Commit Graph

22 Commits

Author SHA1 Message Date
Hui Wang
1908d44cf8
dashboards: fix description about pending datapoints (#7235)
Some checks failed
build / Build (push) Has been cancelled
CodeQL Go / Analyze (push) Has been cancelled
CodeQL JS/TS / Analyze (push) Has been cancelled
main / lint (push) Has been cancelled
main / test (test-full) (push) Has been cancelled
main / test (test-full-386) (push) Has been cancelled
main / test (test-pure) (push) Has been cancelled
See [our
playground](https://play-grafana.victoriametrics.com/d/oS7Bi_0Wz_vm/victoriametrics-cluster-vm?orgId=1&var-ds=P996FABE17B5F6D1E&var-job=All&var-job_insert=All&var-job_select=All&var-job_storage=All&var-instance=All)
for reference.

(cherry picked from commit d3f110373c)
2024-10-11 14:28:23 +02:00
Roman Khavronenko
deb2f87074
deployment: add panel and alerts for displying go scheduler latency (#7078)
The panel and alerting rule should help to understand whether VM
component doesn't have enough CPU resources or gets throttled. The alert
is applicable for all VM components.
The panel was added to vmalert, vmagent, vmsingle, vm clusert and
victorialogs dashes.

-------------------

This alerting rule should have help us identify resource shortage for
sandbox vmagent - see [this
link](https://play.victoriametrics.com/select/accounting/1/6a716b0f-38bc-4856-90ce-448fd713e3fe/prometheus/graph/#/?g0.range_input=23d13h25m25s424ms&g0.end_input=2024-09-23T14%3A11%3A00&g0.relative_time=none&g0.tab=0&g0.expr=histogram_quantile%280.99%2C+sum%28rate%28go_sched_latencies_seconds_bucket%7Bjob%3D%22vmagent-monitoring-vmagent%22%7D%5B5m%5D%29%29+by+%28le%2C+job%2C+instance%29%29+%3E+0.1)
for example. We weren't aware of resource shortage, because VM metrics
assumed this vmagent had 1vCPU while in fact its limit was 0.2vCPU.

Signed-off-by: hagen1778 <roman@victoriametrics.com>
(cherry picked from commit 4d0b41e63b)
2024-09-24 16:58:14 +02:00
hagen1778
89819f2054
dashboards: use $__interval variable for offsets and look-behind windows in annotations
This should improve precision of `restarts` and `version change` annotations when
 zooming-in/zooming-out on the dashboards.

 The change also makes `restarts` dashboard visible on the panels, so user can disable it from
 displaying if needed. This could be useful when restarts overlap with version change events.

Signed-off-by: hagen1778 <roman@victoriametrics.com>
(cherry picked from commit 9dd9b4442f)
2024-05-22 16:40:08 +02:00
hagen1778
0f72ab8ef6
deployment: bump Grafana version to 10.4.2
Signed-off-by: hagen1778 <roman@victoriametrics.com>
(cherry picked from commit 9256df17fa)
2024-04-30 10:30:00 +02:00
hagen1778
8f48747802
dashboards: add Network Usage panel to Resource Usage row
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/4478
Signed-off-by: hagen1778 <roman@victoriametrics.com>
(cherry picked from commit 8606b48ce5)
2024-04-30 10:29:58 +02:00
Dima Lazerka
d821e13c24
deployment/dashboards: properly show version for non-stable docker images (#6150)
re: .*-(?:tags|heads)-(.*)-(?:0|dirty)-.*

cases:
victoria-metrics-20240419-160209-heads-enterprise-single-node-0-g08f933ab0c
enterprise-single-node

victoria-metrics-20240201-133950-tags-v1.97.1-enterprise-0-g760a8733b
v1.97.1-enterprise

victoria-metrics-20240419-160209-heads-rotation-part-2-0-ge2367b6d1-dirty-848b54cd
rotation-part-2-0-ge2367b6d1

victoria-metrics-20240419-160209-heads-lts-1.93-enterprise-search-contention-0-g30ef4aad21-amd64
lts-1.93-enterprise-search-contention

victoria-metrics-20240425-150852-tags-v1.101.0-enterprise-0-g718138c64
v1.101.0-enterprise

Signed-off-by: hagen1778 <roman@victoriametrics.com>
Co-authored-by: Dzmitry Lazerka <dlazerka@gmail.com>
Co-authored-by: hagen1778 <roman@victoriametrics.com>
(cherry picked from commit 564463259a)
2024-04-30 10:29:56 +02:00
Zakhar Bessarab
88677f179c
dashboards/victoria-metrics-single: allow selecting multiple instance values (#5870)
Allowing to select multiple instance IPs makes it much easier to view
metrics for longer periods of time in dynamic environments such as
Kubernetes. In k8s update will also cause IP to change making it harder
to use dashboard to check the status.

See: https://github.com/VictoriaMetrics/VictoriaMetrics/issues/5869

---------

Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>
Signed-off-by: hagen1778 <roman@victoriametrics.com>
Co-authored-by: hagen1778 <roman@victoriametrics.com>
(cherry picked from commit 6b493582da)
2024-04-30 10:29:54 +02:00
Aliaksandr Valialkin
64938732e3
all: replace old https://docs.victoriametrics.com/MetricsQL.html url with the new one - https://docs.victoriametrics.com/metricsql/ 2024-04-18 02:15:33 +02:00
hagen1778
51745ec5ff
dashboards: update links in various panels
* use docs.victoriametrics.com instead of github docs
* add links to common terms used in VictoriaMetrics

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-03-04 17:00:54 +02:00
hagen1778
9b173c2f01
dashboards: follow-up 4369bc1df2
* add more details to changelog
* simplify panels description
* remove capacity planning recommendation, as it proves it incompetent

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-02-08 12:55:42 +02:00
Hui Wang
0cd0ddc1c1
deployment/dashboards: fix Storage full ETA panels (#5747)
During background downsampling, rate(vm_deduplicated_samples_total{type="merge"}) could be much bigger than 
rate(vm_rows_added_to_storage_total) and it could last quite some time,
 which causes negative values of Storage full ETA and confuses users, see playground.

Instead of trying to get more accurate results during downsampling, I think it's ok to ignore 
vm_deduplicated_samples_total at all, it's more reasonable to see Storage full ETA increase after downsampling.

---------

Co-authored-by: Aliaksandr Valialkin <valyala@victoriametrics.com>
2024-02-08 12:54:31 +02:00
hagen1778
bdbab7bed5
dashboards/all: add new panel CPU spent on GC
It should help identifying cases when too much CPU is spent on garbage collection,
 and advice users on how this can be addressed.

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-02-05 11:42:28 +02:00
hagen1778
3dab94a6c1
dashboards: update to grafana/grafana:10.3.1
Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-02-05 10:50:36 +02:00
hagen1778
151247c9b9
dashboards/single: fix typo in query for version annotation
The typo falsely produced many version change events.

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-01-31 10:28:24 +02:00
hagen1778
5aa0f77d8c
dashboards: specify where to see details about dropped labels
Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-01-29 17:23:38 +01:00
hagen1778
6ef6b83b33
dashboards: reflect dashboard rename in copy script
This is a follow-up for ff33e60a3d

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-01-22 18:44:18 +02:00
hagen1778
a2d3fce05f
deployment/dashboards: change title VictoriaMetrics to VictoriaMetrics - single-node
The new title should provide better understanding of this dashboard purpose.

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-01-17 01:06:07 +02:00
Aliaksandr Valialkin
5c43f2261e
dashboards: remove path!="/favicon.ico" filter from requests rate graphs
The `path!="/favicon.ico"` filter has little sense, since there are many other special paths,
which may be filtered out - /metrics, /flags, /health, /ping, /robots.txt, /-/healthy, /-/ready, /reload, etc.
See /lib/httpserver/httpserver.go for more details.
It will be hard or impossible to maintain filters for all these paths, so it is better to drop this filter
in order to simplify queries and improve the consistency of these queries.
2023-11-16 19:29:46 +01:00
hagen1778
7d72474a38
dashboards: use version instead of short_version in annotations
`version` label won't show the difference if various flavors of the same
version were deployed. But `short_version` will.

For example, on the sandbox env we test VM builds before new version release.
Without this change, the version update won't be visible on dashboard.

Signed-off-by: hagen1778 <roman@victoriametrics.com>
(cherry picked from commit d389a4fcf3)
2023-11-16 09:27:42 +01:00
hagen1778
72a40539b0
dashboards: update description for RSS and anonymous memory panels to be consistent for single-node, cluster and vmagent dashboards.
Signed-off-by: hagen1778 <roman@victoriametrics.com>
(cherry picked from commit d3ae2b2f62)
2023-11-14 10:00:11 +01:00
hagen1778
f2195cb914
dashboards/victoriametrics: account for instance filter in annotations
Signed-off-by: hagen1778 <roman@victoriametrics.com>
2023-09-21 09:36:35 +02:00
Roman Khavronenko
c81e90223c
dashboards: provide copies of Grafana dashboards alternated with Vict… (#4905)
dashboards: provide copies of Grafana dashboards alternated with VictoriaMetrics datasource

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2023-08-29 11:20:16 +02:00