Aliaksandr Valialkin
fc902734d9
lib/storage: further tuning for time series selector code
2021-03-15 20:32:37 +02:00
Aliaksandr Valialkin
dc70e7de76
lib/uint64set: reduce the size of bucket16 by storing smallPool by pointer.
...
This reduces CPU time spent on bucket16 copying inside bucket13.addBucketAtPos.
2021-03-15 17:27:03 +02:00
Aliaksandr Valialkin
cc6b1a5941
lib/uint64set: optimize Set.AddMulti for large sorted sets
2021-03-15 17:10:54 +02:00
Aliaksandr Valialkin
f023e92f1a
lib/uint64set: optimize bucket16.add and bucket16.addMulti a bit
2021-03-15 16:58:24 +02:00
Aliaksandr Valialkin
1c26020080
lib/storage: tune per-day index search
2021-03-15 13:36:36 +02:00
Aliaksandr Valialkin
7f52aae20c
lib/promscrape: an attempt to reduce memory usage when vmagent scrapes targets with varying number of metrics
...
Do not cache too big byte buffers and too big writeRequestCtx objects,
since it is cheaper to re-create them instead of wasting RAM for their caching.
This reverts 7f6f350ee1
2021-03-15 11:49:29 +02:00
Aliaksandr Valialkin
33cd6c26d3
lib/promscrape: return back the logic for flushing big buffers to storage from the commit 3fd8653b40
...
This should reduce memory usage when vmagent scrapes targets with big number of metrics and `-promscrape.streamParse` isn't enabled
2021-03-14 22:25:37 +02:00
Aliaksandr Valialkin
894246176f
lib/promscrape/discovery/kubernetes: do not start object watcher until initial objects are loaded
2021-03-14 21:56:16 +02:00
Aliaksandr Valialkin
9e55db4a53
lib/promscrape: retry service discovery in a few seconds if it starts returning 0 targets
...
This should reduce recovery time from temporary issues during service discovery
2021-03-14 21:56:16 +02:00
Aliaksandr Valialkin
3b46ae1c05
lib/promscrape: remove duplicate target word in error message
2021-03-14 21:56:16 +02:00
Aliaksandr Valialkin
b0b28eeb93
lib/promscrape/discovery/kubernetes: further optimize kubernetes service discovery for the case with many scrape jobs
...
Do not re-calculate labels per each scrape job - reuse them instead for scrape jobs with identical Kubernetes role
2021-03-14 21:16:41 +02:00
Aliaksandr Valialkin
620f05cd2c
lib/promscrape/discovery: fixes after 133b288681
...
- Removed a deadlock in addAPIWatcher
- Do not create unused ScrapeWork objects
- Do not spend CPU resources on creating objectByKey map in addAPIWatcher
This work is based on https://github.com/VictoriaMetrics/VictoriaMetrics/pull/1125
2021-03-13 15:22:38 +02:00
Aliaksandr Valialkin
54f902467d
lib/proxy: there is no need in cloning tlsCfg, which has been created two lines above
2021-03-12 10:48:01 +02:00
Aliaksandr Valialkin
72a8fa484b
lib/proxy: set proxy address in tls.Config.ServerName instead of the target address
...
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1116
2021-03-12 10:41:25 +02:00
Aliaksandr Valialkin
60e0280a94
lib/promscrape: add ability to configure proxy options via proxy_tls_config, proxy_basic_auth, proxy_bearer_token and proxy_bearer_token_file
...
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1116
2021-03-12 03:36:11 +02:00
Aliaksandr Valialkin
b2732575f7
lib/storage: further tune filters sorting logic
2021-03-12 00:51:35 +02:00
Aliaksandr Valialkin
8fc29ffc67
lib/promscrape/discovery/kubernetes: use a single watcher per apiURL
...
Previously multiple scrape jobs could create multiple watchers for the same apiURL. Now only a single watcher is used.
This should reduce load on Kubernetes API server when many scrape job configs use Kubernetes service discovery.
2021-03-11 17:04:14 +02:00
Aliaksandr Valialkin
8b8d4cbcfe
lib/proxy: do not show inline basic auth passwords when logging errors related to proxy_url
2021-03-11 13:44:14 +02:00
Aliaksandr Valialkin
41f641b132
lib/promscrape/discovery/kubernetes: localize Bookmark parsing code
...
This is a follow-up for e772d1c920
2021-03-11 13:08:56 +02:00
Aliaksandr Valialkin
6c9cd3f7c1
lib/promscrape/discovery/kubernetes: reduce load on Kubernetes API server by using watch bookmarks
...
This allows continuing object watch from the last bookbark instead of reloading all the objects
on watch errors or timeouts.
See https://kubernetes.io/docs/reference/using-api/api-concepts/#watch-bookmarks
2021-03-10 15:08:40 +02:00
Aliaksandr Valialkin
bd8b7a88a7
lib/httpserver: export vm_available_memory_bytes and vm_available_cpu_cores metrics
...
These metrics are useful for tracking the available memory and CPU cores for VictoriaMetrics apps.
2021-03-10 12:08:26 +02:00
Aliaksandr Valialkin
e15f3f4f2a
lib/proxy: pass proxy hostname in Host header of the CONNECT request
...
This should resolve the following issue when connecting to tls proxy:
cannot validate certificate for ... because it doesn't contain any IP SANs
2021-03-09 20:41:18 +02:00
Aliaksandr Valialkin
9d8223eafb
lib/proxy: set missing ServerName in TLS config for proxy_url.
...
While at it, allow setting Proxy-Authorization for `proxy_url` via `basic_auth` and `bearer_token` configs.
2021-03-09 19:01:14 +02:00
Nikolay
1310f84122
Changes tlsConfig init for proxy connections ( #1121 )
...
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1116
2021-03-09 19:01:13 +02:00
Aliaksandr Valialkin
0554430d7e
lib/promscrape: apply sample_limit after metric relabeling is applied as Prometheus does
...
See the description for `sample_limit` option from Prometheus docs:
Per-scrape limit on number of scraped samples that will be accepted.
If more than this number of samples are present after metric relabeling
the entire scrape will be treated as failed. 0 means no limit.
https://prometheus.io/docs/prometheus/latest/configuration/configuration/#scrape_config
2021-03-09 15:52:41 +02:00
Aliaksandr Valialkin
7b66c8cbf8
lib/promscrape/discovery/kubernetes: remove too verbose logs about starting and stopping the watchers
...
Log the number of objects loaded per each watch url
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1113
2021-03-09 15:07:12 +02:00
John Belmonte
edf39aa225
spelling fix: adjacent ( #1115 )
2021-03-09 09:19:16 +02:00
Aliaksandr Valialkin
502fab797a
lib/promscrape: add scrape_offset option to scrape_config
...
This option can be used for specifying the particular offset per each scrape interval for target scraping
2021-03-08 11:59:32 +02:00
Aliaksandr Valialkin
c4a0bd5eac
lib/storage: go fmt
2021-03-08 11:59:31 +02:00
Aliaksandr Valialkin
c76a904bb0
lib/storage: tune loopsCount estimations in getMetricIDsForTagFilterSlow
...
The adjusted estmations give up to 2x lower median response times on 200qps /api/v1/query_range workload
2021-03-07 21:17:48 +02:00
Aliaksandr Valialkin
c04505e585
lib/promscrape/discovery/kubernetes: reduce memory usage further when big number of scrape jobs are configured for the same kubernetes_sd_config role
...
Serialize reloading per-role objects, so they don't occupy too much memory when objects for many scrape jobs are simultaneously refreshed.
Do not reload per-role objects if they were already refreshed by concurrent goroutines. This should reduce load on Kubernetes API server
when big number of scrape jobs are configured for the same Kubernetes role.
This is a follow-up for 17b87725edhttps://github.com/VictoriaMetrics/VictoriaMetrics/issues/1113
2021-03-07 20:03:22 +02:00
Aliaksandr Valialkin
175466bb41
lib/decimal: prevent exponent overflow when processing values close to zero
...
See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1114
2021-03-05 18:53:41 +02:00
Aliaksandr Valialkin
5807ff57f3
lib/promscrape/discovery/kubernetes: reduce memory usage when Kubernetes service discovery is configured on a big number of scrape jobs
...
Previously vmagent was creating a separate Kubernetes object cache per each scrape job.
This could result in increased memory usage when monitoring a Kubernetes cluster with big number of objects (pods / nodes / services, etc.)
as seen at https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1113
Now it uses a shared map of scrape objects across multiple scrape jobs.
2021-03-05 17:32:33 +02:00
Aliaksandr Valialkin
92ddb8f197
lib/promscrape/discovery/kubernetes: move apiWatcher code to a separate file
2021-03-05 17:32:32 +02:00
Aliaksandr Valialkin
02c0959380
lib/promscrape: make cluster membership calculations consistent across 32-bit and 64-bit architectures
2021-03-05 09:06:08 +02:00
Aliaksandr Valialkin
133fb9fc00
lib/promscrape: add -promscrape.cluster.replicationFactor command-line flag for replicating scrape targets among vmagent instances in the cluster
2021-03-04 10:21:27 +02:00
Aliaksandr Valialkin
bae7a1b47a
lib/promscrape/discovery/kubernetes: fix tests after e154f4a644
2021-03-03 22:42:04 +02:00
Nikolay
7d92ef3acd
Fix ingress discovery api ( #1110 )
2021-03-03 10:45:50 +02:00
Aliaksandr Valialkin
25f453ce1a
lib/promscrape/discovery/kubernetes: properly check for nil pointer inside interface
...
See https://mangatmodi.medium.com/go-check-nil-interface-the-right-way-d142776edef1
This fixes a panic when the ScrapeWork is filtered out in swcFunc.
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1108
2021-03-03 10:42:54 +02:00
Aliaksandr Valialkin
fea27b9845
lib/promscrape: go fmt
2021-03-02 21:20:23 +02:00
Aliaksandr Valialkin
c67a07b469
lib/handshake: log read/write operation duration on connection errors
...
This improve debuggability of network errors
2021-03-02 21:20:20 +02:00
Aliaksandr Valialkin
c8dde1fd6b
lib/storage: typo fix: umarshal -> unmarshal
2021-03-02 20:48:44 +02:00
Aliaksandr Valialkin
3fbe2bf1c8
lib/promscrape: pre-allocate space for a map in mergeLabels
...
This should reduce the number of memory allocations when discovering big number of targets
2021-03-02 18:42:44 +02:00
Aliaksandr Valialkin
ac5c47a9f5
lib/promscrape/discovery: properly track vm_promscrape_discovery_kubernetes_objects_removed_total metric
2021-03-02 18:33:29 +02:00
Aliaksandr Valialkin
62d7e07ff7
lib/promrelabel: remove unneded optimizations for labeldrop and labelkeep actions
...
These optimizations may slow down code execution by matching the same label against regexp two times instead of a single time
2021-03-02 18:01:08 +02:00
Aliaksandr Valialkin
f9c1fe3852
lib/promscrape/discovery/kubernetes: cache ScrapeWork objects as soon as the corresponding k8s objects are changed
...
This should reduce CPU usage and memory usage when Kubernetes contains tens of thousands of objects
2021-03-02 16:44:19 +02:00
Aliaksandr Valialkin
f686174329
lib/promscrape/discovery/ec2: follow-up after f6114345de
2021-03-02 13:47:35 +02:00
Nikolay
fd8ca7df50
Adds webIndentity token for aws ( #1099 )
...
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1080
2021-03-02 13:47:34 +02:00
Aliaksandr Valialkin
e45c399467
lib/protoparser/prometheus: properly unescape label values in Prometheus exposition format
...
Unescape only `\n`, `\"` and `\\` sequences as Prometheus does. Other escape sequences shouldn't be unescaped.
2021-03-02 13:22:10 +02:00
Aliaksandr Valialkin
f4969a624d
lib/protoparser/graphite: fix parsing of a Graphite line with empty tags such as foo; 1 2
...
See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1100
2021-03-01 17:17:01 +02:00
Aliaksandr Valialkin
b89a4fac2f
lib/promscrape/discovery/kubernetes: deflake tests; a follow-up for 05fb08713c
2021-03-01 14:31:44 +02:00
Aliaksandr Valialkin
c3bf72992f
lib/promscrape: explicitly stop and cleanup service discovery routines when new config is read from -promscrape.config
...
This should reduce memory usage when `-promscrape.config` file frequently changes
2021-03-01 14:15:16 +02:00
Aliaksandr Valialkin
8af9370bf2
lib/promscrape: use target arg in ScrapeWork cache
2021-03-01 12:31:11 +02:00
Aliaksandr Valialkin
5328506c3e
lib/promscrape: typo fix, which prevented from caching ScrapeWork entries
2021-03-01 12:13:13 +02:00
Aliaksandr Valialkin
d5058caccc
lib/promscrape: add vm_promscrape_scrapework_cache_* metrics for tracking ScrapeWork cache effectiveness
2021-03-01 12:05:58 +02:00
Aliaksandr Valialkin
ee5d26a546
lib/httpserver: make make errcheck happy after the commit 9fc7726d84
2021-03-01 00:35:30 +02:00
Aliaksandr Valialkin
109bfaadad
lib/promscrape: reduce CPU usage an memory allocations when constructing scrapeWorkKey
2021-02-28 22:30:36 +02:00
Aliaksandr Valialkin
9e644ef111
lib/httpserver: make sure the gzipResponseWriter.Write() is called on Flush() and Close() calls
...
This should fix the `http: superfluous response.WriteHeader call` issue
See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1078
2021-02-28 19:23:26 +02:00
Aliaksandr Valialkin
9a2bf65134
lib/promscrape: add ability to spread scrape targets among multiple vmagent instances
...
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1084
2021-02-28 18:40:42 +02:00
Aliaksandr Valialkin
3e44d9947e
lib/promscrape/discovery/kubernetes: properly account the number of objects when watcher is stopped
...
A follow-up for b21b110b7a
2021-02-28 17:06:49 +02:00
Aliaksandr Valialkin
0ef7a94056
lib/promscrape/discovery/kubernetes: add vm_promscrape_discovery_kubernetes_* metrics for monitoring internal state of k8s service discovery
2021-02-28 16:58:45 +02:00
Aliaksandr Valialkin
f52bdbe2a3
lib/promscrape/discovery/kubernetes: remove resourceVersionMatch=NotOlderThan query arg when watching for k8s object changes, since it cannot be used when watch=1 query arg is passed
2021-02-28 16:08:43 +02:00
Aliaksandr Valialkin
6f8866a40a
lib/promscrape: fix possible deadlock in parallel execution of target relabeling
2021-02-28 16:05:32 +02:00
Aliaksandr Valialkin
5c9e657808
lib/promscrape/discovery/kubernetes: fix deadlock in startWatcherForURL
...
reloadObjects must be called without holding aw.mu lock
2021-02-28 15:25:33 +02:00
Aliaksandr Valialkin
e77f2f8630
lib/promscrape/discovery/kubernetes: typo fix after 241ffd1f3b
2021-02-28 15:15:27 +02:00
Aliaksandr Valialkin
82441537ff
lib/promscrape/discovery/kubernetes: pre-populate labelsByKey in reloadObject()
2021-02-28 15:09:43 +02:00
Aliaksandr Valialkin
e003453941
lib/promscrape/discovery/kubernetes: compare sorted sets of labels in tests
...
This should deflake tests where the order of labels isn't stable
2021-02-28 14:12:32 +02:00
Aliaksandr Valialkin
6a21ef87b7
lib/promscrape: add missing startWatchersForRole() call at the beginning of apiWatcher.getLabelsForRole
2021-02-28 14:00:00 +02:00
Aliaksandr Valialkin
6d0e7fb8b0
lib/promscrape/discovery/kubernetes: reload k8s resources on every error
...
This is needed for obtaining fresh resourceVersion
2021-02-27 01:46:59 +02:00
Aliaksandr Valialkin
7f1302688f
lib/fs: follow-up after f3a03c4164
2021-02-27 01:09:37 +02:00
Nikolay
d88fa5ebe4
Adds windows build ( #1040 )
...
* fixes windows compilation,
adds signal impl for windows,
adds free space usage for windows,
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/70
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1036
NOTE victoria metrics database still CANNOT work under windows system,
only vmagent is supported.
To completly port victoria metrics, you have to fix issues with separators,
parsing and posix file removall
* rollback separator
* Adds windows setInformation api,
it must behave like unix, need to test it.
changes procutil
* check for invlaid param
* Fixes posix delete semantic
* refactored a bit
* fixes openbsd build
* removed windows api call
* Fixes code after windows add
* Update lib/procutil/signal_windows.go
Co-authored-by: Aliaksandr Valialkin <valyala@gmail.com>
2021-02-27 01:06:22 +02:00
Aliaksandr Valialkin
44975e28fe
lib/promscrape: yet another typo fix after ed8441ec52
2021-02-26 23:36:34 +02:00
Aliaksandr Valialkin
c1b8729bd8
lib/fs: properly handle stale NFS file handle error during file deletion
...
This error can appear when -storageDataPath points to NFS volume and the given file has been already removed.
2021-02-26 23:24:46 +02:00
Aliaksandr Valialkin
50e74c439c
lib/promscrape: typo fix after ed8441ec52
2021-02-26 23:04:40 +02:00
Aliaksandr Valialkin
fa3ce450fb
lib/promscrape: cache ScrapeWork
...
This should reduce the time needed for updating big number of scrape targets.
2021-02-26 21:43:41 +02:00
Aliaksandr Valialkin
efcdf613c2
lib/promscrape/discovery/kubernetes: cache target labels
...
This should reduce CPU usage on repeated SDConfig.GetLabels() calls.
2021-02-26 20:24:29 +02:00
Aliaksandr Valialkin
22822feea3
lib/promscrape/discovery/kubernetes: errcheck fix
2021-02-26 19:09:12 +02:00
Aliaksandr Valialkin
dc8c045378
lib/promscrape: cleanup after 9b2246c29b
...
Main points:
* Revert changes outside lib/promscrape/discovery/kuberntes . These changes can be applied later in a separate commit
* Minimize changes in lib/promscrape/discovery/kubernetes compared to a93e644001
2021-02-26 19:09:12 +02:00
Nikolay
cf9262b01f
vmagent kubernetes watch stream discovery. ( #1082 )
...
* started work on sd for k8s
* continue work on watch sd
* fixes
* continue work
* continue work on sd k8s
* disable gzip
* fixes typos
* log errror
* minor fix
Co-authored-by: Aliaksandr Valialkin <valyala@gmail.com>
2021-02-26 19:09:12 +02:00
Aliaksandr Valialkin
4cfac70cde
lib/promscrape: remove duplicate code a bit
2021-02-26 15:54:07 +02:00
Aliaksandr Valialkin
55eb983193
lib/promscrape: reduce processing time for big number of discovered targets by processing them in parallel
2021-02-26 12:48:48 +02:00
Aliaksandr Valialkin
f8baf1a76d
lib/promrelabel: optimize labeldrop and labelkeep relabeling for prefix.* and prefix.+ regexps
2021-02-24 17:58:11 +02:00
Aliaksandr Valialkin
06676c8feb
lib/storage: consistency renaming: durationsPerDateTagFilterCache -> loopsPerDateTagFilterCache
2021-02-23 15:50:08 +02:00
faceair
b1409a7413
lib/storage: correct tagfilter match cost ( #1079 )
2021-02-22 21:54:37 +02:00
Aliaksandr Valialkin
197ecca426
lib/promrelabel: add more optimizations for relabeling for common cases
2021-02-22 16:36:54 +02:00
Aliaksandr Valialkin
63c16c3fdf
lib/promrelabel: optimize relabeling performance for common cases
2021-02-22 00:51:07 +02:00
Aliaksandr Valialkin
8b87398333
lib/promscrape: export vm_promscrape_target_relabel_duration_seconds metric
2021-02-21 23:21:35 +02:00
Aliaksandr Valialkin
587132555f
lib/mergeset: reduce memory usage for inmemoryBlock by using more compact items representation
...
This also should reduce CPU time spent by GC, since inmemoryBlock.items don't have pointers now,
so GC doesn't need visiting them.
2021-02-21 22:09:10 +02:00
Aliaksandr Valialkin
bd3bcdc43c
lib/storage: do not re-calculate stats for heavy tag filters
...
This should reduce the number of slow queries when stats for heavy tag filters was recalculated.
2021-02-21 21:43:37 +02:00
Aliaksandr Valialkin
b8a5ee2e93
lib/{mergeset,storage}: allow merging smaller number of small parts
...
While this may increase CPU and disk IO usage needed for background merge,
this also recudes CPU usage during queries in production. This is because
such queries tend to read recently added data and it is better to have lower number
of parts for such data in order to reduce CPU usage.
This partially reverts ebf8da3730
2021-02-21 21:43:37 +02:00
Aliaksandr Valialkin
34195218e1
lib/{mergeset,storage}: do not use pools for indexBlock and inmemoryBlock during their caching, since this results in higher memory usage in production without any performance gains
2021-02-21 21:43:37 +02:00
Aliaksandr Valialkin
f09b46b2b1
lib/promscrape: typo fix after the commit f26162ec99
2021-02-19 00:34:18 +02:00
Aliaksandr Valialkin
72eef964d9
app/vmagent: properly perform graceful shutdown, which was broken in the commit 1d1ba889fe
...
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1065
2021-02-19 00:34:17 +02:00
Aliaksandr Valialkin
502d0e2524
lib/promscrape: add scrape_align_interval config option into scrape config
...
This option allows aligning scrapes to a particular intervals.
2021-02-18 23:53:04 +02:00
Aliaksandr Valialkin
10ccb92e4d
lib/storage: use composite index for a query with a name filter and negative filters
2021-02-18 18:57:45 +02:00
Aliaksandr Valialkin
418de71509
lib/storage: properly handle queries containing a filter on metric name plus any number of negative filters and zero non-negative filters
...
Example: `node_cpu_seconds_total{mode!="idle"}`
2021-02-18 18:33:05 +02:00
Aliaksandr Valialkin
628b8eb55e
lib/storage: prevent from running identical heavy tag filters in concurrent queries when measuring the number of loops for such tag filter.
...
This should reduce CPU usage spikes when measuring the number of loops needed for heavy tag filters
2021-02-18 14:01:18 +02:00
Aliaksandr Valialkin
fd41f070db
lib/storage: sort tag filters by the number of loops they need for the execution
...
This metric should work better than the filter execution duration, since it cannot be distorted
by concurrently running queries.
2021-02-18 12:52:29 +02:00
Aliaksandr Valialkin
9566015a36
Revert "lib/mergeset: tune lifetime for entries inside block caches"
...
This reverts commit 458c89324d
2021-02-17 20:42:15 +02:00
Aliaksandr Valialkin
b4c7d5992b
lib/storage: move composite filters to the top during sorting
2021-02-17 20:26:32 +02:00
