Commit graph

6051 commits

Author SHA1 Message Date
Aliaksandr Valialkin
dff47c73b7
app/vmselect: improve logging when the incoming query cannot be executed because of timeout in the wait queue 2023-01-11 01:06:05 -08:00
Aliaksandr Valialkin
65b9dcfcca
app/vmselect/promql: typo fix after 0771d57860 2023-01-11 01:05:31 -08:00
Aliaksandr Valialkin
0771d57860
app/vmselect/promql: make a copy of per-series timestamps before their modification
The per-series timestamps are usually shared among series, so it is unsafe modifying them.

The issue has been appeared after the optimization at 2f3ddd4884
2023-01-11 00:59:13 -08:00
Aliaksandr Valialkin
31fc29599f
app/vmselect/promql: move the eval function args in parallel query trace outside the loop 2023-01-10 22:23:30 -08:00
Aliaksandr Valialkin
a0f9cb27f9
docs/CHANGELOG.md: document v1.79.7 LTS release 2023-01-10 21:24:28 -08:00
Aliaksandr Valialkin
4faf7ea41e
vendor: make vendor-update 2023-01-10 18:58:34 -08:00
Aliaksandr Valialkin
c449714c0a
deployment/docker: update Go builder from v1.19.4 to v1.19.5
See https://github.com/golang/go/issues?q=milestone%3AGo1.19.5+label%3ACherryPickApproved
2023-01-10 18:43:04 -08:00
Aliaksandr Valialkin
37fe04999c
docs/CHANGELOG.md: add release date for v1.86.0 2023-01-10 17:45:57 -08:00
Aliaksandr Valialkin
eda940106a
deployment/docker: update Docker tag for VictoriaMetrics components from v1.85.3 to v1.86.0 2023-01-10 17:26:37 -08:00
Aliaksandr Valialkin
393e7636be
Merge branch 'public-single-node' into pmm-6401-read-prometheus-data-files 2023-01-10 16:23:24 -08:00
Aliaksandr Valialkin
28df7c2a96
docs/CHANGELOG.md: cut v1.86.0 2023-01-10 16:22:26 -08:00
Aliaksandr Valialkin
2d294cca59
deployment/docker: update Alpine base image from v3.17.0 to v3.17.1
See https://alpinelinux.org/posts/Alpine-3.17.1-released.html
2023-01-10 16:14:40 -08:00
Aliaksandr Valialkin
ebc200846c
Merge branch 'public-single-node' into pmm-6401-read-prometheus-data-files 2023-01-10 16:11:27 -08:00
Aliaksandr Valialkin
95ce1ba6ce
lib/httpserver: directly pass flag value to CheckAuthFlag()
There is no sense in passing a pointer to flag value there.

This is a follow-up for 4225a0bd75
2023-01-10 15:52:23 -08:00
Zakhar Bessarab
4225a0bd75
Use httpAuth.* flags as a fallback for endpoints protected by *AuthKey flags (#3582)
* {lib/server, app/}: use `httpAuth.*` flag as fallback for `*AuthKey` if it is not set

* lib/ingestserver/opentsdbhttp: fix opentdb HTTP handler not respecting `httpAuth.*` flags

* Apply suggestions from code review

Co-authored-by: Aliaksandr Valialkin <valyala@victoriametrics.com>
2023-01-10 15:46:13 -08:00
Dmytro Kozlov
0811000bb0
app/vmctl: Add insecure skip verify flag for remote read protocol (#3611)
* app/vmctl: Add insecure skip verify flag for remote read protocol
2023-01-10 23:18:49 +01:00
Artem Navoiev
37c52ccaf4
vmagent: add minimal scrape file exampe for vmagent quick start (#3627)
* vmagent: add minimal scrape file exampe for vmagent quick start

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

* replace example with link to your prometheus.yml in docker

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

Signed-off-by: Artem Navoiev <tenmozes@gmail.com>
2023-01-10 23:16:10 +01:00
Aliaksandr Valialkin
cbe62f23ba
lib/promscrape/discovery/gce: follow-up for b2ccdaaa2f
- Use promutils.Labels.GetLabels() instead of comparing promutils.Labels.Labels to nil.
  This make the code more consistent with other places.

- Mention the release where the issue has been introduced at docs/CHANGELOG.md.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3624
2023-01-10 13:51:03 -08:00
Aliaksandr Valialkin
53d871d0b1
app/vmselect/netstorage: reduce tail latency during query processing
Previously the selected time series were split evenly among available CPU cores
for further processing - e.g unpacking the data and applying the given rollup
function to the unpacked data.
Some time series could be processed slower than others.
This could result in uneven work distribution among available CPU cores,
e.g. some CPU cores could complete their work sooner than others.
This could slow down query execution.

The new algorithm allows stealing time series to process from other CPU cores
when all the local work is done. This should reduce the maximum time
needed for query execution (aka tail latency).

The new algorithm should also scale better on systems with many CPU cores,
since every CPU processes locally assigned time series without inter-CPU communications.

The inter-CPU communications are used only when all the local work is finished
and the pending work from other CPUs needs to be stealed.
2023-01-10 13:43:14 -08:00
Zakhar Bessarab
b2ccdaaa2f
lib/promscrape/discovery/gce: fix crash in case instance does not have any labels set (#3625)
Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>
2023-01-10 11:07:11 +01:00
Denys Holius
b06e795a1e
docs/Release-Guide.md: added missed link to rpm repository (#3623) 2023-01-10 10:14:56 +01:00
Aliaksandr Valialkin
e640ff72f1
app/vmselect/netstorage: reduce memory allocations when unpacking time series
Unpack time series with less than 400K samples in the currently running goroutine.
Previously a new goroutine was being started for unpacking the samples.
This was requiring additional memory allocations.
2023-01-09 23:18:17 -08:00
Aliaksandr Valialkin
9a563a6aef
app/vmselect/promql: eliminate memory allocation when sorting values inside float64s 2023-01-09 23:06:46 -08:00
Aliaksandr Valialkin
30ed33fae0
app/vmselect/promql: pre-allocate memory for values to be merged in mergeTimeseries()
This should reduce the number of memory re-allocations
2023-01-09 22:51:17 -08:00
Aliaksandr Valialkin
645c24dc5f
app/vmselect/promql: consistently intern series names obtained from marshalMetricNameSorted
This reduces memory allocations when the returned series names are used as map keys later
2023-01-09 22:45:40 -08:00
Aliaksandr Valialkin
2f3ddd4884
app/vmselect/promql: avoid memory allocations and copying from source timeseries to the returned result at timeseriesToResult() 2023-01-09 22:38:59 -08:00
Aliaksandr Valialkin
26cf680468
app/vmselect/promql: remove memory allocations from sortMetricTags() 2023-01-09 22:22:15 -08:00
Aliaksandr Valialkin
4f0c11ee93
app/vmselect/promql: intern output series names inside timeseriesToResult()
This reduces the number of memory allocations for repeated queries,
which return (almost) the same set of time series.
2023-01-09 22:19:56 -08:00
Aliaksandr Valialkin
562d6bca08
app/vmselect/promql: intern output series names during normal aggregation 2023-01-09 22:15:24 -08:00
Aliaksandr Valialkin
21ee9a1fab
app/vmselect/promql: intern output series names during incremental aggregation
This should reduce the number of memory allocations for repeated queries
2023-01-09 22:11:36 -08:00
Aliaksandr Valialkin
df2a494a7c
app/vmselect/netstorage: pre-allocate 4 block references per each time series during querying
Usually the number of blocks returned per each time series during queries is around 4.
So it is a good idea to pre-allocate 4 block references per time series
in order to reduce the number of memory allocations.
2023-01-09 22:03:23 -08:00
Aliaksandr Valialkin
c5e0f527bc
app/vmselect/netstorage: cache canonical MetricName for time series returned from the storage
This reduces memory allocations for repeated queries, which return (almost) the same set of time series.
2023-01-09 21:53:10 -08:00
Aliaksandr Valialkin
7afcca0c51
all: use metricsql.CompileRegexp instead of regexp.Compile for compiling regexps used in graphite queries
This should speed up repeated queries, since metricsql.CompileRegexp returns regexps from the cache
on subsequent calls for the same input regexp.
2023-01-09 21:43:08 -08:00
Aliaksandr Valialkin
67ab49baa9
vendor: make vendor-update 2023-01-09 21:34:34 -08:00
Aliaksandr Valialkin
e5eca54951
lib/promscrape/discovery/nomad: sync nomad_sd_configs fields with the Prometheus implementation
See the list of configs supported by Prometheus at f88a0a7d83/discovery/nomad/nomad.go (L76-L84)

- Removed "token" option. In can be set either via NOMAD_TOKEN env var or via `bearer_token` config option.
- Removed "scheme" option. It is automatically detected depending on whether the `tls_config` is set.
- Removed "services" and "tags" options, since they aren't supported by Prometheus.
- Added "region" option. If it is missing, then the region is read from NOMAD_REGION env var.
  If this var is empty, then it is set to "global" in the same way as Nomad client does.
  See 865ee8d37c/api/api.go (L297)
  and 865ee8d37c/api/api.go (L555-L556)
- If the "server" option is missing, then it is read from NOMAD_ADDR in the same way
  as Nomad client does - see 865ee8d37c/api/api.go (L294-L296)

This is a follow-up for 8aee209c53

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3367
2023-01-09 21:14:48 -08:00
Aliaksandr Valialkin
c38a10e143
app/vmselect/netstorage: eliminate memory allocation for sortBlocksHeap arg when calling mergeSortBlocks() 2023-01-09 21:08:51 -08:00
Aliaksandr Valialkin
1f9d605988
app/vmselect/netstorage: consistently select the sample with the biggest value out of samples with identical timestamps
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3333

This fix is based on https://github.com/VictoriaMetrics/VictoriaMetrics/pull/3620 ,
but doesn't slow down the common case with merging replicated data blocks so significantly.

Benchmark results:

Before the change:

BenchmarkMergeSortBlocks/replicationFactor-1-4         	   13968	     85643 ns/op	 956.53 MB/s	    1700 B/op	       1 allocs/op
BenchmarkMergeSortBlocks/replicationFactor-2-4         	   10806	    109171 ns/op	1500.77 MB/s	    2191 B/op	       1 allocs/op
BenchmarkMergeSortBlocks/replicationFactor-3-4         	    8887	    130623 ns/op	1881.45 MB/s	    2660 B/op	       1 allocs/op
BenchmarkMergeSortBlocks/replicationFactor-4-4         	    7440	    157348 ns/op	2082.52 MB/s	    3174 B/op	       1 allocs/op
BenchmarkMergeSortBlocks/replicationFactor-5-4         	    6534	    184473 ns/op	2220.38 MB/s	    3612 B/op	       1 allocs/op
BenchmarkMergeSortBlocks/overlapped-blocks-bestcase-4  	   13419	     85205 ns/op	 961.44 MB/s	    2213 B/op	       1 allocs/op
BenchmarkMergeSortBlocks/overlapped-blocks-worstcase-4 	     579	   1894900 ns/op	  43.23 MB/s	   46760 B/op	       1 allocs/op

After the change:

BenchmarkMergeSortBlocks/replicationFactor-1-4         	   13832	     85298 ns/op	 960.40 MB/s	    1716 B/op	       1 allocs/op
BenchmarkMergeSortBlocks/replicationFactor-2-4         	    8833	    134222 ns/op	1220.66 MB/s	    2675 B/op	       1 allocs/op
BenchmarkMergeSortBlocks/replicationFactor-3-4         	    6487	    184830 ns/op	1329.65 MB/s	    3636 B/op	       1 allocs/op
BenchmarkMergeSortBlocks/replicationFactor-4-4         	    4977	    236318 ns/op	1386.61 MB/s	    4733 B/op	       1 allocs/op
BenchmarkMergeSortBlocks/replicationFactor-5-4         	    4088	    296734 ns/op	1380.36 MB/s	    5761 B/op	       1 allocs/op
BenchmarkMergeSortBlocks/overlapped-blocks-bestcase-4  	   14083	     84067 ns/op	 974.47 MB/s	    2110 B/op	       1 allocs/op
BenchmarkMergeSortBlocks/overlapped-blocks-worstcase-4 	     536	   2043534 ns/op	  40.09 MB/s	   50511 B/op	       1 allocs/op
2023-01-09 13:01:48 -08:00
Denys Holius
fe0e199859
deployment/docker: update Alertmanager tag from v0.24.0 to v0.25.0 in docker-compose files (#3619)
deployment/docker: bump alertmanager to latest v0.25.0
2023-01-09 12:37:14 +01:00
Roman Khavronenko
8aee209c53
lib/promscrape: remove datacenter field from nomad_sd_config (#3612)
Looks like `datacenter` field isn't part of `/v1/services` API.
See https://developer.hashicorp.com/nomad/api-docs/services#list-services
and https://developer.hashicorp.com/nomad/api-docs/services#read-service

Related issues:
https://github.com/traefik/traefik/issues/9109
https://github.com/prometheus/prometheus/issues/11776

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2023-01-09 09:07:40 +01:00
Aliaksandr Valialkin
28f8dc41b0
lib/promscrape/discoveryutils: cleanup after 5df9fddaf2
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3468
2023-01-07 01:26:54 -08:00
Zakhar Bessarab
5df9fddaf2
lib/promscrape/discoveryutils: use correct timeout for blocking requests (#3609)
Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>

Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>
2023-01-07 01:13:03 -08:00
Aliaksandr Valialkin
41e00a0df7
lib/storage: simplify the fix from 488940502c
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3566
2023-01-07 01:04:43 -08:00
Dmytro Kozlov
488940502c
lib/storage: fix returning camelcase label names (#3608)
* lib/storage: fix returning camelcase label names

* doc: add change log

* Update docs/CHANGELOG.md

* Update docs/CHANGELOG.md

Co-authored-by: Aliaksandr Valialkin <valyala@victoriametrics.com>
2023-01-07 00:50:14 -08:00
Aliaksandr Valialkin
5fe7ff24c2
lib/streamaggr: limit the the number of concurrent flushes of the aggregate data to the exact number of available CPUs
This should reduce the maximum memory usage during concurrent flushes of the aggregate data
2023-01-07 00:18:51 -08:00
Aliaksandr Valialkin
ad5bfe3089
lib/promscrape: reduce the number of concurrently executed processScrapedData calls from 2x of the number of CPUs to the number of CPUs
This should reduce the maximum memory usage for processScrapedData() function by 2x.
The only part, which can be IO-bound in the processScrapedData() is pushData() call,
when it buffers data to persistent queue if the remote storage cannot keep up
with the data ingestion speed. In this case it is OK if the scrape pace will be limited.
2023-01-07 00:14:30 -08:00
Aliaksandr Valialkin
af263fe881
all: small improvements in error messages and command-line flag descriptions related to concurrency limiters 2023-01-07 00:11:44 -08:00
Aliaksandr Valialkin
45f39e291e
lib/writeconcurrencylimiter: moved the error generation from incConcurrency() to the caller place 2023-01-06 23:45:58 -08:00
Aliaksandr Valialkin
986a05e18d
lib/promscrape: limit the concurrency during parsing and relabeling the scraped samples
This should reduce memory usage when scraping big number of targets,
since this limits the summary memory usage during concurrent parsing and relabeling
by the number of available CPU cores.
2023-01-06 22:59:17 -08:00
Aliaksandr Valialkin
293e4dc77b
app/{vminsert,vmstorage}: add comments on why storage.AddRows() is called without limiting the number of concurrent calls 2023-01-06 22:40:07 -08:00
Aliaksandr Valialkin
5c4bd4f7c1
lib/streamaggr: limit the number of concurrent flushes of aggregate metrics in order to limit memory usage 2023-01-06 22:39:13 -08:00