Commit graph

8939 commits

Author SHA1 Message Date
hagen1778
5334f0c2ce
docs/CHANGELOG.md: cut v1.101.0
Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-04-25 16:25:09 +02:00
hagen1778
7fd9325e62
app/vmselect: run make vmui-update
Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-04-25 15:51:03 +02:00
hagen1778
a62551f773
docs: add change for c0e4ccb7b5
Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-04-25 15:37:37 +02:00
Hui Wang
dd0d2c77c8
app/vmselect: implement cmd-line flags -search.disableImplicitConversions and -search.logImplicitConversions (#6180)
address https://github.com/VictoriaMetrics/VictoriaMetrics/issues/4338
support disable or log [implicit
conversions](https://docs.victoriametrics.com/metricsql/#implicit-query-conversions)
for subquery with cmd-line flags `-search.disableImplicitConversion` and
`-search.logImplicitConversion`

Signed-off-by: hagen1778 <roman@victoriametrics.com>
Co-authored-by: hagen1778 <roman@victoriametrics.com>
2024-04-25 12:54:42 +02:00
Yury Molodov
57b7d16259
vmui: improve error message for server response issues (#6177)
Updates error messages for better clarity and guidance on server
response issues.
2024-04-25 12:52:13 +02:00
Yury Molodov
6193fa3dcf
vmui: trigger auto-suggestion at any cursor position (#6155)
- Implemented auto-suggestion triggers for mid-string cursor positions
in vmui.
- Improved the suggestion list positioning to appear directly beneath
the active text editing area.

https://github.com/VictoriaMetrics/VictoriaMetrics/issues/5864
2024-04-25 12:48:49 +02:00
hagen1778
6aaf1768f4
Revert "docs: removed code-style highlighting for commanad-line flags of VM components (#6147)"
This reverts commit 9bedbcfa2f.
2024-04-25 11:23:02 +02:00
Andrii Chubatiuk
07496d7d92
deployment: update makefile package-* targets (#6172)
Updated package targets in a same manner, how it's done for publish ones
in
7958f38864
2024-04-24 18:29:32 +02:00
hagen1778
2d9bbe1934
docs/changelog: mention downsampling fixes for ENT version of VM
Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-04-24 17:03:21 +02:00
hagen1778
679844feaf
Revert "app/vmbackup: introduce new flag type URL (#6152)"
This reverts commit 029060af60.
2024-04-24 13:47:57 +02:00
Roman Khavronenko
029060af60
app/vmbackup: introduce new flag type URL (#6152)
The new flag type is supposed to be used for specifying URL values which
could contain sensitive information such as auth tokens in GET params or
HTTP basic authentication.

The URL flag also allows loading its value from files if `file://`
prefix is specified. As example, the new flag type was used in
app/vmbackup as it requires specifying `authKey` param for making the
snapshot.

See related issue
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/5973

Thanks to @wasim-nihal for initial implementation
https://github.com/VictoriaMetrics/VictoriaMetrics/pull/6060

---------

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-04-24 10:57:54 +02:00
Github Actions
a14acd4fcc
Automatic update operator docs from VictoriaMetrics/operator@6e1a876 (#6179) 2024-04-24 00:41:23 -07:00
Artem Navoiev
592a9fe9f2
docs: vmalert remove new lines in configuration section
Signed-off-by: Artem Navoiev <tenmozes@gmail.com>
2024-04-23 22:30:31 +02:00
Artem Navoiev
aeba3f39db
docs: vmalert remove new lines in configuration section
Signed-off-by: Artem Navoiev <tenmozes@gmail.com>
2024-04-23 21:47:17 +02:00
hagen1778
035de57e5e
dashboards: show max number of active merges instead of cumulative
The cumulative number of active merges could be red herring
as it its value depends on the number of vmstorages.
For example, vmstorage could be added or removed and this will affect
the panel.
Or, each vmstorage could start a merging process (i.e. for downsampling)
and visiually it could look like a massive change.

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-04-23 16:41:48 +02:00
hagen1778
22cfab1ea2
docs/changelog: mention icnreased CPU usage in 1.100.1 release for ENT distributions
Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-04-23 16:34:35 +02:00
hagen1778
4251292708
app/vmagent: mention corner case with dangling queues and identical URLs
See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/6140

We don't cover this corner case as it has low chance for reproduction.
Precisely, the requirements are following:
1. vmagent need to be configured with multiple identical `remoteWrite.url` flags;
2. At least one of the persistent queues need to be non-empty, which already
signalizes about issues with setup;
3. vmagent need to be restarted with removing of one of `remoteWrite.url` flags.

We do not document this case in vmagent.md as it seems to be a rare corner case
and its explanation will require too much of explanation and confuse users.

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-04-23 14:49:45 +02:00
Github Actions
301bd387d4
Automatic update operator docs from VictoriaMetrics/operator@e343a8e (#6168) 2024-04-23 12:12:35 +02:00
Andrii Chubatiuk
7958f38864
deployment/docker: allow cross-platform building on arm64 platform (#6158)
Added x86_64 libraries to allow building cross-platform images on arm64
2024-04-23 12:12:10 +02:00
Denys Holius
bde4693a90
deployment/docker/victorialogs/fluentbit-docker/docker-compose.yml: update fluentbit version from v2.1.4 to v3.0.2 (#6120)
see also https://fluentbit.io/announcements/v3.0.0/
2024-04-23 01:07:45 -07:00
Roman Khavronenko
5f487c7090
app/vmalert: fix links with anchors in vmalert's UI (#6146)
Starting from v1.99.0 vmalert could ignore anchors pointing to specific
rule groups if `search` param was present in URL.
This change makes anchors compatible with `search` param in UI.

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-04-22 15:02:10 +02:00
Github Actions
6dbb4c1671
Automatic update operator docs from VictoriaMetrics/operator@7e2ba6f (#6165) 2024-04-22 14:55:41 +02:00
Denys Holius
9bedbcfa2f
docs: removed code-style highlighting for commanad-line flags of VM components (#6147)
Using `sh` or `console` formatting doesn't do word-breaking on render. This makes flags description
harder to read, as users need to scroll the web page horizontally.
Removing the formatting renders the description with normal word-breaking.
2024-04-22 14:53:26 +02:00
hagen1778
bae3874e6a
app/streamaggr: follow-up after c0e4ccb7b5
* rm vmagent mentions from vminsert flags
* improve documentation wording, add links to related sections
* mention `ignore_first_intervals` in the stream aggr options
* update flags description
* add basic test for config parsing validation

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-04-22 14:22:59 +02:00
Andrii Chubatiuk
c0e4ccb7b5
lib/streamaggr: add option to ignore first N aggregation intervals (#6137)
Stream aggregation may yield inaccurate results if it processes incomplete data. 
This issue can arise when data is sourced from clients that maintain a queue of unsent data, such as Prometheus or vmagent.
 If the queue isn't fully cleared within the aggregation interval, only a portion of the time series may be included in that period, leading to distorted calculations. 
To mitigate this we add an option to ignore first N aggregation intervals. It is expected, that client queues
will be cleared during the time while aggregation ignores first N intervals and all subsequent aggregations
will be correct.
2024-04-22 13:52:04 +02:00
Aliaksandr Valialkin
81b2fb88cc
docs/MetricsQL.md: add links to raw samples chapter at https://docs.victoriametrics.com/keyconcepts/ 2024-04-21 00:21:51 +02:00
Aliaksandr Valialkin
3f8f7f2c41
docs/CONTRIBUTING.md: small re-wording 2024-04-21 00:07:44 +02:00
Aliaksandr Valialkin
0e75581dd9
docs: remove duplicate information regarding contribution guidelines to VictoriaMetrics
Substitute duplicate information with the link to https://docs.victoriametrics.com/contributing/ across all the docs.
2024-04-21 00:00:50 +02:00
Aliaksandr Valialkin
bbdd7743a8
docs/CONTRIBUTING.md: more formatting fixes and clarifications 2024-04-20 23:48:56 +02:00
Aliaksandr Valialkin
c52f73b3e1
.github: move pull_request_template.md from .github/PULL_REQUEST_TEMPLATE/ to .github/
I couldn't figure out how to show the provided pull request template to users
according to https://docs.github.com/en/communities/using-templates-to-encourage-useful-issues-and-pull-requests/creating-a-pull-request-template-for-your-repository ,
I let's try 'global pull request for the whole repository' approach.
2024-04-20 23:42:57 +02:00
Aliaksandr Valialkin
d18968f8cb
docs/CONTRIBUTING.md: small corrections after a402847eb6 2024-04-20 23:21:54 +02:00
Aliaksandr Valialkin
a402847eb6
Move CONTRIBUTING.md to docs/
It is better from visibility PoV if the CONTRIBUTING.md file is visible at https://docs.victoriametrics.com/contributing/ .
This page can be indexed by search engines and searched by our users later.
It is also easier to provide a link to this page now comparing to the old link, which is much harder to remember:
https://github.com/VictoriaMetrics/VictoriaMetrics/blob/master/CONTRIBUTING.md

While at it, syncrhonize docs/CONTRIBUTING.md with .github/PULL_REQUEST_TEMPLATE/pull_request_template.md ,
so they do not contain duplicate information, which can be outdated over time. Now all the relevant information
is located at docs/CONTRIBUTING.md, while .github/PULL_REQUEST_TEMPLATE/pull_request_template.md refers this doc.

This is a follow-up for c006db1798

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/pull/6040
2024-04-20 23:11:22 +02:00
Aliaksandr Valialkin
4770294732
lib/protoparser: substitute hybrid channel-based pools with plain sync.Pool
Using plain sync.Pool simplifies the code without increasing memory usage and CPU usage.
So it is better to use plain sync.Pool from readability and maintainability PoV.

This is a follow-up for 8942f290eb
2024-04-20 21:59:51 +02:00
Aliaksandr Valialkin
8942f290eb
app/vminsert: replace hybrid sync.Pool+channel-based pool scheme for poolCtx with plain sync.Pool
This simplifies the code, while doesn't increase memory usage under low and high data ingestion rate.

This is a follow-up for 1decbcf6eb
2024-04-20 21:44:53 +02:00
Aliaksandr Valialkin
1decbcf6eb
app/vminsert/influx: replace hybrid channel-based pool+sync.Pool with plain sync.Pool for pushCtx
The memory usage for plain sync.Pool doesn't increase comparing to the memory usage for the hybrid scheme,
so it is better to use plain sync.Pool in order to simplify the code and make it more readable and maintainable.

This is a follow-up for c22da2f917
2024-04-20 21:41:06 +02:00
Aliaksandr Valialkin
c22da2f917
app/vmagent/influx: replace hybrid channel-based pool + sync.Pool with plain sync.Pool for pushCtx
Data ingestion benchmark doesn't show memory usage difference between two approaches,
so let's use simpler approach in order to improve code readability and maintainability.

This is a follow-up for 77c597738c
2024-04-20 21:38:11 +02:00
Aliaksandr Valialkin
77c597738c
app/vmagent/common: use plain sync.Pool instead of a mix of sync.Pool with channel-based pool for PushCtx
This scheme was used for reducing memory usage when vmagent runs on a machine with big number of CPU cores
and the ingestion rate isn't too big. The scheme with channel-based pool could reduce memory usage,
since it minimizes the number of PushCtx structs in the pool in this case.

Performance tests didn't reveal significant difference in memory usage under both low and high ingestion rate
between plain sync.Pool and the current hybrid scheme, so replace the scheme with plain sync.Pool in order
to simplify the code.
2024-04-20 21:27:05 +02:00
Aliaksandr Valialkin
7531e9084a
all: use clear() built-in Go function for clearing []prompbmarshal.TimeSeries and []prompbmarshal.Label slices
This makes the code a bit clear.
2024-04-20 21:00:03 +02:00
Aliaksandr Valialkin
3e728c41f6
app/vminsert/common: remove obsolete optimization for reducing memory usage for InsertCtx pool
This optimization is no longer needed according to benchmarks with ingestion rate.

This simplifies the code a bit.
2024-04-20 20:51:53 +02:00
Bartosz Fenski
cceb4c62a3
typo (#6159)
s/infromation/information/
2024-04-20 05:12:15 -07:00
Github Actions
b11d4ccee3
Automatic update Grafana datasource docs from VictoriaMetrics/grafana-datasource@5b8a0ba (#6156) 2024-04-20 02:31:36 -07:00
Aliaksandr Valialkin
34e253f9d6
app/vmselect/promql: add support for matching against multiple numeric constants via q == (c1,...,cN) and q != (c1,...,cN) syntax 2024-04-19 17:56:29 +02:00
Aliaksandr Valialkin
249a467ea4
docs/CHANGELOG.md: document v1.97.4 LTS release
See https://github.com/VictoriaMetrics/VictoriaMetrics/releases/tag/v1.97.4
2024-04-19 13:10:21 +02:00
Aliaksandr Valialkin
2529f3e8eb
docs/CHANGELOG.md: document v1.93.14 LTS release
See https://github.com/VictoriaMetrics/VictoriaMetrics/releases/tag/v1.93.14
2024-04-19 12:52:21 +02:00
Aliaksandr Valialkin
ec17d4390b
docs/LTS-releases.md: change the latest v1.93.x LTS release from v1.93.13 to v1.93.14
See https://github.com/VictoriaMetrics/VictoriaMetrics/releases/tag/v1.93.14
2024-04-19 12:50:32 +02:00
Aliaksandr Valialkin
8f535f9e76
app/vmagent/remotewrite: add support for replication additionally to sharding when both -remoteWrite.shardByURL and -remoteWrite.shardByURLReplicas=RF command-line flags are set
This allows setting up data replication among failure domains if the replication factor is smaller than the number of failure domains.
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/6054

See https://docs.victoriametrics.com/vmagent/#sharding-among-remote-storages
2024-04-19 11:27:33 +02:00
Github Actions
54605f5dd1
Automatic update operator docs from VictoriaMetrics/operator@3d7bcba (#6149) 2024-04-19 09:17:37 +02:00
Hui Wang
a84491324d
vmalert: avoid blocking APIs when alerting rule uses template functio… (#6129)
* vmalert: avoid blocking APIs when alerting rule uses template function `query`

* app/vmalert: small refactoring

* simplify labels and templates expanding
* simplify `newAlert` interface
* fix `TestGroupStart` which mistakenly skipped annotations
and response labels check

Signed-off-by: hagen1778 <roman@victoriametrics.com>

* reduce alerts lock time when restore

---------

Signed-off-by: hagen1778 <roman@victoriametrics.com>
Co-authored-by: hagen1778 <roman@victoriametrics.com>
2024-04-19 09:16:26 +02:00
Roman Khavronenko
316b19a5d1
app/vmalert: make TestGroupStart more reliable (#6130)
There was a sleep statement in the test, waiting for Group
to perform a couple of evaluation. But looks like
it worked unreliable for some CI tests like the one below
https://github.com/VictoriaMetrics/VictoriaMetrics/actions/runs/8718213844/job/23915007958?pr=6115

This commit changes the sleep statement on a function that
waits for a specific number of evaluations. It should make this
test faster in general case, and more reliable for slow environemnts.
2024-04-19 09:06:40 +02:00
Aliaksandr Valialkin
565dcefc29
docs/Cluster-VictoriaMetrics.md: typo fix after dadeb6620a: folling -> following 2024-04-19 00:24:46 +02:00