New Features and Improvements in VictoriaMetrics - Q3 2025

VictoriaMetrics 0 views 29 slides Oct 13, 2025
Slide 1
Slide 1 of 29
Slide 1
1
Slide 2
2
Slide 3
3
Slide 4
4
Slide 5
5
Slide 6
6
Slide 7
7
Slide 8
8
Slide 9
9
Slide 10
10
Slide 11
11
Slide 12
12
Slide 13
13
Slide 14
14
Slide 15
15
Slide 16
16
Slide 17
17
Slide 18
18
Slide 19
19
Slide 20
20
Slide 21
21
Slide 22
22
Slide 23
23
Slide 24
24
Slide 25
25
Slide 26
26
Slide 27
27
Slide 28
28
Slide 29
29

About This Presentation

New 7 releases in Q3
Search in documentation - now with AI
vmui - new alerting page
Dashboards: alert statistics -available at https://grafana.com/orgs/victoriametrics/dashboards
Alerting best practices - read at https://victoriametrics.com/blog/alerting-best-practices
Metrics usage tracker enabled...


Slide Content

Simple, Reliable, Efficient Monitoring victoriametrics.com

What's new in VictoriaMetrics
Roman Khavronenko
Q3 2025

LTS:
●v1.110 - v1.110.19
●v1.122 - v1.122.4 - new LTS line
●v1.102 - deprecated
New 7 releases in Q3
Upstream:
●v1.120.0 - v1.126.0

Search in documentation

Search in documentation

Search in documentation - now with AI! ✨??????????????????????????????
Try it now at https://docs.victoriametrics.com/

Search in documentation
Thanks to @makasim, @andrewchubatiuk, @tenmozes and @kacperwalter!

vmui - new alerting page, v1.125.0

vmui - new alerting page

vmui - new alerting page
Thanks to @andrewchubatiuk, @haleygo, @arturminchukov and @loori-r!
Try it now at https://play.victoriametrics.com/

Dashboards: alert statistics
Available at https://grafana.com/orgs/victoriametrics/dashboards

Dashboards: alert statistics
Try it now at https://play-grafana.victoriametrics.com/

Alerting best practices
Read at https://victoriametrics.com/blog/alerting-best-practices

Metrics usage tracker enabled by default, v1.122.0

Log new series toggle, v1.123.0
1.Call /internal/log_new_series?seconds=60 on vmstorage or vmsingle
2.Check the logs:
2025-10-02T08:40:38 info enabling logging of new series for the next 1m0s. This may increase resource usage during this period.
2025-10-02T08:40:39 info new series created:
vminsert_request_duration_seconds_bucket{job="victoria-metrics",instance="self",vmrange="1.136e-06...1.292e-06"}
2025-10-02T08:40:39 info new series created: vminsert_request_duration_seconds_sum{job="victoria-metrics",instance="self"}
2025-10-02T08:40:39 info new series created: vminsert_request_duration_seconds_count{job="victoria-metrics",instance="self"}

3. Logging disables automatically in `?seconds=<duration>`

Log new series toggle, v1.123.0

Thanks to @leiwingqueen!

Usability improvements
vmsingle and vmstorage in VictoriaMetrics cluster : The -snapshotsMaxAge flag default
have been changed to 3d.

This enables automatic deletion of snapshots older than 3 days.

If you want to keep the previous behavior (never automatically deleting snapshots), please set
-snapshotsMaxAge=0.

See #9344 for the details.

Usability improvements: share queries for debugging

Usability improvements: export raw data

Usability improvements: export raw data
●Export in Json or CSV format
●Raw data is essential for troubleshooting
●https://docs.victoriametrics.com/victoriametrics/troubl
eshooting/#unexpected-query-results
●Raw data can be easily imported into VM

Usability improvements: labels limit per target
vmagent and Single-node VictoriaMetrics : support label_limit scrape parameter.
It sets labels limit for each scraped sample from a target. If limit is exceeded, the scrape will be
rejected and target marked with corresponding scrape error.
The label_limit can be defined on global level, per job or during relabeling via
__label_limit__ label.
See #7660 and #3233 issues.

Performance optimizations
FEATURE: improve performance on systems with many CPU cores by removing the top sources of false sharing at global variables. See
#8682 . Thanks to @tIGO for raising this issue and for the initial attempt to fix it at the PR #8683 .
FEATURE: vmselect in VictoriaMetrics cluster : dynamically adjusts the concurrent dial limit between 8 and 64 based on
-search.maxConcurrentRequests . Additionally, goroutines now have the opportunity to access available connections while
awaiting the dial limit token. This enables faster connection establishment when sudden requests arrive and reduces the blocking time
during the availability check for connections. See #8922
FEATURE: vmsingle and vmstorage in VictoriaMetrics cluster : reduce CPU usage caused by garbage collection during indexDB
lookups on cache misses. See #9324 for details.
FEATURE: vmagent and vmsingle : improve Kubernetes service discovery performance with large amount of configured role selectors.
See this #9354 issue for details. Thanks to the @fxrlv
FEATURE: vmsingle and vmstorage in VictoriaMetrics cluster : optimize /api/v1/labels and /api/v1/label/TAG/values
requests with a single match or extra_filters filter containing time series metric name. See this PR #9489 for details.

One more thing…

One more thing…

VictoriaTraces - database for storing traces

●Accepts trace spans in OpenTelemetry protocol
●Supports Jaeger Query Service JSON APIs:
○Integrates with Grafana via Jaeger datasource
○Integrates with Jaeger UI
●Uses up to 3.7x less RAM and up to 2.6x less CPU than other solutions such as
Grafana Tempo
●Supports alerting via vmalert

VictoriaTraces - database for storing traces

Developer's Note: Research on Distributed Tracing,
Comparing With Tempo and ClickHouse


VictoriaTraces - database for storing traces

VictoriaTraces - database for storing traces

Opentelemetry-demo: logs, metrics and traces with VM

Useful Links slide
●https://docs.victoriametrics.com/changelog/
●https://github.com/VictoriaMetrics/operator/releases
●https://github.com/VictoriaMetrics/helm-charts/releases
●https://github.com/VictoriaMetrics/victorialogs-datasource/releases
●https://github.com/VictoriaMetrics/victoriametrics-datasource/releases
●https://play-grafana.victoriametrics.com/
●https://play.victoriametrics.com/
Simple, Reliable, Efficient Monitoring victoriametrics.com