Commit graph

8156 commits

Author SHA1 Message Date
Aliaksandr Valialkin
8e03bc6b53
app/vmselect/promql: do not spend CPU time on verifying whether the rollup cache needs to be reset for the given metric rows when it has been already instructed to reset 2024-01-26 21:13:38 +01:00
Aliaksandr Valialkin
4ac7e3a355
docs/Makefile: mention that the Makefile rules must be run from VictoriaMetrics repository root 2024-01-26 21:01:40 +01:00
Aliaksandr Valialkin
32c064a401
app/vmauth: return 503 service unavailable status code when the backend returns response with unsupported status code, but the request cannot be re-tried.
While at it, properly close response body. This should prevent from possible http keep-alive connection leak to backends because of unclosed response bodies.

This is a follow-up for 3c0aa14b5b
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/pull/5688
2024-01-26 20:43:11 +01:00
Artem Navoiev
60fc2da6c1
docs: fix key concepts image and links
Signed-off-by: Artem Navoiev <tenmozes@gmail.com>
2024-01-26 20:30:29 +01:00
Artem Navoiev
25165656bb
docs: change [image] to img as far we support it in release guide
Signed-off-by: Artem Navoiev <tenmozes@gmail.com>
2024-01-26 19:01:20 +01:00
Artem Navoiev
41e99765cc
docs: remoev vmanomaly as far we have dedicated section with alredy exists redirects
Signed-off-by: Artem Navoiev <tenmozes@gmail.com>
2024-01-26 18:38:01 +01:00
Artem Navoiev
bc033a2b30
docs: vmanomaly fix images
Signed-off-by: Artem Navoiev <tenmozes@gmail.com>
2024-01-26 17:59:37 +01:00
Daria Karavaieva
105cb44884
Vmanomaly Guide dashboard provisioning (#5679)
* dashboard provisioning

* delete dashboard filter, new query

* dashboard screens, guide fixes
2024-01-26 17:12:58 +01:00
Artem Navoiev
9ded04e643
docs: remove raw and endraw tags as they are not needed for the new v… (#5696)
* docs: remove raw and endraw tags as they are not needed for the new version of site

Signed-off-by: Artem Navoiev <tenmozes@gmail.com>

* revert formating in vmaler

Signed-off-by: Artem Navoiev <tenmozes@gmail.com>

---------

Signed-off-by: Artem Navoiev <tenmozes@gmail.com>
2024-01-26 07:30:45 -08:00
Github Actions
fae801edd3
Automatic update operator docs from VictoriaMetrics/operator@0628def (#5694) 2024-01-26 10:21:52 +01:00
Github Actions
2582b1e15a
Automatic update Grafana datasource docs from VictoriaMetrics/grafana-datasource@c644bec (#5691) 2024-01-26 11:44:02 +04:00
Roman Khavronenko
b11f4ef5ea
app/vmalert: autogenerate ALERTS_FOR_STATE time series for alerting rules with for: 0 (#5680)
* app/vmalert: autogenerate `ALERTS_FOR_STATE` time series for alerting rules with `for: 0`

 Previously, `ALERTS_FOR_STATE` was generated only for alerts with `for > 0`.
 This behavior differs from Prometheus behavior - it generates ALERTS_FOR_STATE
 time series for alerting rules with `for: 0` as well. Such time series can
 be useful for tracking the moment when alerting rule became active.

 Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/5648
 https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3056

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

* app/vmalert: support ALERTS_FOR_STATE in `replay` mode

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

---------

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-01-25 15:42:57 +01:00
Github Actions
a95246d885
Automatic update operator docs from VictoriaMetrics/operator@e75a096 (#5690) 2024-01-25 15:33:01 +01:00
hagen1778
d71218d6ce
docs: simplify instructions for spinning up docker env
Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-01-25 15:27:57 +01:00
Github Actions
e29fe0933b
Automatic update operator docs from VictoriaMetrics/operator@f6b9c08 (#5676) 2024-01-25 15:10:23 +01:00
Alexander Marshalov
3c0aa14b5b
vmauth: fix vmauth_user_request_backend_errors_total metric calc logic for use case when only one backend is available - if we get an error from the retry_status_codes list, but cannot execute retry, we increment vmauth_user_request_backend_errors_total as well (#5688) 2024-01-25 14:04:20 +01:00
hagen1778
56310ffb47
docs: fix the issue link
Signed-off-by: hagen1778 <roman@victoriametrics.com>
2024-01-25 13:35:41 +01:00
Alexander Marshalov
495fa9800a
Follow up after https://github.com/VictoriaMetrics/VictoriaMetrics/pull/5685 (#5686) 2024-01-25 10:03:46 +01:00
Alexander Marshalov
d5682858c0
Make a makefile work on a MacOS and Linux (#5685) 2024-01-25 09:58:01 +01:00
Aliaksandr Valialkin
c3a585cfe5
lib/storage: rename *AssistedMerges to *AssistedMergesCount in order to make these field names less misleading
These fields are counters, not gauges, so adding Count suffix to them makes easier to understand this while reading the code
2024-01-25 10:19:32 +02:00
Alexander Marshalov
806c07ddd5
vmsingle/vmselect returns http status 429 (TooManyRequests) instead of 503 (ServiceUnavailable) when max concurrent requests limit is reached. (#5682) 2024-01-24 17:55:06 +01:00
Aliaksandr Valialkin
18df07e824
lib/mergeset: start assisted merge for file parts only if the number of file parts is bigger than maxFileParts
The maxFileParts usage has been accidentally removed in fa566c68a6

While at it, add Count suffix to *AssistedMerges counter names in order to make them less misleading.
Previously their names were falsely suggesting that these are gauges, which show the number of concurrently
executed assisted merges.
2024-01-24 15:08:42 +02:00
Aliaksandr Valialkin
ac5b740750
lib/promscrape/discovery/kubernetes: typo fix in the comment for ContainerStateTerminated struct
This is a follow-up for ef12598ad4
2024-01-24 15:06:46 +02:00
Aliaksandr Valialkin
ef12598ad4
lib/promscrape/discovery/kubernetes: do not generate targets for already terminated pods and containers
Already terminated pods and containers cannot be scraped and will never resurrect,
so there is zero sense in creating scrape targets for them.
2024-01-24 14:57:53 +02:00
Aliaksandr Valialkin
4d961c70f7
app/{vmselect,vmstorage}: return compression of the data passed from vmstorage to vmselect
This reverts cd4f641d32 , since it has been appeared that the disabled compression
for vmstorage->vmselect data increase network bandwidth usage by more than 10x on typical production workloads,
while it decreases CPU usage at vmstorage by up to 10% and improves query latency by up to 10%.

The 10x increase in network usage is too high price for 10% improvements on query latency and vmstorage CPU usage.
This may result in network bandwidth bottlenecks, which can reduce the overall performance and stability
of VictoriaMetrics cluster. That's why return back the vmstorage->vmselect data compression by default.

The vmstorage->vmselect compression can be disabled by passing -rpc.disableCompression command-line flag to vmstorage.
The vmselect->vmselect compression in multi-level cluster setup can be disabled by passing -clusternative.disableCompression command-line flag.
2024-01-24 13:39:28 +02:00
Aliaksandr Valialkin
f888a019fe
lib/streamaggr: expand %{ENV} placeholders in stream aggregation configs 2024-01-24 12:31:27 +02:00
Aliaksandr Valialkin
fa566c68a6
lib/mergeset: really limit the number of in-memory parts to 15
It has been appeared that the registration of new time series slows down linearly
with the number of indexdb parts, since VictoriaMetrics needs to check every indexdb part
when it searches for TSID by newly ingested metric name.

The number of in-memory parts grows when new time series are registered
at high rate. The number of in-memory parts grows faster on systems with big number
of CPU cores, because the mergeset maintains per-CPU buffers with newly added entries
for the indexdb, and every such entry is transformed eventually into a separate in-memory part.

The solution has been suggested in https://github.com/VictoriaMetrics/VictoriaMetrics/pull/5212
by @misutoth - to limit the number of in-memory parts with buffered channel.
This solution is implemented in this commit. Additionally, this commit merges per-CPU parts
into a single part before adding it to the list of in-memory parts. This reduces CPU load
when searching for TSID by newly ingested metric name.

The https://github.com/VictoriaMetrics/VictoriaMetrics/pull/5212 recommends setting the limit on the number
of in-memory parts to 100, but my internal testing shows that much lower limit 15 works with the same efficiency
on a system with 16 CPU cores while reducing memory usage for `indexdb/dataBlocks` cache by up to 50%.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/5190
2024-01-24 03:38:12 +02:00
Aliaksandr Valialkin
5543c04061
docs/Cluster-VictoriaMetrics.md: document that vmstorage doesnt compress data it sends to vmselect by default
This is a follow-up for cd4f641d32
2024-01-23 23:22:18 +02:00
Aliaksandr Valialkin
8fb8b71295
lib/encoding: remove uneeded re-slicing of byte slice before passing it to binary.BigEndian.Uint* 2024-01-23 22:50:29 +02:00
Aliaksandr Valialkin
1c58c00618
app/vmselect/netstorage: limit the initial size for brsPoolCap with 32Kb
This should reduce the number of expensive memory allocations with sizes bigger than 32Kb
2024-01-23 22:29:39 +02:00
Aliaksandr Valialkin
43ecd5d258
app/vmselect/netstorage: pre-allocate memory for metricNamesBuf
This should reduce the number of metricNamesBuf re-allocations in append()
2024-01-23 21:34:16 +02:00
Aliaksandr Valialkin
ae643ef1f1
lib/{storage,mergeset}: reduce the maxium compression level for the stored data
This reduces CPU usage a bit, while doesn't increase resulting file sizes according to synthetic tests.
2024-01-23 17:46:50 +02:00
Github Actions
05c9a4d7ce
Automatic update operator docs from VictoriaMetrics/operator@1470569 (#5668) 2024-01-23 16:22:16 +01:00
Aliaksandr Valialkin
6c214397ed
lib/storage: compress metricIDs, which match the given filters, before storing them in tagFiltersToMetricIDsCache
This allows reducing the indexdb/tagFiltersToMetricIDs cache size by 8 on average.
The cache size can be checked via vm_cache_size_bytes{type="indexdb/tagFiltersToMetricIDs"} metric exposed at /metrics page.
2024-01-23 16:09:55 +02:00
Aliaksandr Valialkin
4d78954158
lib/storage: do not sort metricIDs passed to Storage.prefetchMetricNames, since the caller is responsible for the sorting 2024-01-23 16:08:38 +02:00
Aliaksandr Valialkin
6d84b1beef
lib/filestream: do not measure read / write duration from / to in-memory buffers
Measuring read / write duration from / to in-memory buffers has little sense,
since it will be always fast. It is better to measure read / write duration from / to
real files at vm_filestream_write_duration_seconds_total and vm_filestream_read_duration_seconds_total metrics.
This also reduces overhead on time.Now() and Histogram.UpdateDuration() calls
per each filestream.Reader.Read() and filestream.Writer.Write() call when the data is read / written from / to in-memory buffers.

This is a follow-up for 2f63dec2e3
2024-01-23 14:52:22 +02:00
Aliaksandr Valialkin
41456d9569
app/vmselect/netstorage: limit the maximum brsPool size to 32Kb at ProcessSearchQuery()
This avoids slow path in Go runtime for allocating objects bigger than 32Kb -
see 704401ffa0/src/runtime/malloc.go (L11)

This also reduces memory usage a bit for vmselect and single-node VictoriaMetrics
after the commit 5dd37ad836 .

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/pull/5527
2024-01-23 14:04:49 +02:00
Aliaksandr Valialkin
1f1768d7af
app/vmselect/netstorage: limit the size of metricNamesBuf to 32Kb in order to avoid slow path at Go runtime for allocating a byte slice of bigger size
See 704401ffa0/src/runtime/malloc.go (L11)

This also reduces the average memory usage a bit for vmselect and single-node VictoriaMetrics
after the commit 508c608062

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/pull/5527
2024-01-23 13:46:37 +02:00
Aliaksandr Valialkin
fac7c30f4e
docs/vmagent.md: clarify how -promscrape.seriesLimitPerTarget command-line flag, series_limit config option and __series_limit__ label interact with each other
This is a follow-up for https://github.com/VictoriaMetrics/VictoriaMetrics/pull/5663
See also 89e3c70ccd
2024-01-23 13:14:50 +02:00
Roman Khavronenko
89e3c70ccd
lib/promscrape: respect 0 value for series_limit param (#5663)
* lib/promscrape: respect `0` value for `series_limit` param

Respect `0` value for `series_limit` param in `scrape_config`
even if global limit was set via `-promscrape.seriesLimitPerTarget`.
Previously, `0` value will be ignored in favor of `-promscrape.seriesLimitPerTarget`.

This behavior aligns with possibility to override `series_limit` value via
relabeling with `__series_limit__` label.

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

* Update docs/CHANGELOG.md

---------

Signed-off-by: hagen1778 <roman@victoriametrics.com>
Co-authored-by: Aliaksandr Valialkin <valyala@victoriametrics.com>
2024-01-23 13:09:14 +02:00
Aliaksandr Valialkin
1c5163ae51
lib/mergeset: make sure that the first and the last items are in the original range after prepareBlock()
Previously the checks were to strict by requiring to leave the same first and last items by prepareBlock()

Thanks to @ahfuzhang for the suggestion at https://github.com/VictoriaMetrics/VictoriaMetrics/issues/5655
2024-01-23 12:58:32 +02:00
Fred Navruzov
2adb38a9c4
- fix 404 errors after page remaning (#5664)
- slight text fixes
2024-01-23 01:56:42 -08:00
Aliaksandr Valialkin
15a15e5b99
app/vmselect/vmui: run make vmui-update in order to sync recent changes in app/vmui 2024-01-23 04:31:44 +02:00
Aliaksandr Valialkin
114822d585
app/{vmstorage,vmselect}: disable vmstorage->vmselect RPC compression by default in order to improve query performance 2024-01-23 04:24:57 +02:00
Zakhar Bessarab
bf4742526d
lib/storage: print tenant ID in log when discarding or truncating labels (#5658)
Previously, it was not possible to determine which tenant sends metrics with excessive amount of labels of label values.

Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>
Co-authored-by: Aliaksandr Valialkin <valyala@victoriametrics.com>
2024-01-23 04:24:56 +02:00
Yury Molodov
38231d5994
vmui: query report (#5497)
* vmui: add query analyzer page

* vmui: fix tabs for query analyzer

* vmui: add help to export query

* vmui: add time params to query analyzer

* docs/vmui: add query analyzer

* vmui: fix validation JSON form

---------

Co-authored-by: Aliaksandr Valialkin <valyala@victoriametrics.com>
2024-01-23 04:23:26 +02:00
Yury Molodov
eb6def0695
vmui: add flag for default timezone setting (#5611)
* vmui: add flag for default timezone setting #5375

* vmui: validate timezone before client return

* Update app/vmselect/vmui.go

---------

Co-authored-by: Aliaksandr Valialkin <valyala@victoriametrics.com>
2024-01-23 04:11:19 +02:00
Yury Molodov
633e6b48ad
vmui: fix cache autocomplete (#5591)
* vmui: fix the logic of closing the popper #5470

* vmui: fix the logic of caching autocomplete results #5472

---------

Co-authored-by: Aliaksandr Valialkin <valyala@victoriametrics.com>
2024-01-23 04:06:14 +02:00
Aliaksandr Valialkin
980338861f
lib/mergeset: skip comparison for every item in the block during merge if the last item in the block is smaller than the first item in the next block
Thanks to @ahfuzhang for the suggestion at https://github.com/VictoriaMetrics/VictoriaMetrics/issues/5651
2024-01-23 03:15:52 +02:00
Aliaksandr Valialkin
bc7d19c8ca
app/vmselect/promql: remove superflouos memory allocations at aggrPrepareSeries()
While at it, also remove unneeded map lookup
2024-01-23 02:28:31 +02:00