Commit graph

3096 commits

Author SHA1 Message Date
Aliaksandr Valialkin
5bc240bffe vendor: make vendor-update 2021-07-13 14:30:19 +03:00
Aliaksandr Valialkin
244d0fe5d7 deployment/docker: update Go builder from v1.16.5 to v1.16.6
Ths Go release has the following bugfixes: https://github.com/golang/go/issues?q=milestone%3AGo1.16.6+label%3ACherryPickApproved
2021-07-13 14:25:41 +03:00
Aliaksandr Valialkin
a925d5a3e1 app/vmselect/promql: duration handling improvements in MetricsQL queries
- Support durations anywhere in MetricsQL queries. E.g. sum_over_time(m[1h])/1h is equivalent to sum_over_time(m[1h])/3600
- Support durations without suffix. E.g. rate(m[300]) is equivalent to rate(m[5m])
2021-07-12 17:16:41 +03:00
Aliaksandr Valialkin
f9de546139 lib/storage: reset perKeyMisses stats less frequently
This should reduce CPU usage for queries executed with intervals higher than 30 seconds
2021-07-12 14:33:42 +03:00
Aliaksandr Valialkin
4f80b2f230 lib/storage: properly limit the size of storage/date_metricID cache 2021-07-12 14:25:44 +03:00
Aliaksandr Valialkin
f3a5465ece docs/CHANGELOG.md: document the change from bfba4c28a4
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1444
2021-07-12 12:44:06 +03:00
Aliaksandr Valialkin
bfba4c28a4 app/vmalert: accept Prometheus-like durations in interval config option inside group section 2021-07-12 12:35:17 +03:00
Aliaksandr Valialkin
a684408e27 docs: update http://slack.victoriametrics.com to https://slack.victoriametrics.com 2021-07-12 10:57:16 +03:00
Aliaksandr Valialkin
8ca2799478 lib/storage: properly determine when the deduplication is needed in needsDedup
Previously needsDedup() could return true if the de-duplication wasn't needed for the following case:

         d < interval
           /     \
   |        v | v        |
     interval   interval

Now it properly returns false for this case
2021-07-12 10:53:30 +03:00
Aliaksandr Valialkin
f539772ca6 docs: sync with the cluster branch 2021-07-10 12:45:38 +03:00
Aliaksandr Valialkin
8c764e88f0 app/vmui: move source code from https://github.com/VictoriaMetrics/vmui to app/vmui
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1413
2021-07-09 17:15:23 +03:00
Aliaksandr Valialkin
2be340a4c9 docs: clarify what does "workload" mean in capacity planning docs 2021-07-09 12:49:53 +03:00
Aliaksandr Valialkin
c5f0b454f0 app/vmselect: follow-up after aa11ef6d3b
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1413
2021-07-07 17:43:35 +03:00
tony
e9e35a7d6a add vmui for vmselect component (#1431)
Co-authored-by: Aliaksandr Valialkin <valyala@gmail.com>
2021-07-07 17:33:02 +03:00
Aliaksandr Valialkin
92b8380fdf docs/Cluster-VictoriaMetrics.md: improve capacity planning recommendations 2021-07-07 16:22:51 +03:00
Aliaksandr Valialkin
1152de4321 vendor: make vendor-update 2021-07-07 16:05:04 +03:00
Aliaksandr Valialkin
7e77980608 vendor: update github.com/VictoriaMetrics/metrics from v1.17.2 to v1.17.3 2021-07-07 16:00:25 +03:00
Aliaksandr Valialkin
6e0553c92e lib/mergeset: cache indexBlock items only on the second request
This should reduce the indexdb/indexBlocks cache size, since it won't contain one-time-wonders items.
2021-07-07 15:23:06 +03:00
Aliaksandr Valialkin
ed944313b0 app/{vminsert,vmselect}: export vminsert_request_duration_seconds and vmselect_request_duration_seconds histograms 2021-07-07 13:25:21 +03:00
Aliaksandr Valialkin
766edbc421 lib/httpserver: print full requestURI in httpserver.Errorf
This should simplify debugging.
2021-07-07 13:09:40 +03:00
Aliaksandr Valialkin
c55a64ba08 docs: clarify capacity planning docs 2021-07-07 12:46:55 +03:00
Aliaksandr Valialkin
e843bd7bd7 lib/storage: do not cache inmemoryBlock entries requested only once (aka one-time-wonder items)
This should reduce the cache size and memory usage for the indexdb/dataBlocks cache
2021-07-07 10:58:51 +03:00
Aliaksandr Valialkin
8b262d4ba7 lib/storage: periodically reset prefetchedMetricIDs cache in order to limit its size under high churn rate 2021-07-07 10:58:51 +03:00
Roman Khavronenko
2f54559c89
alerts: sync alert expression for DiskRunsOutOfSpaceIn3Days with dashboard (#1436) 2021-07-07 10:31:09 +03:00
Aliaksandr Valialkin
a7694092b8 Revert "lib/uint64set: allow reusing bucket16 structs inside uint64set.Set via uint64set.Release method"
This reverts commit 7c6d3981bf.

Reason for revert: high contention at bucket16Pool on systems with big number of CPU cores.
This slows down query processing significantly.
2021-07-06 18:21:35 +03:00
Aliaksandr Valialkin
8aa9bba9bd lib/{mergeset,storage}: switch from sync.Pool to chan-based pool for inmemoryPart objects
This should reduce memory usage on systems with big number of CPU cores,
since every inmemoryPart object occupies at least 64KB of memory and sync.Pool maintains
a separate pool inmemoryPart objects per each CPU core.

Though the new scheme for the pool worsens per-cpu cache locality, this should be amortized
by big sizes of inmemoryPart objects.
2021-07-06 16:28:41 +03:00
Aliaksandr Valialkin
7c6d3981bf lib/uint64set: allow reusing bucket16 structs inside uint64set.Set via uint64set.Release method
This reduces the load on memory allocator in Go runtime in production workload.
2021-07-06 15:35:03 +03:00
Aliaksandr Valialkin
78c9174682 lib/mergeset: increase pool capacity for inmemoryBlock according to collected profiles from production workload
CPU and memory profiles show that the pool capacity for inmemoryBlock objects is too small.
This results in the increased load on memory allocation code in Go runtime.
Increase the pool capacity in order to reduce the load on Go runtime.
2021-07-06 13:41:34 +03:00
Aliaksandr Valialkin
f71e4d1853 lib/mergeset: limit the frequency for flushCallback calls to once per 10 seconds
This should improve hit ratio for tagFiltersCache when big number of new time series are constantly registered
(aka high churn rate). This, in turn, should reduce CPU usage for queries over such time series.
2021-07-06 12:17:17 +03:00
Aliaksandr Valialkin
f3acf065c9 lib/storage: consistency renaming: tagCache -> tagFiltersCache
This improves code readability
2021-07-06 11:03:51 +03:00
Aliaksandr Valialkin
0020b9f904 lib/workingsetcache: properly update stats for requests and cache misses
Previously the stats for cache misses could be improperly counted, because it had inflated cache misses
if the entry was missing in the curr cache, but was existing in the prev cache.

The same applies to cache requests - they were inflated if the entry was missing in the curr cache.
2021-07-06 10:53:32 +03:00
Roman Khavronenko
75f12bfe78
add option to add Copy button for code snippets (#1433)
To add a Copy button wrap code snippet with the following element:
```
<div class="with-copy" markdown="1">

<your-code-snippet>

</div>
```

See the changes to `Kubernetes monitoring with VictoriaMetrics Single` for details.
2021-07-06 08:23:39 +03:00
Aliaksandr Valialkin
4cf47163c1 lib/workingsetcache: fix cache capacity calculations after 4f0003f182 2021-07-05 17:11:57 +03:00
Aliaksandr Valialkin
4f0003f182 lib/workingsetcache: typo fixes after d0c830039d 2021-07-05 15:35:37 +03:00
Aliaksandr Valialkin
d0c830039d lib/storage: tune cache sizes according to production workload 2021-07-05 15:16:11 +03:00
Aliaksandr Valialkin
8f973e34fb lib/workingsetcache: properly switch to whole mode
Previously the switch from `split` to `whole` mode had been performed too early,
e.g. when the current cache size became bigger than 1/4 of the allowed cache size.

Now it is performed when the current cache size becomes bigger than 1/2 of the allowed cache size.

This change can reduce memory usage for data ingestion path when big number of active time series are ingested.
2021-07-05 15:16:11 +03:00
Aliaksandr Valialkin
43103be011 lib/{storage,mergeset}: increase cache timeout for data and index blocks from a minute to two minutes
One minute cache timeout result in slower queries in some production workloads where the interval
between query execution is in the range 1 minute - 2 minutes.
2021-07-05 15:16:11 +03:00
Aliaksandr Valialkin
54b9e1d3cb lib/cgroup: set GOGC to 50 by default if it isn't set
This should reduce memory usage for typical VictoriaMetrics workloads by up to 50%
2021-07-05 15:16:11 +03:00
Roman Khavronenko
bd6b8f7e31
move github-pages docs to the main repo (#1432)
* move github-pages docs to the main repo

* rm github actions for copying docs to VictoriaMetrics/VictoriaMetrics.github.io
2021-07-05 14:34:10 +03:00
Aliaksandr Valialkin
888d62e40c docs/CHANGELOG.md: document the bugfix for vm_merge_need_free_disk_space metric at 9a83e9018d 2021-07-05 12:01:08 +03:00
Aliaksandr Valialkin
6bef227388 docs/Articles.md: add an url to https://medium.com/ibm-garage/monitoring-of-multiple-openshift-clusters-with-victoriametrics-d4f0979e2544 2021-07-05 11:51:37 +03:00
Aliaksandr Valialkin
9a83e9018d lib/storage: properly detect free disk space shortage during data merge
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1373
2021-07-02 17:40:54 +03:00
Aliaksandr Valialkin
f6bb130898 app/{vmselect,vmstorage}: clarify the description for -dedup.minScrapeInterval command-line flag
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1426
2021-07-02 15:05:57 +03:00
Aliaksandr Valialkin
7088f17494 lib/promscrape/discovery/consul: use case-insensitive comparison for service names
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1424
2021-07-02 14:49:27 +03:00
Aliaksandr Valialkin
6e406083f2 lib/promauth: cache the client TLS certificate for up to a second
This should reduce CPU usage when TLS connections are established at a high rate.
2021-07-02 13:21:51 +03:00
Aliaksandr Valialkin
a93da746c0 vendor: update github.com/VictoriaMetrics/fasthttp from v1.0.15 to v1.0.16
This should help with proper copying of tls.Config struct after it has been set up in lib/promauth.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1420
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/470
2021-07-02 12:03:32 +03:00
Aliaksandr Valialkin
1e52f11e87 docs/Cluster-VictoriaMetrics.md: typo fix: siplify -> simplify 2021-07-02 10:50:55 +03:00
Aliaksandr Valialkin
5beb099ff9 docs/Cluster-VictoriaMetrics.md: add a chapter describing a toy cluster setup on a single host
While at it, refer to available tools, which can simplify cluster setup
2021-07-02 10:48:56 +03:00
Aliaksandr Valialkin
158c50c0ee lib/promauth: reload TLS certificates from disk on every mTLS connection as Prometheus does
This allows updating client certificates without the need to restart vmagent and/or single-node VictoriaMetrics.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1420
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/470
2021-07-01 15:27:47 +03:00
Aliaksandr Valialkin
02ed4b8b46 docs/CHANGELOG.md: document ae485c2bfd 2021-07-01 11:51:19 +03:00