scrape targets in kubernetes_sd_config. Prevoiusly pod objects could be updated after the related endpoints object update. This could lead to missing scrape targets. See this issue.

  • BUGFIX: vmagent: properly spread scrape targets among vmagent replicas if -promscrape.cluster.replicationFactor exceeds 1. See this pull request.

  • BUGFIX: vmagent: limit scrape_timeout by scrape_interval. This guarantees that only a single sample is lost during the configured scrape_interval when scrape target responds slowly. See this comment for details.

  • BUGFIX: properly remove stale parts outside the configured retention if -retentionPeriod is smaller than one month. Previously stale parts could remain active for up to a month after they go outside the retention.

  • BUGFIX: stop the process on panic errors, since such errors may leave the process in inconsistent state. Previously panics could be recovered, which could result in unexpected hard-to-debug further behavior of running process.

  • BUGFIX: vminsert, vmagent: make sure data ingestion connections are closed before completing graceful shutdown. Previously the connection may remain open, which could result in trailing samples loss.

  • BUGFIX: vmauth, vmalert: properly re-use HTTP keep-alive connections to backends and datasources. Previously only 2 keep-alive connections per backend could be re-used. Other connections were closed after the first request. See this issue for details.

  • BUGFIX: vmalert: fix false positive error result contains metrics with the same labelset after applying rule labels, which could be triggered when recording rules generate unique metrics. See this issue.

  • BUGFIX: vmctl: properly import InfluxDB rows if they have a field and a tag with identical names. See this issue.

  • BUGFIX: properly reload configs if SIGHUP signal arrives during service initialization. Previously such SIGHUP signal could be ingonred and configs weren't reloaded.

  • BUGFIX: vmalert: properly import default rules from OpenShift. See this issue.

  • BUGFIX: reduce the probability of the removal queue is full panic when highly loaded VictoriaMetrics stores data on NFS. See this issue.

v1.59.0

Released at 01-05-2021

  • FEATURE: improved new time series registration speed on systems with many CPU cores. See this issue. Thanks to @waldoweng for the idea and draft implementation.

  • FEATURE: vmalert: use the same technique as Grafana for determining evaluation timestamps for recording rules. This should make consistent graphs for series generated by recording rules compared to graphs generated for queries from recording rules in Grafana. See this issue.

  • FEATURE: vmauth: add ability to set madatory query args in url_prefix. For example, url_prefix: http://vm:8428/?extra_label=team=dev would add extra_label=team=dev query arg to all the incoming requests. See the example for more details.

  • FEATURE: vmctl: add OpenTSDB migration option. See more details here. Thanks to @johnseekins!

  • FEATURE: log metrics with dropped labels if the number of labels in the ingested metric exceeds -maxLabelsPerTimeseries. This should simplify debugging for this case.

  • FEATURE: vmagent: list user-visible endpoints at http://vmagent:8429/. See this issue.

  • BUGFIX: vmagent: properly update role: endpoints and role: endpointslices scrape targets if the underlying service objects are updated in kubernetes_sd_config. See this issue.

  • BUGFIX: vmagent: apply scrape_timeout on receiving the first response byte from stream_parse: true scrape targets. Previously it was applied to receiving and processing the full response stream. This could result in false timeout errors when scrape target exposes millions of metrics as described here.

  • BUGFIX: vmagent: eliminate possible data race when obtaining value for the metric vm_persistentqueue_bytes_pending. The data race could result in incorrect value for this metric.

  • BUGFIX: vmstorage: remove empty directories on startup. Such directories can be left after unclean shutdown on NFS storage. Previously such directories could lead to crashloop until manually removed. See this issue.

v1.58.0

Released at 08-04-2021

  • FEATURE: vminsert and vmagent: add -sortLabels command-line flag for sorting metric labels before pushing them to vmstorage. This should reduce the size of MetricName -> internal_series_id cache (aka vm_cache_size_bytes{type="storage/tsid"}) when ingesting samples for the same time series with distinct order of labels. For example, foo{k1="v1",k2="v2"} and foo{k2="v2",k1="v1"} represent a single time series. Labels sorting is disabled by default, since the majority of established exporters preserve the order of labels for the exported metrics.

  • FEATURE: allow specifying label value alongside label name for the others sum time series returned from topk_* and bottomk_* functions from MetricsQL. For example, topk_avg(3, max(process_resident_memory_bytes) by (instance), "instance=other_sum") would return top 3 series from max(process_resident_memory_bytes) by (instance) plus a series containing the sum of other series. The others sum series will have {instance="other_sum"} label.

  • FEATURE: do not delete dst_label when applying label_copy(q, "src_label", "dst_label") and label_move(q, "src_label", "dst_label") to series without src_label and with non-empty dst_label. See more details at MetricsQL docs.

  • FEATURE: update Go builder from v1.16.2 to v1.16.3. This should fix these issues.

  • FEATURE: vmagent: add support for follow_redirects option to scrape_configs section in the same way as Prometheus 2.26 does.

  • FEATURE: vmagent: add support for authorization section in -promscrape.config in the same way as Prometheus 2.26 does.

  • FEATURE: vmagent: add support for socks5 proxy in proxy_url config option. See this issue.

  • FEATURE: vmagent: add support for socks5 over tls proxy in proxy_url config option. It can be set up with the following config: proxy_url: "tls+socks5://proxy-addr:port".

  • FEATURE: vmagent: reduce memory usage when -remoteWrite.queues is set to a big value. See this issue.

  • FEATURE: vmagent: add AWS IAM roles for tasks support for EC2 service discovery according to these docs.

  • FEATURE: vmagent: add support for proxy_tls_config, proxy_authorization, proxy_basic_auth, proxy_bearer_token and proxy_bearer_token_file options in consul_sd_config, dockerswarm_sd_config and eureka_sd_config sections.

  • FEATURE: vmagent: pass X-Prometheus-Scrape-Timeout-Seconds header to scrape targets as Prometheus does. In this case scrape targets can limit the time needed for performing the scrape. See this comment for details.

  • FEATURE: vmagent: drop corrupted persistent queue files at -remoteWrite.tmpDataPath instead of throwing a fatal error. Corrupted files can appear after unclean shutdown of vmagent such as OOM kill or hardware reset. See this issue.

  • FEATURE: vmauth: add support for authorization via bearer token. See the docs for details.

  • FEATURE: publish arm64 and amd64 binaries for cluster version of VictoriaMetrics at releases page.

  • BUGFIX: properly handle /api/v1/labels and /api/v1/label/<label_name>/values queries on big start ... end time range. This should fix big resource usage when VictoriaMetrics is queried with Promxy v0.0.62 or newer versions.

  • BUGFIX: do not break sort order for series returned from topk*, bottomk* and outliersk MetricsQL functions. See this issue.

  • BUGFIX: vmagent: properly work with simple HTTP proxies which don't support CONNECT method. For example, PushProx. See this issue.

  • BUGFIX: vmagent: properly discover targets if multiple namespace selectors are put inside kubernetes_sd_config. See this issue.

  • BUGFIX: vmagent: properly discover role: endpoints and role: endpointslices targets in kubernetes_sd_config. See this issue.

  • BUGFIX: properly generate filename for *.tar.gz archive inside _checksums.txt file posted at releases page. See this issue.

v1.57.1

Released at 30-03-2021

  • FEATURE: publish vmutils for GOOS=arm on releases page.

  • BUGFIX: prevent from possible incomplete query results after timed out query.

  • BUGFIX: vmselect: remove -search.storageTimeout command-line flag, since it has the same meaning as -search.maxQueryDuration. See this issue.

  • BUGFIX: vminsert: return back type label to per-tenant metric vm_tenant_inserted_rows_total. See this issue.

v1.57.0

Released at 29-03-2021

  • FEATURE: optimize query performance by up to 10x on systems with many CPU cores. See this tweet.

  • FEATURE: add the following metrics at /metrics page for every VictoraMetrics app:

    • process_resident_memory_anon_bytes - RSS share for memory allocated by the process itself. This share cannot be freed by the OS, so it must be taken into account by OOM killer.
    • process_resident_memory_file_bytes - RSS share for page cache memory (aka memory-mapped files). This share can be freed by the OS at any time, so it must be ignored by OOM killer.
    • process_resident_memory_shared_bytes - RSS share for memory shared with other processes (aka shared memory). This share can be freed by the OS at any time, so it must be ignored by OOM killer.
    • process_resident_memory_peak_bytes - peak RSS usage for the process.
    • process_virtual_memory_peak_bytes - peak virtual memory usage for the process.
  • FEATURE: accept and enforce extra_label=<label_name>=<label_value> query arg at Graphite APIs.

  • FEATURE: use InfluxDB field as metric name if measurement is empty and -influxSkipSingleField command-line is set. See this issue.

  • FEATURE: vmagent: add -promscrape.consul.waitTime command-line flag for tuning the maximum wait time for Consul service discovery. See this issue.

  • FEATURE: vmagent: add vm_promscrape_discovery_kubernetes_stale_resource_versions_total metric for monitoring the frequency of too old resource version errors during Kubernetes service discovery.

  • FEATURE: single-node VictoriaMetrics: log metrics with timestamps older than -search.cacheTimestampOffset compared to the current time. See these docs for details.

  • BUGFIX: prevent from infinite loop on {__graphite__="..."} filters when a metric name contains *, { or [ chars.

  • BUGFIX: prevent from infinite loop in /metrics/find and /metrics/expand Graphite Metrics API handlers when they match metric names or labels with *, { or [ chars.

  • BUGFIX: do not merge duplicate time series during requests to /api/v1/query. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1141

  • BUGFIX: vmagent: properly handle too old resource version error messages from Kubernetes watch API. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1150

  • BUGFIX: vmagent: do not retry sending data blocks if remote storage returns 400 Bad Request error. The number of dropped blocks due to such errors can be monitored with vmagent_remotewrite_packets_dropped_total metrics. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1149

  • BUGFIX: properly calculate summarize and *Series functions in Graphite Render API.

v1.56.0

Released at 17-03-2021

  • FEATURE: add the following functions to MetricsQL:

    • histogram_avg(buckets) - returns the average value for the given buckets.
    • histogram_stdvar(buckets) - returns standard variance for the given buckets.
    • histogram_stddev(buckets) - returns standard deviation for the given buckets.
  • FEATURE: export vm_available_memory_bytes and vm_available_cpu_cores metrics, which show the number of available RAM and available CPU cores for VictoriaMetrics apps.

  • FEATURE: export vm_index_search_duration_seconds histogram, which can be used for troubleshooting time series search performance.

  • FEATURE: vmagent: add ability to replicate scrape targets among vmagent instances in the cluster with -promscrape.cluster.replicationFactor command-line flag. See these docs.

  • FEATURE: vmagent: accept scrape_offset option at scrape_config. This option may be useful when scrapes must start at the specified offset of every scrape interval. See these docs for details.

  • FEATURE: vmagent: support proxy_tls_config, proxy_basic_auth, proxy_bearer_token and proxy_bearer_token_file options at scrape_config section for configuring proxies specified via proxy_url. See these docs.

  • FEATURE: vmauth: allow using regexp paths in url_map. See this issue for details.

  • FEATURE: accept round_digits query arg at /api/v1/query and /api/v1/query_range handlers. This option can be set at Prometheus datasource in Grafana for limiting the number of digits after the decimal point in response values.

  • FEATURE: add -influx.databaseNames command-line flag, which can be used for accepting data from some Telegraf plugins such as fluentd plugin. See this issue.

  • FEATURE: add -logNewSeries command-line flag, which can be used for debugging the source of time series churn rate.

  • FEATURE: publish Windows builds for vmagent, vmalert, vmauth and vmctl at vmutils-windows-*.zip archives at releases page.

  • FEATURE: listen for IPv6 UDP if -enableTCP6 command-line flag is passed to VictoriaMetrics. See this issue.

  • BUGFIX: vmagent: prevent from high CPU usage bug during failing scrapes with small scrape_timeout (less than a few seconds).

  • BUGFIX: vmagent: reduce memory usage when Kubernetes service discovery is used in big number of distinct scrape config jobs by sharing Kubernetes object cache. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1113

  • BUGFIX: vmagent: apply sample_limit only after metric_relabel_configs are applied as Prometheus does. Previously the sample_limit was applied before metrics relabeling.

  • BUGFIX: vmagent: properly apply tls_config, basic_auth and bearer_token to proxy connections if proxy_url option is set. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1116

  • BUGFIX: vmagent: properly scrape targets via https proxy specified in proxy_url if insecure_skip_verify flag isn't set in tls_config section. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1116

  • BUGFUX: avoid duplicate time series error if prometheus_buckets() covers a time range with distinct set of buckets.

  • BUGFIX: prevent exponent overflow when processing extremely small values close to zero such as 2.964393875E-314. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1114

  • BUGFIX: do not include datapoints with a timestamp t-d when returning results from /api/v1/query?query=m[d]&time=t as Prometheus does.

  • BUGFIX: do not crash if a query contains histogram_over_time() function name with uppercase chars. For example, Histogram_Over_Time(m[5m]).

v1.55.1

Released at 03-03-2021

v1.55.0

Released at 02-03-2021

  • FEATURE: add sign(q) and clamp(q, min, max) functions, which are planned to be added in the upcoming Prometheus release . The last_over_time(m[d]) function is already supported in MetricsQL.

  • FEATURE: vmagent: add scrape_align_interval config option, which can be used for aligning scrapes to the beginning of the configured interval. See these docs for details.

  • FEATURE: expose io-related metrics at /metrics page for every VictoriaMetrics component:

    • process_io_read_bytes_total - the number of bytes read via io syscalls such as read and pread
    • process_io_written_bytes_total - the number of bytes written via io syscalls such as write and pwrite
    • process_io_read_syscalls_total - the number of read syscalls such as read and pread
    • process_io_write_syscalls_total - the number of write syscalls such as write and pwrite
    • process_io_storage_read_bytes_total - the number of bytes read from storage layer
    • process_io_storage_written_bytes_total - the number of bytes written to storage layer
  • FEATURE: vmagent: add ability to spread scrape targets among multiple vmagent instances. See these docs and this issue for details.

  • FEATURE: vmagent: use watch API for Kubernetes service discovery. This should reduce load on Kubernetes API server when it tracks big number of objects (for example, 10K pods). This should also reduce the time needed for k8s targets discovery. See this issue for details.

  • FEATURE: vmagent: export vm_promscrape_target_relabel_duration_seconds metric, which can be used for monitoring the time spend on relabeling for discovered targets.

  • FEATURE: vmagent: optimize relabeling performance for common cases.

  • FEATURE: add increase_pure(m[d]) function to MetricsQL. It works the same as increase(m[d]) except of various edge cases. See this issue for details.

  • FEATURE: increase accuracy for buckets_limit(limit, buckets) results for small limit values. See MetricsQL docs for details.

  • FEATURE: vmagent: initial support for Windows build with CGO_ENABLED=0 GOOS=windows go build -mod=vendor ./app/vmagent. See this and this issue.

  • FEATURE: vmagent: support WebIdentityToken auth in EC2 service discovery. See this issue for details.

  • FEATURE: vmalert: properly process query params in -datasource.url and -remoteRead.url command-line flags. See this issue for details.

  • BUGFIX: vmagent: properly apply -remoteWrite.rateLimit when -remoteWrite.queues is greater than 1. Previously there was a data race, which could prevent from proper rate limiting.

  • BUGFIX: vmagent: properly perform graceful shutdown on SIGINT and SIGTERM signals. The graceful shutdown has been broken in v1.54.0. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1065

  • BUGFIX: reduce the probability of duplicate time series errors when querying Kubernetes metrics.

  • BUGFIX: properly calculate histogram_quantile() over time series with only a single non-zero bucket with {le="+Inf"}. Previously NaN was returned, now the value for the last bucket before {le="+Inf"} is returned like Prometheus does.

  • BUGFIX: vmselect: do not cache partial query results on timeout when receiving data from vmstorage nodes. See https://github.com/VictoriaMetrics/VictoriaMetrics/pull/1085

  • BUGFIX: properly handle stale NFS file handle error.

  • BUGFIX: properly cache query results when extra_label query arg is used. Previously the cached results could clash for different extra_label values. See https://github.com/VictoriaMetrics/VictoriaMetrics/pull/1095

  • BUGFIX: fix http: superfluous response.WriteHeader call issue. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1078

  • BUGFIX: fix arm64 builds due to the issue in github.com/golang/snappy. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1074

  • BUGFIX: fix index out of range [1024819115206086200] with length 27 panic, which could occur when 1e-9 value is passed to VictoriaMetrics histogram. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1096

  • BUGFIX: fix parsing for Graphite line with empty tags such as foo; 123 456. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1100

  • BUGFIX: unescape only \\, \n and \" in label names when parsing Prometheus text exposition format as Prometheus does. Previously other escape sequences could be improperly unescaped.

v1.54.1

Released at 18-02-2021

  • BUGFIX: properly handle queries containing a filter on metric name plus any number of negative filters and zero non-negative filters. For example, node_cpu_seconds_total{mode!="idle"}. The bug was introduced in v1.54.0.

v1.54.0

Released at 18-02-2021

  • FEATURE: optimize searching for matching metrics for metric{<label_filters>} queries if <label_filters> contains at least a single filter. For example, the query up{job="foobar"} should find the matching time series much faster than previously.

  • FEATURE: reduce execution times for q1 <binary_op> q2 queries by executing q1 and q2 in parallel.

  • FEATURE: switch from Go1.15 to Go1.16 for building prod binaries.

  • FEATURE: single-node VictoriaMetrics now accepts requests to handlers with /prometheus and /graphite prefixes such as /prometheus/api/v1/query. This improves compatibility with handlers from VictoriaMetrics cluster.

  • FEATURE: expose process_open_fds and process_max_fds metrics. These metrics can be used for alerting when process_open_fds reaches process_max_fds. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/402 and https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1037

  • FEATURE: vmalert: add -datasource.appendTypePrefix command-line option for querying both Prometheus and Graphite datasource in cluster version of VictoriaMetrics. See these docs for details.

  • FEATURE: vmauth: add ability to route requests from a single user to multiple destinations depending on the requested paths. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1064

  • FEATURE: remove dependency on external programs such as cat, grep and cut when detecting cpu and memory limits inside Docker or LXC container.

  • FEATURE: vmagent: add __meta_kubernetes_endpoints_label_*, __meta_kubernetes_endpoints_labelpresent_*, __meta_kubernetes_endpoints_annotation_* and __meta_kubernetes_endpoints_annotationpresent_* labels for role: endpoints in Kubernetes service discovery. These labels where added in Prometheus 2.25.

  • FEATURE: reduce the minimum supported retention period for inverted index (aka indexdb) from one month to one day. This should reduce disk space usage for <-storageDataPath>/indexdb folder if -retentionPeriod is set to values smaller than one month.

  • FEATURE: vmselect: export per-tenant metrics vm_vmselect_http_requests_total and vm_vmselect_http_requests_duration_ms_total . Other per-tenant metrics are available as a part of enterprise package. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/932 for details.

  • BUGFIX: properly convert regexp tag filters containing escaped dots to non-regexp tag filters. For example, {foo=~"bar\.baz"} should be converted to {foo="bar.baz"}. Previously it was incorrectly converted to {foo="bar\.baz"}, which could result in missing time series for this tag filter.

  • BUGFIX: do not spam error logs when discovering Docker Swarm targets without dedicated IP. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1028 .

  • BUGFIX: properly embed timezone data into VictoriaMetrics apps. This should fix -loggerTimezone usage inside Docker containers.

  • BUGFIX: properly build Docker images for non-amd64 architectures (arm, arm64, ppc64le, 386) on Docker hub. Previously these images were incorrectly based on amd64 base image, so they didn't work.

  • BUGFIX: vmagent: return back unsent block to the queue during graceful shutdown. Previously this block could be dropped if remote storage is unavailable during vmagent shutdown. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1065 .

v1.53.1

Released at 03-02-2021

  • BUGFIX: vmselect: fix the bug peventing from proper searching by Graphite filter with wildcards such as {__graphite__="foo.*.bar"}.

v1.53.0

Released at 03-02-2021

  • FEATURE: added vmctl tool to VictoriaMetrics release process. Now it is packaged in vmutils-*.tar.gz archive on the releases page. Source code for vmctl tool has been moved from github.com/VictoriaMetrics/vmctl to github.com/VictoriaMetrics/VictoriaMetrics/app/vmctl.

  • FEATURE: added -loggerTimezone command-line flag for adjusting time zone for timestamps in log messages. By default UTC is used.

  • FEATURE: added -search.maxStepForPointsAdjustment command-line flag, which can be used for disabling adjustment for points returned by /api/v1/query_range handler if such points have timestamps closer than -search.latencyOffset to the current time. Such points may contain incomplete data, so they are substituted by the previous values for step query args smaller than one minute by default.

  • FEATURE: vmselect: added ability to use Graphite-compatible filters in MetricsQL via {__graphite__="foo.*.bar"} syntax. This expression is equivalent to {__name__=~"foo[.][^.]*[.]bar"}, but it works faster and it is easier to use when migrating from Graphite to VictoriaMetrics. This feature deprecates the usage of -search.treatDotsAsIsInRegexps command-line flag.

  • FEATURE: vmselect: added ability to set additional label filters, which must be applied during queries. Such label filters can be set via optional extra_label query arg, which is accepted by querying API handlers. For example, the request to /api/v1/query_range?extra_label=tenant_id=123&query=<query> adds {tenant_id="123"} label filter to the given <query>. It is expected that the extra_label query arg is automatically set by auth proxy sitting in front of VictoriaMetrics. Contact us if you need assistance with such a proxy. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1021 .

  • FEATURE: vmalert: added -datasource.queryStep command-line flag for passing optional step query arg to /api/v1/query endpoint. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1025

  • FEATURE: vmalert: added ability to query Graphite datasource when evaluating alerting and recording rules. See these docs for details.

  • FEATURE: vmagent: added -remoteWrite.roundDigits command-line option for rounding metric values to the given number of decimal digits after the point before sending the metric to the corresponding -remoteWrite.url. This option can be used for improving data compression on the remote storage, because values with lower number of decimal digits can be compressed better than values with bigger number of decimal digits.

  • FEATURE: vmagent: added -remoteWrite.rateLimit command-line flag for limiting data transfer rate to -remoteWrite.url. This may be useful when big amounts of buffered data is sent after temporarily unavailability of the remote storage. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1035

  • FEATURE: vmagent: export the following additional metrics, which may be useful during troubleshooting:

    • vm_promscrape_scrapes_failed_per_url_total
    • vm_promscrape_scrapes_skipped_by_sample_limit_per_url_total
    • vm_promscrape_discovery_requests_total
    • vm_promscrape_discovery_retries_total
    • vm_promscrape_scrape_retries_total
    • vm_promscrape_service_discovery_duration_seconds
  • FEATURE: vmselect: initial implementation for Graphite Render API.

  • BUGFIX: vmagent: reduce HTTP reconnection rate for scrape targets. Previously vmagent could errorneusly close HTTP keep-alive connections more frequently than needed.

  • BUGFIX: vmagent: retry scrape and service discovery requests when the remote server closes HTTP keep-alive connection. Previously disable_keepalive: true option could be used under scrape_configs section when working with such servers.

v1.52.0

Released at 13-01-2021

v1.51.0

Released at 27-12-2020

v1.50.2

Released at 19-12-2020

v1.50.1

Released at 15-12-2020

  • FEATURE: vmagent: export vmagent_remotewrite_blocks_sent_total and vmagent_remotewrite_blocks_sent_total metrics for each -remoteWrite.url.

  • BUGFIX: vmagent: properly delete unregistered scrape targets from /targets and /api/v1/targets pages. They weren't deleted due to the bug in v1.50.0.

v1.50.0

Released at 15-12-2020

  • FEATURE: automatically reset response cache when samples with timestamps older than now - search.cacheTimestampOffset are ingested to VictoriaMetrics. This makes unnecessary disabling response cache during data backfilling or resetting it after backfilling is complete as described in these docs. This feature applies only to single-node VictoriaMetrics. It doesn't apply to cluster version of VictoriaMetrics because vminsert nodes don't know about vmselect nodes where the response cache must be reset.

  • FEATURE: vmalert: add query, first and value functions to alert templates. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/539

  • FEATURE: vmagent: return user-friendly HTML page when requesting /targets page from web browser. The page is returned in the old plaintext format when requesting via curl or similar tool.

  • FEATURE: allow multiple whitespace chars between measurements, fields and timestamp when parsing InfluxDB line protocol. Though InfluxDB line protocol denies multiple whitespace chars between these entities, some apps improperly put multiple whitespace chars. This workaround allows accepting data from such apps.

  • FEATURE: export vm_promscrape_active_scrapers{type="<sd_type>"} metric for tracking the number of active scrapers per each service discovery type.

  • FEATURE: export vm_promscrape_scrapers_started_total{type="<sd_type>"} and vm_promscrape_scrapers_stopped_total{type="<sd_type>"} metrics for tracking churn rate for scrapers per each service discovery type.

  • FEATURE: vmagent: allow setting per--remoteWrite.url command-line flags for -remoteWrite.sendTimeout and -remoteWrite.tlsInsecureSkipVerify.

  • BUGFIX: properly handle * and [...] inside curly braces in query passed to Graphite Metrics API. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/952

  • BUGFIX: vmagent: fix memory leak when big number of targets is discovered via service discovery.

  • BUGFIX: vmagent: properly pass datacenter filter to Consul API server. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/574#issuecomment-740454170

  • BUGFIX: properly handle CPU limits set on the host system or host container. The bugfix may result in lower memory usage on systems with CPU limits. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/946

  • BUGFIX: prevent from duplicate name tag returned from /tags/autoComplete/tags handler. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/942

  • BUGFIX: do not enable strict parsing for -promscrape.config if -promscrape.config.dryRun comand-line flag is set. Strict parsing can be enabled with -promscrape.config.strictParse command-line flag. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/944

  • BUGFIX: vminsert: properly update vm_rpc_rerouted_rows_processed_total metric. Previously it wasn't updated. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/955

  • BUGFIX: vmagent: properly recover when opening incorrectly stored persistent queue. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/964

  • BUGFIX: vmagent: properly handle scrape errors when stream parsing is enabled with -promscrape.streamParse command-line flag or with stream_parse: true per-target config option. Previously such errors weren't reported at /targets page. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/967

  • BUGFIX: assume the previous value is 0 when calculating increase() for the first point on the graph if its value doesn't exceed 100 and the delta between two first points equals to 0. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/962

v1.49.0

Released at 05-12-2020

  • FEATURE: optimize Consul service discovery speed when discovering big number of services. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/574

  • FEATURE: add label_uppercase(q, label1, ... labelN) and label_lowercase(q, label1, ... labelN) function to MetricsQL for uppercasing and lowercasing values for the given labels. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/936

  • FEATURE: add count_eq_over_time(m[d], N) and count_ne_over_time(m[d], N) for counting the number of samples for m over d that (equal / not equal) to N.

  • FEATURE: do not print usage info for all the command-line flags when incorrect command-line flag is passed. Previously it could be hard reading the error message about incorrect command-line flag because of too big usage info for all the flags.

  • FEATURE: upgrade Go builder from v1.15.5 to v1.15.6 . This fixes issues found in Go since v1.15.5.

  • BUGFIX: properly parse timestamps in OpenMetrics format - they are exposed as floating-point number in seconds instead of integer milliseconds unlike in Prometheus exposition format. See the docs.

  • BUGFIX: return nan for a >bool b query when a equals to nan like Prometheus does. Previously 0 was returned in this case. This applies to any comparison operation with bool modifier. See these docs for details.

  • BUGFIX: properly parse hex numbers in MetricsQL. Previously hex numbers with non-decimal digits such as 0x3b couldn't be parsed.

  • BUGFIX: handle time() cmp_op metric like Prometheus does - i.e. return metric value if cmp_op comparison is true. Previously time() value was returned.

  • BUGFIX: return nan for minute(m) query when m equals to nan like Prometheus does. This applies to all the time-related functions such as day_of_month, day_of_week, days_in_month, hour, month and year.

v1.48.0

Released at 26-11-2020

v1.47.0

Released at 16-11-2020

v1.46.0

Released at 07-11-2020

v1.45.0

Released at 02-11-2020

v1.44.0

Released at 13-10-2020

  • FEATURE: automatically add missing label filters to binary operands as described at https://utcc.utoronto.ca/~cks/space/blog/sysadmin/PrometheusLabelNonOptimization . This should improve performance for queries with missing label filters in binary operands. For example, the following query should work faster now, because it shouldn't fetch and discard time series for node_filesystem_files_free metric without matching labels for the left side of the expression:

       node_filesystem_files{ host="$host", mountpoint="/" } - node_filesystem_files_free
    
  • FEATURE: vmagent: add Docker Swarm service discovery (aka dockerswarm_sd_config). See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/656

  • FEATURE: add ability to export data in CSV format. See these docs for details.

  • FEATURE: vmagent: add -promscrape.suppressDuplicateScrapeTargetErrors command-line flag for suppressing duplicate scrape target errors. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/651 and https://docs.victoriametrics.com/vmagent.html#troubleshooting .

  • FEATURE: vmagent: show original labels before relabeling is applied on duplicate scrape target errors. This should simplify debugging for incorrect relabeling. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/651

  • FEATURE: vmagent: /targets page now accepts optional show_original_labels=1 query arg for displaying original labels for each target before relabeling is applied. This should simplify debugging for target relabeling configs. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/651

  • FEATURE: add -finalMergeDelay command-line flag for configuring the delay before final merge for per-month partitions. The final merge is started after no new data is ingested into per-month partition during -finalMergeDelay.

  • FEATURE: add vm_rows_added_to_storage_total metric, which shows the total number of rows added to storage since app start. The sum(rate(vm_rows_added_to_storage_total)) can be smaller than sum(rate(vm_rows_inserted_total)) if certain metrics are dropped due to relabeling. The sum(rate(vm_rows_added_to_storage_total)) can be bigger than sum(rate(vm_rows_inserted_total)) if replication is enabled.

  • FEATURE: keep metric name after applying MetricsQL functions, which don't change time series meaning. The list of such functions:

    • keep_last_value
    • keep_next_value
    • interpolate
    • running_min
    • running_max
    • running_avg
    • range_min
    • range_max
    • range_avg
    • range_first
    • range_last
    • range_quantile
    • smooth_exponential
    • ceil
    • floor
    • round
    • clamp_min
    • clamp_max
    • max_over_time
    • min_over_time
    • avg_over_time
    • quantile_over_time
    • mode_over_time
    • geomean_over_time
    • holt_winters
    • predict_linear See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/674
  • BUGFIX: properly handle stale time series after K8S deployment. Previously such time series could be double-counted. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/748

  • BUGFIX: return a single time series at max from absent() function like Prometheus does.

  • BUGFIX: vmalert: accept days, weeks and years in for: part of config like Prometheus does. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/817

  • BUGFIX: fix mode_over_time(m[d]) calculations. Previously the function could return incorrect results.

v1.43.0

Released at 06-10-2020

v1.42.0

Released at 30-09-2020

Previous releases

See releases page.