github-mirrors/VictoriaMetrics
1
0
Fork 0
mirror of https://github.com/VictoriaMetrics/VictoriaMetrics.git synced 2024-11-21 14:44:00 +00:00
Code Issues Projects Releases Packages Wiki Activity
3274 commits 150 branches 748 tags 347 MiB
Commit graph

3274 commits

This branch
This branch
All branches
Author SHA1 Message Date
Aliaksandr Valialkin
5d64ed73c5 lib/protoparser/clusternative: do not pool unmarshalWork structs, since they can occupy big amounts of memory (more than 100MB per each struct) 
This should reduce memory usage for vmstorage under high ingestion rate when the vmstorage runs on a system with big number of CPU cores
 2021-06-23 15:46:50 +03:00
Aliaksandr Valialkin
cdfae0117a app/vmselect/promql: return the last timestamp for the max / min value from tmax_over_time() and tmin_over_time() function as most users expect 2021-06-23 14:19:00 +03:00
Aliaksandr Valialkin
70e2852376 docs/CHANGELOG.md: document the bugfix for incorrect stats collection for concurrently executed tag filter 
Follow up for c22114c6f0
 2021-06-23 14:05:28 +03:00
Aliaksandr Valialkin
94f3e40ab3 app/vminsert/netstorage: sort the -storageNode list passed to vminsert nodes 
This should reduce resource usage (CPU, RAM, disk IO) at vmstorage nodes
if the addresses of vmstorage nodes are passed in random order to vminsert nodes.
 2021-06-23 14:01:41 +03:00
Aliaksandr Valialkin
c22114c6f0 lib/storage: tune tag filters search logic 
Tune the logic according to the logs provided at https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1338#issuecomment-864293624

The previous logic had a race when multiple concurrent queries execute the same tag filter without prior stats.
This could result in incorrectly stored stats for such tag filter, which then could result in non-optimal sorting of tag filters
for further queries.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1338
 2021-06-23 13:29:39 +03:00
Aliaksandr Valialkin
e8a5bb92b7 lib/promscrape/discovery/consul: properly pass namespace to Consul watcher 
Follow-up for 58a2989fe7
 2021-06-22 17:42:41 +03:00
Aliaksandr Valialkin
ac54f34f9e lib/promscrape/discovery/http: follow up after e307bbb29a 2021-06-22 13:40:33 +03:00
Aliaksandr Valialkin
755040a171 lib/promscrape/discovery: support generic auth configs in Consul service discovery in the same way as Prometheus 2.28 does 2021-06-22 13:34:02 +03:00
Aliaksandr Valialkin
59e7755df9 docs/CHANGELOG.md: document the support for Consul namepsace 
See 58a2989fe7
 2021-06-22 13:34:02 +03:00
Nikolay
e307bbb29a
adds http_sd (#1399) 
* adds http_sd

* adds X-Prometheus-Refresh-Interval-Seconds header

* Update lib/promscrape/discovery/http/api.go

Co-authored-by: Aliaksandr Valialkin <valyala@gmail.com>
 2021-06-22 13:33:37 +03:00
Nikolay
58a2989fe7
adds consul enterprise namespace support (#1400) 
* adds consul enterprise namespace support

* Update lib/promscrape/discovery/consul/consul.go

Co-authored-by: Aliaksandr Valialkin <valyala@gmail.com>
 2021-06-22 12:49:44 +03:00
Aliaksandr Valialkin
3bc83d3b17 docs/PerTenantStatistic.md: document that the per-tenant statistic is a part of cluster version of VictoriaMetrics 2021-06-22 12:43:01 +03:00
Roman Khavronenko
da8c901fab
vmctl: add more context to flags description in vm-native mode (#1395) 2021-06-18 19:20:01 +03:00
Aliaksandr Valialkin
a3262daac0 docs/CHANGELOG.md: typo fixes 2021-06-18 19:14:53 +03:00
Aliaksandr Valialkin
83a4db813e app/vmselect: log slow requests to all the /api/v1/* handlers if their execution time exceeds -search.logSlowQueryDuration 2021-06-18 19:04:42 +03:00
Aliaksandr Valialkin
4cfedc5931 docs/Single-server-VictoriaMetrics.md: mention that it is recommended to use a single scrape_interval across all the scrape targets 2021-06-18 15:39:33 +03:00
Aliaksandr Valialkin
570f36b344
app/vmctl: limit JSON line size by 10K samples (#1394) 
This should reduce the maximum memory usage at VictoriaMetrics when importing time series with big number of samples.
 2021-06-18 15:26:47 +03:00
Aliaksandr Valialkin
eb1af09a04 docs/Cluster-VictoriaMetrics.md: clarify docs about VictoriaMetrics cluster architecture 2021-06-18 14:36:39 +03:00
Aliaksandr Valialkin
cbd9159a22 docs/CHANGELOG.md: document the reduced disk write IO usage 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1338
 2021-06-18 14:02:59 +03:00
Aliaksandr Valialkin
fb72a2133f lib/promscrape: show jobs with empty scrape targets on /targets page 2021-06-18 10:53:52 +03:00
Aliaksandr Valialkin
d8ab409418 docs/{vmgateway,vmbackupmanager}: explicitly mention that these components are a part of an enterprise package 2021-06-17 17:19:49 +03:00
Nikolay
6c434b260e
fixes DO service discovery labels (#1389) 
adds test for digitalocean sd
 2021-06-17 15:12:20 +03:00
Aliaksandr Valialkin
dcbc22552f lib/storage: fix infinite loop introduced in aa9b56a046 2021-06-17 14:28:10 +03:00
Aliaksandr Valialkin
aa9b56a046 lib/{mergeset,storage}: reduce the number of fsync calls on data ingestion path on systems with many cpu cores 
VictoriaMetrics maintains a buffer per CPU core for the ingested data. These buffers are flushed to disk every second.
These buffers are flushed to disk in parallel starting from the commit 56b6b893ce .
This resulted in increased write disk IO usage on systems with many cpu cores
as described at https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1338#issuecomment-863046999 .

This commit merges the per-CPU buffers into bigger in-memory buffers before flushing them to disk.
This should reduce the rate of fsync syscalls and, consequently, the write disk IO on systems with many CPU cores.

This should help https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1338
See also https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1244
 2021-06-17 13:52:08 +03:00
Aliaksandr Valialkin
12a83d25bf app/vmagent/remotewrite: go fmt after 0a796f7c3a 2021-06-17 13:52:06 +03:00
Aliaksandr Valialkin
9eb3fc346f docs/vmagent.md: sync with app/vmagent/README.md via make docs-sync 2021-06-16 12:36:49 +03:00
Aliaksandr Valialkin
6d17a4e12d docs/CHANGELOG.md: document the changed -remoteWrite.queues value 
This is a follow-up for 0a796f7c3a

See https://github.com/VictoriaMetrics/VictoriaMetrics/pull/1385
 2021-06-16 12:35:46 +03:00
Zongyang
0a796f7c3a
Change default value of '-remoteWrite.queues' to cgroup.AvailableCPUs * 2 (#1385) 
* Change default value of '-remoteWrite.queues' to cgroup.AvailableCPUS() * 2 to reduce scrape interval

Default value of vmagent option '-remotewrite.queues' is 4 and default
size of vmagent ScheudleUnmarshalWorkers is number of CPUs, when available
CPUs is much greater than 4, e.g 32, worker are competing push queues
which will increase scrape interval and may cause scrape timeout.

* Update README and flag description

Co-authored-by: xiaozy <xiaozy01@fenbi.com>
 2021-06-16 12:16:44 +03:00
Roman Khavronenko
fb4f758715
promql: fix increase_pure calculation for cases with stale series (#1381) 
Due to staleness handling, increase_pure were using incorrect previous value
during calculation in cases where series disappears for period longer
than staleness period and then returns back. The fix suppose to account
for a real datapoint value before staleness takes place. The fix should
remove unexpected spikes while using `increase_pure` for staled series.
 2021-06-15 17:37:19 +03:00
Aliaksandr Valialkin
6a8369f0fc docs/Single-server-VictoriaMetrics.md: mention that VictoriaMetrics works great with APM workloads (aka Application Performance Monitoring) 2021-06-15 17:32:52 +03:00
Aliaksandr Valialkin
84fb59b0ba lib/storage: move deletedMetricIDs set from indexDB to Storage 
This makes consitent the list of deleted metricIDs when it is used from both the current indexDB and the previous indexDB (aka extDB).
This should fix the issue, which could lead to storing new samples under deleted metricIDs after indexDB rotation.
See more details at https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1347#issuecomment-861232136 .

Thanks to @tangqipengleoo for the initial analysis and the pull request - https://github.com/VictoriaMetrics/VictoriaMetrics/pull/1383 .

This commit resolves the issue in more generic way compared to https://github.com/VictoriaMetrics/VictoriaMetrics/pull/1383 .

The downside of the commit is the deletedMetricIDs set isn't cleaned from the metricIDs outside the retention. It needs app restart.
This should be OK in most cases.
 2021-06-15 15:04:30 +03:00
Aliaksandr Valialkin
e028ad241a lib/protoparser: stop reading the input stream as soon as the callback provided by the caller returns error 
This is a follow-up for af90c3c43b
 2021-06-14 15:18:49 +03:00
faceair
af90c3c43b
lib/protoparser: stop read when callback error (#1380) 2021-06-14 15:10:58 +03:00
Aliaksandr Valialkin
36d55bff66 lib/promscrape: show the number of samples collected during the last scrape at /targets and /api/v1/targets pages 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1377
 2021-06-14 14:04:00 +03:00
Aliaksandr Valialkin
7b283ee91c vendor: update github.com/klauspost/compress from v1.13.0 to v1.13.1 2021-06-14 13:42:46 +03:00
Roman Khavronenko
a90012ef26
dashboard: bump version requirements (#1378) 2021-06-14 13:31:59 +03:00
Aliaksandr Valialkin
05bc9667c1 docs/CHANGELOG.md: document the addition of DigitalOcean service discovery 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1367
 2021-06-14 13:18:19 +03:00
Nikolay
729c4eeb9c
adds digital ocean sd (#1376) 
* adds digital ocean sd config

* adds digital ocean sd
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1367

* typo fix
 2021-06-14 13:15:04 +03:00
Roman Khavronenko
b8526e88d3
Dashboard single (#1374) 
* dashboard: update single version dash

The update contains the following changes:
* display anonymous memory usage metric. This metric suppose to reflect
memory usage of the process which can't be freed by OS;
* add legends to all panels. This is important for cases when users share
the screenshots;
* modify panels for Grafana v8.0.0

* dashboard: update single version dash tags

* dashboard: update vmagent dash

The update contains the following changes:
* display anonymous memory usage metric. This metric suppose to reflect
memory usage of the process which can't be freed by OS;
* add legends to all panels. This is important for cases when users share
the screenshots;
* modify panels for Grafana v8.0.0
 2021-06-14 13:03:23 +03:00
Aliaksandr Valialkin
06b8e7d148 lib/promscrape: increase the duration for reading the full response in stream parsing mode 
Increase the duration from 10x to 30x of the configured `scrape_interval'.

This should help https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1365
 2021-06-14 12:28:09 +03:00
Aliaksandr Valialkin
48210130ac lib/protoparser: measure the duration for reading the whole block of data instead of a single read operation 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1365
 2021-06-14 12:25:52 +03:00
Aliaksandr Valialkin
3e5b6bae66 docs/Cluster-VictoriaMetrics.md: add lists for command-line flags for cluster components 2021-06-14 12:22:05 +03:00
Aliaksandr Valialkin
3c4366806c lib/protoparser/common: log the duration for reading a block of data in ReadLinesBlockExt on error 
This may help debugging issues like https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1365
 2021-06-14 12:22:04 +03:00
Aliaksandr Valialkin
8a519f1518 docs/vmalert.md: follow-up after 6d5a8c28cd 2021-06-14 11:37:26 +03:00
Roman Khavronenko
6d5a8c28cd
Vmalert docs (#1372) 
* vmalert: mention what happens if `for` is set to 0 or omitted

* vmalert: add more context to docs
 2021-06-11 13:25:53 +03:00
Aliaksandr Valialkin
2ef2e3d6f8 docs/CHANGELOG.md: cut v1.61.1 2021-06-11 13:01:31 +03:00
Aliaksandr Valialkin
ed83558646 app/vmauth: properly handle http.ErrAbortHandler panic 
This panic can be raised by the reverseProxy on aborted request to the backend.
So handle it (e.g. suppress) at reverseProxy.ServeHTTP call.

Do not suppress the panic at lib/httpserver generic HTTP handler,
since it may result in an inconsistent state left after the panicking handler.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1353
 2021-06-11 12:50:25 +03:00
Aliaksandr Valialkin
c4f3fbfa5d lib/storage: reset cache on disk during series deletion and during indexdb rotation 
This should prevent from inconsistent behavior (aka partially missing data for some time series) after unclean shutdown.

See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1347
 2021-06-11 12:42:28 +03:00
Aliaksandr Valialkin
d979e14da2 docs/CHANGELOG.md: document the bugfix from 7adfe878e1 2021-06-11 11:28:10 +03:00
Roman Khavronenko
7adfe878e1
vmalert: fix mistake with object reuse while parsing response (#1370) 
* vmalert: fix mistake with object reuse while parsing response

During the refactoring, the wrong optimisations was applied in
parse function which caused metric fields reset. The change removes
optimisation.

https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1369

* vmalert: add test to cover multiple metrics in one response
 2021-06-11 11:22:05 +03:00