Commit graph

1859 commits

Author SHA1 Message Date
Nikolay
c2380f73e6
lib/netutil: log only parsing errors for proxy-protocol (#3985)
* lib/netutil: log only parsing errors for proxy-protocol

Previosly every error was logged. With configured TCP health checks at load-balancer or kubernetes, vmauth spams a lot of false positive error message into logs

* Update docs/CHANGELOG.md

Co-authored-by: Roman Khavronenko <roman@victoriametrics.com>

* Update lib/netutil/tcplistener.go

Co-authored-by: Roman Khavronenko <roman@victoriametrics.com>

---------

Co-authored-by: Aliaksandr Valialkin <valyala@victoriametrics.com>
Co-authored-by: Roman Khavronenko <roman@victoriametrics.com>
2023-03-24 22:14:32 -07:00
Aliaksandr Valialkin
e7c20f0c75
lib/{mergeset,storage}: prevent from long wait time when creating a snapshot under high data ingestion rate
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3551
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/pull/3873
2023-03-24 22:12:08 -07:00
Dmytro Kozlov
67a526f5f7
lib/storage: fix collect downsampling metrics (#489)
* lib/storage: fix downsampling

* lib/storage: update logic

* lib/storage: fix comments, removed unneeded check
2023-03-19 23:34:33 -07:00
Nikolay
ed2e4685d9
lib/vmselectapi: fixes regression for disable compression setting (#3932)
after vmselect api refactoring it wasn't possible to disable response cache.
This patch restores correct behavior for rpc.disableCompression flag
2023-03-12 01:59:38 -08:00
Nikolay
ea038a8940
lib/storage: correctly handle io.EOF error for pre-fetched metrics (#3946)
io.EOF shouldn't be returned from this function. It breaks all search
API logic and may result in empty query results.
2023-03-12 01:29:05 -08:00
Nikolay
ce950d3b06
lib{mergset,storage}: prevent possible race condition with logging st… (#3900)
lib{mergset,storage}: prevent possible race condition with logging stats for merges

Previously partwrapper could be release by background process and reference for part may be invalid
during logging stats. It will lead to panic at vmstorage
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3897
2023-03-12 01:23:55 -08:00
Nikolay
c80d0aaaf0
lib/netutil: fixes panic at proxy protocol (#3905)
it may occur if non proxy protocol message received by tcp server.
Listener Accept method must return only non-recoverable errors.
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3335
2023-03-12 01:12:53 -08:00
Zakhar Bessarab
3f57d4d489
lib/promscrape: correctly register vm_promscrape_config_* metrics (#3876)
* lib/promscrape: set `vm_promscrape_config_last_reload_successful` to 1 if there was no promscrape config provided

Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>

* lib/promscrape: register `vm_promscrape_config_*` metrics only in case promscrape config is used

Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>

---------

Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>
Co-authored-by: Aliaksandr Valialkin <valyala@victoriametrics.com>
2023-02-27 12:06:22 -08:00
Zakhar Bessarab
c9547f4fba
lib/{fs,mergeset,storage}: skip .must-remove. dirs when creating snapshot (#3858) (#3867) 2023-02-24 12:44:35 -08:00
Aliaksandr Valialkin
95b66f3de5
lib/promscrape: follow-up for 43e104a83f
- Return immediately on context cancel during the backoff sleep.
  This should help with https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3747

- Add a comment describing why the second attempt to obtain the response from remote side
  is perfromed immediately after the first attempt.

- Remove fasthttp dependency from lib/promscrape/discoveryutils

- Set context deadline before calling doRequestWithPossibleRetry().
  This simplifies the doRequestWithPossibleRetry() a bit.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3293
2023-02-24 12:26:33 -08:00
Zakhar Bessarab
3f64dae46b
fix: do not use exponential backoff for first retry of scrape request (#3824)
* fix: do not use exponential backoff for first retry of scrape request (#3293)

* lib/promscrape: refactor `doRequestWithPossibleRetry` backoff to simplify logic

Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>

* Update lib/promscrape/client.go

Co-authored-by: Roman Khavronenko <roman@victoriametrics.com>

* lib/promscrape: refactor `doRequestWithPossibleRetry` to make it more straightforward

Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>

---------

Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>
Co-authored-by: Roman Khavronenko <roman@victoriametrics.com>
2023-02-24 12:26:32 -08:00
Mattias Ängehov
13d0a2a3e7
Azure Service Discovery - Fix token fetch for Container Apps/App Services (#3832)
* Modify API version when running in Container App

* Handle expires on from token response

Response from IMDS does not always contain expires in value which is
currently used to get the token expiry time. An example resources that
doesn't provide it are Container Apps and App Service.

Signed-off-by: Mattias Ängehov <mattias.angehov@castoredc.com>

* Fix client id parameter for user assigned identity

* Apply suggestions from code review

---------

Signed-off-by: Mattias Ängehov <mattias.angehov@castoredc.com>
Co-authored-by: Aliaksandr Valialkin <valyala@gmail.com>
2023-02-23 19:30:23 -08:00
Zakhar Bessarab
3dfb9e732c
lib/promscrape: fix cancelling in-flight scrape requests during configuration reload (#3853)
* lib/promscrape: fix cancelling in-flight scrape requests during configuration reload (see #3747)

Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>

* lib/promscrape: fix order of params for `doRequestWithPossibleRetry` to follow codestyle

Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>

* lib/promscrape: accept deadline explicitly and extend passed context for local use

Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>

---------

Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>
2023-02-22 18:28:28 -08:00
Droxenator
732dd5357a
fixed opentsdbListenAddr timestamp conversion (#3810)
Co-authored-by: Andrei Ivanov <a.ivanov@corp.mail.ru>
2023-02-14 22:47:45 -08:00
Aliaksandr Valialkin
ddd8d01cf5
lib/{mergeset,storage}: allow at least 3 concurrent flushes during background merges on systems with 1 or 2 CPU cores
This should prevent from data ingestion slowdown and query performance degradation
on systems with small number of CPU cores (1 or 2), when big merge is performed.

This should help https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3790

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3337
2023-02-11 12:17:46 -08:00
Zakhar Bessarab
bbf663bd04
lib/promscrape: fix cancelling in-flight scrape requests during configuration reload (#3791)
* lib/promscrape: fix cancelling in-flight scrape requests during configuration reload when using `streamParse` mode (see #3747)

Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>

* Update docs/CHANGELOG.md

---------

Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>
Co-authored-by: Aliaksandr Valialkin <valyala@victoriametrics.com>
2023-02-09 11:18:36 -08:00
Aliaksandr Valialkin
146b3bd088
lib/backup/azremote: fix after upgrading github.com/Azure/azure-sdk-for-go/sdk/storage/azblob from v0.6.1 to v1.0.0 2023-02-08 09:19:10 -08:00
Karan Sharma
004a24c950
sd/nomad: panic in nomad watcher because of nil map (#3784)
properly initialize url.Values
2023-02-08 08:37:02 -08:00
Aliaksandr Valialkin
f5595233c2
lib/writeconcurrencylimiter: initialize concurrencyLimitCh before exporting vm_concurrent_insert_capacity and vm_concurrent_insert_current metrics
This will result in proper calculations for the the alerting rule:

 avg_over_time(vm_concurrent_insert_current[1m]) >= vm_concurrent_insert_capacity

See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3761
2023-02-07 11:08:39 -08:00
Aliaksandr Valialkin
ac695f36bb
lib/promscrape: add a comment explaining the logic behind adding exported_ perfix to metric names
This is a follow-up for 7b87fac8e7

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3557
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3406
2023-02-01 12:02:05 -08:00
Dmytro Kozlov
3c1e455805
lib/promscrape: fix honor_labels behavior (#3739)
Co-authored-by: Aliaksandr Valialkin <valyala@victoriametrics.com>
2023-02-01 12:02:04 -08:00
Nikolay
554876cc38
lib/storage: fixes finalDedup for backfilled data (#3737)
previously historical data backfilling may trigger force merge for previous month every hour
it consumes cpu, disk io and decrease cluster performance.
Following commit fixes it by applying deduplication for InMemoryParts
2023-02-01 09:57:02 -08:00
Aliaksandr Valialkin
a522bbc8b4
lib/bytesutil/internstring.go: increase the limit on the maximum string lengths, which can be interned
The limit has been increased from 300 bytes to 500 bytes according to the collected production stats.
This allows reducing CPU usage without significant increase of RAM usage in most practical cases.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3692
2023-01-31 11:04:09 -08:00
Aliaksandr Valialkin
855d560789
lib/promscrape/discovery/azure: add __meta_azure_machine_size label in the same way as Prometheus does
See https://github.com/prometheus/prometheus/pull/11650
2023-01-27 17:07:57 -08:00
Aliaksandr Valialkin
134f7622d6
lib/promscrape/discovery/kubernetes: add support for __meta_kubernetes_pod_container_id
See https://github.com/prometheus/prometheus/issues/11843
and https://github.com/prometheus/prometheus/pull/11844
2023-01-27 16:33:57 -08:00
Aliaksandr Valialkin
bccbe07c33
lib/netutil: move IsTrivialNetworkError() function there, since it is used in multiple places across the code 2023-01-27 13:24:44 -08:00
Aliaksandr Valialkin
eb10102521
lib/netutil: typo fix in the error message 2023-01-27 11:31:05 -08:00
Aliaksandr Valialkin
b17857c7a4
lib/netutil: limit the time needed for reading proxy protocol headers
This should prevent from misconfigured proxies and from possible Slowloris-type DoS attacks
(see https://en.wikipedia.org/wiki/Slowloris_(computer_security) )

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3335
2023-01-26 23:47:06 -08:00
Nikolay
ebebaecd94
lib/netutil: init implimentation of proxy protocol (#3687)
* lib/netutil: init implimentation of proxy protocol
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3335

* wip

Co-authored-by: Aliaksandr Valialkin <valyala@victoriametrics.com>
2023-01-26 23:25:22 -08:00
Nikolay
4af05065d1
lib/storage: properly release parts inMerge lock (#3711)
if storage doesn't have enough disk space, finalDedupWatcher holds inMerge lock for all parts and never release it until storage restart
2023-01-26 08:57:36 -08:00
Aliaksandr Valialkin
5defa99a2e
lib/streamaggr: add ability to de-duplicate input samples before aggregation 2023-01-25 09:22:03 -08:00
Roman Khavronenko
dad25672e2
discover/ec2: bump API version (#3702)
Switch to the actual API version `2016-11-15`,
since the old version doesn't provide access to all
the fields which implementation expects.
For example, old API missing `zone_id` field
in `DescribeAvailabilityZonesResponse` response.

See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3700

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

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2023-01-24 09:10:45 -08:00
Aliaksandr Valialkin
0698467ae5
lib/bytesutil: do not intern long strings, since they may need big amounts of additional memory for the cache
Allow users fine-tuning the maximum string length for interning via -internStringMaxLen command-line flag.
This may be used for fine-tuning RAM vs CPU usage for certain workloads.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3692
2023-01-23 23:37:08 -08:00
Aliaksandr Valialkin
4b3a207705
app/{vmagent,vminsert}: follow-up for 1cfa183c2b
- Call httpserver.GetQuotedRemoteAddr() and httpserver.GetRequestURI() only when the error occurs.
  This saves CPU time on fast path when there are no parsing errors.
- Create a helper function - httpserver.LogError() - for logging the error with the request uri and remote addr context.
2023-01-23 22:41:08 -08:00
Artem Navoiev
0ac0cfdc69
add error handler for parsing prometheus text format to vmagent and v… (#3693)
* add error handler for parsing prometheus text format to vmagent and vminsert

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

* fix typo

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

* typo

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

* fix variables naming and error message

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

Signed-off-by: Artem Navoiev <tenmozes@gmail.com>
2023-01-23 22:36:23 -08:00
Aliaksandr Valialkin
71a170d404
lib/promscrape: follow-up for 393876e52a
- Document the change in docs/CHANGELOG.md
- Reduce memory usage when sending stale markers even more by parsing the response in stream parsing mode
- Update the TestSendStaleSeries

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3668
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3675
2023-01-23 21:56:18 -08:00
Roman Khavronenko
8e2a8a6ae2
lib/promscrape: limit number of sent stale series at once (#3686)
Stale series are sent when there is a difference between current
and previous scrapes. Those series which disappeared in the current scrape
are marked as stale and sent to the remote storage.

Sending stale series requires memory allocation and in case when too many
series disappear in the same it could result in noticeable memory spike.
For example, re-deploy of a big fleet of service can result into
excessive memory usage for vmagent, because all the series with old
pod name will be marked as stale and sent to the remote write storage.

This change limits the number of stale series which can be sent at once,
so memory usage remains steady.

https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3668
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3675
Signed-off-by: hagen1778 <roman@victoriametrics.com>

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2023-01-23 21:56:17 -08:00
Aliaksandr Valialkin
95d4db0506
lib/promscrape: properly log the actual response size after c4229a1bba 2023-01-23 21:13:06 -08:00
Aliaksandr Valialkin
903b2e710c
lib/storage: use deterministic random generator in tests
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3683
2023-01-23 20:12:32 -08:00
Aliaksandr Valialkin
4c7062b408
lib/mergeset: use deterministic random generator in tests
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3683
2023-01-23 19:44:10 -08:00
Aliaksandr Valialkin
f8dcbe4abd
lib/mergeset: fix data race in BenchmarkInmemoryBlockMarshal 2023-01-23 19:44:07 -08:00
Aliaksandr Valialkin
107a056ade
lib/decimal: use consistent randomizer in tests
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3683
2023-01-23 19:24:05 -08:00
Aliaksandr Valialkin
796c7b0ee1
lib/uint64set: use repeatable randomizer in tests
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3683
2023-01-23 19:24:05 -08:00
Aliaksandr Valialkin
dfb1d1ead1
lib/encoding: make deterministic tests which rely on math/rand
Fixes https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3683
2023-01-23 18:43:49 -08:00
Aliaksandr Valialkin
d8329e47cf
lib/vmselectapi: propagate timeout errors from vmselect to vmstorage instead of closing the connection established from vmselect to vmstorage
This is a follow-up for 20e9598254
2023-01-20 19:30:22 -08:00
Tobias Jungel
8bdc63aab9
app/vmbackup: prevent password leaks (#3672)
This prevents vmbackup from leaking passwords into logs like shown below.

2023-01-11T15:00:01.050Z        info    VictoriaMetrics/lib/logger/flag.go:12   build version: vmbackup-20221214-211706-tags-v1.85.1-0-g09a70d3e9
2023-01-11T15:00:01.050Z        info    VictoriaMetrics/lib/logger/flag.go:13   command-line flags
2023-01-11T15:00:01.050Z        info    VictoriaMetrics/lib/logger/flag.go:20     -dst="fs:///vm-backups/latest"
2023-01-11T15:00:01.050Z        info    VictoriaMetrics/lib/logger/flag.go:20     -snapshot.createURL="http://user:super_sercret123@victoriametricspshot/create"
2023-01-11T15:00:01.050Z        info    VictoriaMetrics/lib/logger/flag.go:20     -storageDataPath="/storage"
2023-01-11T15:00:01.050Z        info    VictoriaMetrics/app/vmbackup/main.go:53 Snapshot create url http://user:super_sercret123@victoriametrics:8428/snapshot/create
2023-01-11T15:00:01.050Z        info    VictoriaMetrics/app/vmbackup/main.go:60 Snapshot delete url http://user:super_sercret123@victoriametrics:8428/snapshot/delete
2023-01-18 11:40:52 -08:00
Aliaksandr Valialkin
c5e858461c
lib/{storage,mergeset}: wake up background merges as soon as there is a potential work for them
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3647
2023-01-18 01:10:43 -08:00
Aliaksandr Valialkin
70b5a6fb28
lib/{storage,mergeset}: do not run assisted merges when flushing pending samples to parts
Assisted merges are intended to be performed by goroutines, which accept the incoming samples,
in order to limit the data ingestion rate.

The worker, which converts pending samples to parts, shouldn't be penalized by assisted merges,
since this may result in increased number of pending rows as seen at https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3647#issuecomment-1385039142
when the assisted merge takes too much time.
2023-01-18 00:25:33 -08:00
Aliaksandr Valialkin
0c90b49e4b
lib/storage: use better naming for a function returning new []rawRows - newRawRowsBlock() -> newRawRows() 2023-01-18 00:01:21 -08:00
Aliaksandr Valialkin
a844b97942
lib/promscrape: follow-up for d79f1b106c
- Document the fix at docs/CHANGELOG.md
- Limit the concurrency for sendStaleMarkers() function in order to limit its memory usage
  when big number of targets disappear and staleness markers are sent
  for all the metrics exposed by these targets.
- Make sure that the writeRequestCtx is returned to the pool
  when there is no need to send staleness markers.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3668
2023-01-17 23:13:08 -08:00