Kiu/VictoriaMetrics
Kiu/VictoriaMetrics
1
0
Fork 0
mirror of https://github.com/VictoriaMetrics/VictoriaMetrics.git synced 2024-12-29 23:30:04 +01:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity
2,768 Commits 113 Branches 748 Tags 738 MiB
ede2ba5a45
Commit Graph

284 Commits

Author SHA1 Message Date
Aliaksandr Valialkin
e205975716 lib/promscrape: make a copy of ScrapeWork from discovered []ScrapeWork slice instead of referring to an item in this slice 
This should prevent from holding previously discovered []ScrapeWork slices when a part of discovered targets changes over time.
This should reduce memory usage for the case when big number of discovered scrape targets changes over time.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/825
 2020-11-10 16:13:05 +02:00
Aliaksandr Valialkin
6e668fd480 lib/promscrape: pre-allocate slice for discovered targets based on previously discovered targets 
This should reduce load on GC a bit when discovering big number of scrape targets
 2020-11-10 15:56:51 +02:00
Aliaksandr Valialkin
bcd139362b lib/promscrape: add -promscrape.dropOriginalLabels command-line flag for reducing memory usage when discovering big number of scrape targets 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/878
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/825
 2020-11-10 00:19:57 +02:00
Aliaksandr Valialkin
6c24c5caa3 lib/promscrape: further reduce memory usage for per-scrape target labels by making a copy of actually used labels 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/825
 2020-11-09 10:54:42 +02:00
Aliaksandr Valialkin
83df20b5b5 lib/promscrape: clean references to label name and label value strings after applying per-target relabeling 
This should reduce memory usage when per-target relabeling creates big number of temporary labels
with long names and/or values.

See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/825
 2020-11-07 16:19:59 +02:00
Aliaksandr Valialkin
9e83335ca9 lib/promscrape/discovery/kubernetes: go fmt 2020-11-07 13:03:49 +02:00
Aliaksandr Valialkin
5407eed2f6 lib/promscrape/discovery/kubernetes: reduce memory usage for labels when discovering big number of scrape targets by using string concatenation instead of fmt.Sprintf 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/825
 2020-11-07 13:03:08 +02:00
Aliaksandr Valialkin
188325f0fc lib/promscrape: eliminate data race in stream parse mode 
Previously `-promscrape.streamParse` mode could result in garbage labels for the scraped metrics because of data race.
See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/825#issuecomment-723198247
 2020-11-07 12:45:17 +02:00
Nikolay
4c808d58bf
Adds ready probe (#874) 
* adds leading forward slash check for scrapeURL path
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/835

* adds ready probe for scrape config initialization,
it should prevent metrics loss during vmagent rolling update,
/ready api will return 425 http code, if some scrape config still waits for initialization.

* updates docs

* Update app/vmagent/README.md

* renames var

* Update app/vmagent/README.md

Co-authored-by: Aliaksandr Valialkin <valyala@gmail.com>
 2020-11-04 20:29:18 +02:00
Aliaksandr Valialkin
b2042a1c30 lib/promscrape: docs update after e4182dd896 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/878
 2020-11-04 17:12:30 +02:00
Nikolay
e4182dd896
reduces memory usage for vmagent, (#880) 
* reduces memory usage for vmagent,
limits count of droppedTarget, that can be stored for /api/v1/targets page up to 999 items,
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/878

* Update app/vmagent/README.md

* Update app/vmagent/README.md

Co-authored-by: Aliaksandr Valialkin <valyala@gmail.com>
 2020-11-04 17:03:43 +02:00
Aliaksandr Valialkin
b9e5172aa2 Revert "lib/promscrape: add -promscrape.dropOriginalLabels command-line flag for reducing memory usage when discovering big number of scrape targets" 
This reverts commit 5a9743211f.
 2020-11-04 11:45:02 +02:00
Aliaksandr Valialkin
5a9743211f lib/promscrape: add -promscrape.dropOriginalLabels command-line flag for reducing memory usage when discovering big number of scrape targets 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/878
 2020-11-04 11:08:57 +02:00
Aliaksandr Valialkin
ca8b5745b5 lib/promscrape: reduce memory allocations in promLabelsString() function 
This should help with reducing memory usage in https://github.com/VictoriaMetrics/VictoriaMetrics/issues/878
 2020-11-04 10:38:44 +02:00
Aliaksandr Valialkin
c0f3be824d lib/promscrape: properly handle response body after 301 redirect 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/869
 2020-11-02 01:09:52 +02:00
Aliaksandr Valialkin
e277c3d07b lib/promscrape: add stream parse mode for efficient scraping of targets that expose millions of metrics 2020-11-01 23:35:06 +02:00
Aliaksandr Valialkin
b7638f04a7 app/vmagent: expose /api/v1/targets page according to https://prometheus.io/docs/prometheus/latest/querying/api/#targets 
This page is exposed by vmagent and by a single-node VictoriaMetrics

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/643
 2020-11-01 23:35:06 +02:00
Nikolay
67059caa12
fixes panic at scrape error body formating, (#868) 
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/864
regression after body reuse improvements
 2020-10-29 17:17:52 +03:00
Nikolay
de3fe22815
adds leading forward slash check for scrapeURL path (#855) 
* fixes in-consistency with prometheus behaviour for scrape targets url path.
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/835
 2020-10-29 08:39:42 +03:00
Nikolay Khramchikhin
e6ab69dd88
fixes openstack api endpoint with suffix trim adds openstack (#840) 
api v2.0 check
 2020-10-16 21:20:57 +03:00
Aliaksandr Valialkin
63c4999e06 lib/promscrape: code prettifying after 9bd9f67718 2020-10-12 16:12:36 +03:00
Nikolay Khramchikhin
9bd9f67718
Adds dockerswarm sd (#818) 
* adds dockerswarm service discovery

https://github.com/VictoriaMetrics/VictoriaMetrics/issues/656

 Following roles supported: services, tasks and nodes.
 Basic, token and tls auth supported.
 Added tests for labels generation.

* added unix socket support to discovery utils

Co-authored-by: Aliaksandr Valialkin <valyala@gmail.com>
 2020-10-12 13:38:21 +03:00
Aliaksandr Valialkin
561a7619a5 lib/promscrape: fix tests after 71ea4935de 2020-10-08 19:32:36 +03:00
Aliaksandr Valialkin
71ea4935de lib/promscrape: add -promscrape.suppressDuplicateScrapeTargetErrors command-line flag in order to suppress duplicate scrape target errors 
Show also original labels for duplicate targets in error message in order to simplify debugging the issue.

Now `/targets` endpoint accepts optional `show_original_labels=1` query arg, which shows original labels for each target.
This may simplify debugging for target relabeling.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/651
 2020-10-08 18:58:30 +03:00
Aliaksandr Valialkin
90aa2a8ffd lib/promscrape/discovery/openstack: show expiration time for refreshed OpenStack token in seconds - this is easier to interpret by human 2020-10-06 11:34:09 +03:00
Aliaksandr Valialkin
aba899c298 lib/promscrape/discovery/openstack: code prettifying after cbe3cf683b 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/728
 2020-10-05 18:11:55 +03:00
Nikolay Khramchikhin
cbe3cf683b
Adds openstack sd (#811) 
* adds openstack service discovery

https://github.com/VictoriaMetrics/VictoriaMetrics/issues/728

 implemented hypervisors and instance discovery with openstack v3 api.
 Added tests for labeling and data parsing.
 Added token refresh.

* Apply suggestions from code review

* Apply suggestions from code review

Co-authored-by: Aliaksandr Valialkin <valyala@gmail.com>
 2020-10-05 16:45:33 +03:00
Aliaksandr Valialkin
f42194d817 lib/promrelabel: make a copy of label with new name for action: labelmap in the same way as Prometheus does 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/812
 2020-10-05 16:19:19 +03:00
Aliaksandr Valialkin
2985077c35 all: consistently use "%w" formatting in fmt.Errorf for wrapped errors 2020-09-23 22:46:34 +03:00
Aliaksandr Valialkin
964bc7595c lib/promscrape/discovery/ec2: code prettifying after 312fead9a2 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/771
 2020-09-21 18:43:34 +03:00
Nikolay Khramchikhin
312fead9a2
Add improvements to ec2_sd_discovery (#775) 
* Add improvements to ec2 discovery

https://github.com/VictoriaMetrics/VictoriaMetrics/issues/771

 role_arn support with aws sts
 instance iam_role support
 refreshing temporary tokens

* Apply suggestions from code review

Co-authored-by: Roman Khavronenko <hagen1778@gmail.com>

* changed implementation, removed tests, clean up code

* moved endpoint builder into getEC2APIResponse

Co-authored-by: Roman Khavronenko <hagen1778@gmail.com>
 2020-09-21 16:04:15 +03:00
Aliaksandr Valialkin
1a9ee39b0e lib/promscrape: avoid copying response body when scraping targets. 
This should reduce memory usage when scraping targets with millions of metrics.
 2020-09-18 13:05:43 +03:00
Aliaksandr Valialkin
fa01169c3d lib/promscrape: add a link to troubleshooting docs to error message when duplicate scrape target with identical labels is skipped 2020-09-15 14:16:05 +03:00
Aliaksandr Valialkin
ba74d0c14c lib/promscrape: typo fix 2020-09-12 00:14:21 +03:00
Aliaksandr Valialkin
7d893a234c lib/promscrape: do not reset the remaining rows when pushing a part of data to remote storage during big scrapes 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/753

Thanks to @PerGon and @clmssz for help with debugging.
 2020-09-11 23:39:13 +03:00
Aliaksandr Valialkin
0e19f35af5 lib/promscrape/discovery/dns: add __meta_dns_srv_record_target and __meta_dns_srv_record_port labels 
This syncs dns service discovery with Prometheus 2.21 - see https://github.com/prometheus/prometheus/releases
and https://github.com/prometheus/prometheus/pull/7678 .
 2020-09-11 23:39:13 +03:00
Nikolay Khramchikhin
6c80ae0da8
Added endpointslices discovery to k8s api (#760) 
This is similar to https://github.com/prometheus/prometheus/pull/6838 , which will be added in Prometheus v2.21.
See https://github.com/prometheus/prometheus/releases/tag/v2.21.0-rc.1

* Added endpointslices discovery to k8s api

Started from 1.17 k8s version endpointslices is beta,
it allows to query k8s api for endpoints more efficient.
It presents at scrape_config.yaml as separate role for kubernetes_sd_config.
kubernetes_sd_config:
- role: endpointslices

* fixed typos, changed EndpointConditions signature - with values instead of pointers
 2020-09-11 12:16:45 +03:00
Aliaksandr Valialkin
038358b777 lib/promscrape: use the number of parsed rows as a basis for writeRequestCtxPool leveling 
The previous basis on `cap(sw.labels)` doesn't work anymore after 7785869ccc ,
because `sw.labels` may be reset multiple times when processing big number of rows.
 2020-09-02 18:46:01 +03:00
Aliaksandr Valialkin
b059f194e4 lib/promscrape: fix applying sample_limit when scraping targets with big number of metrics 
This has been broken at 7785869ccc
 2020-09-01 11:08:13 +03:00
Aliaksandr Valialkin
7785869ccc lib/promscrape: reduce memory usage when scraping targets with millions of metrics 
This should help when scraping /federate endpoints from Prometheus instances,
which scrape millions of metrics. See https://prometheus.io/docs/prometheus/latest/federation/
 2020-09-01 10:57:07 +03:00
Aliaksandr Valialkin
acbcad1ece lib/{promscrape,leveledbytebufferpool}: rename getPoolIdAndCapacity to getPoolIDAndCapacity in order to make golint happy 2020-08-28 09:49:32 +03:00
Aliaksandr Valialkin
efc730863b lib/promscrape: reduce memory usage when scraping targets with big number of metrics alongside targets with small number of labels 
Previously targets with big number of metrics and/or labels could generated too big buffers,
which then could be re-used when scraping targets with small number of metrics.
This resulted in memory waste.

Now big buffers are used only for targets with big number of metrics / labels,
while small buffers are used for targets with small number of metrics / labels.
 2020-08-16 22:29:51 +03:00
Aliaksandr Valialkin
147c35ebd4 all: allow using KB, MB, GB, KiB, MiB and GiB suffixes in command-line flag values related to byte sizes or byte rates 2020-08-16 17:05:52 +03:00
Aliaksandr Valialkin
e646674b23 lib/promscrape: use a hint on body length instead of body capacity 
This should reduce memory usage for https://github.com/VictoriaMetrics/VictoriaMetrics/issues/689
 2020-08-14 01:17:52 +03:00
Aliaksandr Valialkin
4628deecd1 lib/promscrape: reduce memory usage when scraping big number of targets 
Thanks to @dxtrzhang for the original idea at https://github.com/VictoriaMetrics/VictoriaMetrics/pull/688

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/689
 2020-08-14 01:04:53 +03:00
Aliaksandr Valialkin
eead3ee8ec lib/promscrape: properly retry requests on the server closed connection before returning the first response byte error during service discover API calls and target scrapes 2020-08-13 22:31:52 +03:00
Aliaksandr Valialkin
c402265e88 all: support %{ENV_VAR} placeholders in yaml configs in all the vm* components 
Such placeholders are substituted by the corresponding environment variable values.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/583
 2020-08-13 17:15:25 +03:00
Aliaksandr Valialkin
1bdfa29ef7 lib/promscrape: optimize per-metric hash calculations 
This increases vmagent performance by up to 10% when scraping big number of metrics
 2020-08-10 19:49:03 +03:00
Aliaksandr Valialkin
455bf50a91 lib/promscrape: show real timestamp and real duration for the scape on /targets page 
Previously the scrape duration may be negative when calculated scrape timestamp drifts away from the real scrape timestamp
 2020-08-10 12:40:25 +03:00
Aliaksandr Valialkin
a499de45cc lib/promscrape: make errcheck happy 2020-08-09 13:17:18 +03:00
Aliaksandr Valialkin
23c9e6b727 lib/promscrape: export scrape_samples_added per-target metric like Prometheus does 
This metric may be useful for detecting targets with high churn rate for the exported metrics.

See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/683
 2020-08-09 12:45:39 +03:00
Aliaksandr Valialkin
d4b6d22987 lib/promscrape: add a test for scrape config for blackbox exporter 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/684
 2020-08-09 12:02:48 +03:00
Roman Khavronenko
829ec4f9cf
Extend metric vm_promscrape_targets with status label (#615) 
The change to `vm_promscrape_targets` metric suppose to improve observability
for `vmagent` so it will be possible to track how many targets are up or down
for every specific scrape group:
```
vm_promscrape_targets{type="static_configs", status="down"} 1
vm_promscrape_targets{type="static_configs", status="up"} 2
```
 2020-07-13 21:52:03 +03:00
Aliaksandr Valialkin
3e1d7d8489 lib/promscrape: send Accept header similar to Prometheus when scraping targets 
This should fix scraping Spring Boot servers, which return incorrect response
unless `Accept: text/plain` request header is set.

See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/608
 2020-07-08 19:48:22 +03:00
Aliaksandr Valialkin
ff6a0955eb lib/promscrape: use HostClient.DoDeadline instead of HostClient.Do in order to guarantee strict deadline across multiple scrape attempts 2020-07-03 21:33:22 +03:00
Aliaksandr Valialkin
8b133e40d5 lib/promscrape: prevent from too big deadline misses on scrape retries 
The maximum deadline miss duration is reduced to 2x scrape_interval in the worst case.
By default it is limited to scrape_interval configured for the given scrape target.
 2020-07-03 20:41:36 +03:00
Aliaksandr Valialkin
44a54b8b3d lib/promscrape: check for nil error before checking for the returned status code when scraping targets 2020-07-03 18:37:14 +03:00
Aliaksandr Valialkin
4dd3de9286 lib/promscrape: add ability to set disable_compression and disable_keepalive options in scrape_config section of the config passed to -promscrape.config 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/580
 2020-07-02 14:19:14 +03:00
Aliaksandr Valialkin
8da3f773ae lib/promscrape: add -promscrape.disableKeepAlive command-line flag for disabling http keep-alive connections when scraping targets 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/580
 2020-07-01 02:20:20 +03:00
Aliaksandr Valialkin
d5dddb0953 all: use %w instead of %s for wrapping errors in fmt.Errorf 
This will simplify examining the returned errors such as httpserver.ErrorWithStatusCode .
See https://blog.golang.org/go1.13-errors for details.
 2020-06-30 23:05:11 +03:00
Aliaksandr Valialkin
586c5be404 lib/promscrape: add missing label sorting for autogenerated metrics 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/592
 2020-06-29 22:36:12 +03:00
Ween
1cd01b5359
Fix Auto metrics relabeled errors (#593) 
* Fix Auto metrics relabeled errors

* Finalize auto-genenated  Labels

* Fix Test Errors

Co-authored-by: xinyulong <xinyulong@kuaishou.com>
 2020-06-29 22:29:29 +03:00
Aliaksandr Valialkin
8f0bcec6cc lib/promscrape: preserve the previously discovered targets on discovery errors per each job_name 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/582
 2020-06-23 15:40:40 +03:00
Aliaksandr Valialkin
4c7f216dfe lib/promscrape: retry performing the request to the server for up to 3 times before giving up when it closes keep-alive connections 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/580
 2020-06-23 12:33:54 +03:00
Aliaksandr Valialkin
e149019c00 lib/promscrape/discovery/consul: reduce load on Consul when discovering big number of targets by using background caching 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/574
 2020-06-20 18:20:01 +03:00
Aliaksandr Valialkin
7bf2cbad32 lib/promscrape: reduce default value for -promscrape.discovery.concurrency from 500 to 100 
This should reduce load on Kubernetes API server and Consul when big number of targets are discovered

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/574
 2020-06-20 17:53:42 +03:00
Aliaksandr Valialkin
6ff821c70d lib/promscrape/discovery/ec2: expose __meta_ec2_ami like the next Prometheus release will do 
See b5d61fb66c for details
 2020-06-20 17:45:23 +03:00
Aliaksandr Valialkin
5820c0ffb7 lib/promrelabel: allows regex capture groups in target_label like Prometheus does 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/569
 2020-06-19 02:21:24 +03:00
Aliaksandr Valialkin
c602284a99 lib/promscrape: mention about -promscrape.maxScrapeSize in the error message when target returns too big response 2020-05-24 14:41:14 +03:00
Aliaksandr Valialkin
f9847352b4 app/vmagent: add -dryRun option for checking all the configs mentioned in command-line flags without running vmagent 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/362
 2020-05-21 15:23:27 +03:00
Aliaksandr Valialkin
d1a9d8aa1c lib/promscrape: add -promscrape.config.dryRun flag for checking -promscrape.config for errors or unsupported options 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/508
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/362
 2020-05-21 14:55:11 +03:00
Aliaksandr Valialkin
cf87b810b7 lib/promscrape: add -promscrape.discovery.concurrency and -promscrape.discovery.concurrentWaitTime flags for tuning the number of concurrent requests to autodiscovery API servers at Consul or Kubernetes 2020-05-19 17:35:53 +03:00
Aliaksandr Valialkin
0afd48d2ee lib: extract common code for returning fast unix timestamp into lib/fasttime 2020-05-14 23:02:07 +03:00
Aliaksandr Valialkin
0a134ace63 app/vmagent: fix scraping mTLS targets, which has been broken in v1.35.1 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/470
 2020-05-12 17:23:03 +03:00
Aliaksandr Valialkin
8300cc17af app/vmagent,lib/promscrape: do not set HostClient.DialDualStack, since it isnt used if HostClient.Dial is set 2020-05-12 15:24:18 +03:00
Aliaksandr Valialkin
68928bf3df lib/promscrape/discovery/gce: discover per-zone instances for gce_sd_config in parallel. This should reduce discovery latency 2020-05-06 15:00:09 +03:00
Aliaksandr Valialkin
3f52a97f9b lib/promscrape: add Prometheus-compatible DNS-based service discovery aka dns_sd_configs 2020-05-06 00:01:58 +03:00
Aliaksandr Valialkin
364789c24c lib/promscrape: properly connect to TCP6 addresses if -enableTCP6 is set 2020-05-06 00:01:57 +03:00
Aliaksandr Valialkin
7b5ef63384 lib/procutil: add NewSighupChan function, which returns a channel, which is triggered on every SIGHUP 2020-05-05 10:54:09 +03:00
Aliaksandr Valialkin
4fa817be10 lib/promscrape: allow explicitly setting empty token via token: "" in consul_sd_config 2020-05-05 07:50:15 +03:00
Aliaksandr Valialkin
40c3ffb359 lib/promscrape: add Prometheus-compatible service discovery for Consul aka consul_sd_configs 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/330
 2020-05-04 20:51:17 +03:00
Aliaksandr Valialkin
218e566647 lib/promscrape: move common code for discovery api config map handling into discoveryutils 2020-05-04 20:51:01 +03:00
Aliaksandr Valialkin
6310b20e72 lib/promscrape/discovery/kubernetes/: unify apiConfig creation 2020-05-04 20:50:49 +03:00
Aliaksandr Valialkin
66b0ae79a5 lib/promscrape: remove debug line left after the commit e4aac6ea40 2020-05-03 17:15:32 +03:00
Aliaksandr Valialkin
69004a5f67 lib/promscrape: fix tests after the commit 658a8742ac 
The original commit copies `__address__` label to `instance` label when generating per-target labels as Prometheus does.

See https://www.robustperception.io/life-of-a-label for details.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/453
 2020-05-03 16:56:15 +03:00
DexterZhang
658a8742ac
fix(vmagent): different behavior as how prometheus deal with labels. [Issue#453] (#454) 2020-05-03 16:51:03 +03:00
Aliaksandr Valialkin
e4aac6ea40 lib/promscrape: make consistent scrape time offsets across reloads for the same ScrapeURL and Labels 
This should make consistent intervals between data points for scrape targets across reloads.
Previously these intervals were random.
 2020-05-03 14:30:21 +03:00
Aliaksandr Valialkin
d9f1b4d6a3 lib/promscrape: fix TestGetFileSDScrapeWorkSuccess after 3b234d82e5 2020-05-03 14:28:31 +03:00
Aliaksandr Valialkin
3b234d82e5 lib/promscrape: reload only modified scrapers on config changes 
This should improve scrape stability when big number of targets are scraped and these targets are frequently changed.

Thanks to @xbsura for the idea and initial implementation attempts at the following pull requests:

- https://github.com/VictoriaMetrics/VictoriaMetrics/pull/449
- https://github.com/VictoriaMetrics/VictoriaMetrics/pull/458
- https://github.com/VictoriaMetrics/VictoriaMetrics/pull/459
- https://github.com/VictoriaMetrics/VictoriaMetrics/pull/460
 2020-05-03 12:45:40 +03:00
Aliaksandr Valialkin
5ec036439d lib/promscrape: set 30 seconds timeout for discovery api requests 
Previously such requests could hang for long time. This could make debugging harder.
 2020-04-29 17:33:34 +03:00
Aliaksandr Valialkin
43c39dc36c vendor: use github.com/VictoriaMetrics/fasthttp instead of github.com/fasthttp/fasthttp 
The upstream fasthttp may contain issues like 996610f021 ,
plus a code that isn't used by VictoriaMetrics. So let's use a private copy under our control instead.
 2020-04-29 17:33:34 +03:00
Aliaksandr Valialkin
521df0e2fc lib/promscrape: handle connection reset when targets responds with http redirect 2020-04-28 02:13:02 +03:00
肖贝贝
2b16c188e8
fix: vmagent not follow 301/302 redirect bug (#445) 
Co-authored-by: xiaobeibei <xiaobeibei@bigo.sg>
 2020-04-28 01:29:37 +03:00
Aliaksandr Valialkin
95942f1ac6 lib/promscrape/discovery/gce: make golangci-lint happy 2020-04-27 19:28:10 +03:00
Aliaksandr Valialkin
b768bc9a6a lib/promscrape: add initial support for Prometheus-compatible service discovery for Amazon EC2 aka ec2_sd_configs 2020-04-27 19:25:53 +03:00
Aliaksandr Valialkin
de59703a16 lib/promscrape/discovery/gce: properly set filter query arg in api url 2020-04-27 16:01:17 +03:00
Aliaksandr Valialkin
0224071ebe lib/promscrape/discovery/gce: allow empty project and zone for gce_sd_config 2020-04-27 11:45:02 +03:00
Aliaksandr Valialkin
6954d0edb7 lib/promscrape/discovery/gce: allow empty zone arg in gce_sd_config - in this case zones for the given project are automatically discovered 2020-04-26 14:34:11 +03:00
Aliaksandr Valialkin
7b8008e0bd lib/promscrape/discovery/gce: make golint happy by ignoring resp.Body.Close() result 2020-04-24 18:13:09 +03:00
Aliaksandr Valialkin
9ef5935552 lib/promscrape: initial implementation for gce_sd_configs aga Prometheus-compatible service discovery for Google Compute Engine 2020-04-24 17:51:22 +03:00
Aliaksandr Valialkin
24461153bf lib/promscrape: query /api/v1/namespaces/* for the configured namespaces in kubernetes_sd_config 
This should fix authroization issues described at https://github.com/VictoriaMetrics/VictoriaMetrics/issues/432
 2020-04-24 14:33:50 +03:00
Aliaksandr Valialkin
00e897119f lib/promscrape: add -promscrape.configCheckInterval command-line flag for automating config checking 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/431
 2020-04-23 23:41:08 +03:00
Aliaksandr Valialkin
a9a7a7175e lib/promscrape: access Config entries by reference, so they can be compared by addresses 2020-04-23 14:38:20 +03:00
Aliaksandr Valialkin
1c5d14a2eb lib/promscrape: move KubernetesSDConfig to lib/promscrape/discovery/kubernetes 2020-04-23 11:34:22 +03:00
Aliaksandr Valialkin
a714568374 lib/promscrape/discovery/kubernetes: hide role switch logic behind GetLabels function 2020-04-22 22:16:11 +03:00
Aliaksandr Valialkin
5454b518a6 lib/promscrape/discovery/kubernetes: reuse a client for empty api_server inside different jobs 2020-04-20 17:07:11 +03:00
Aliaksandr Valialkin
43375df923 lib/promscrape/discovery/kubernetes: update stale comments 2020-04-17 14:06:20 +03:00
Aliaksandr Valialkin
5d1537a395 lib/promscrape: suppress scrape errors if -promscrape.suppressScrapeErrors flag is set 2020-04-16 23:41:30 +03:00
Aliaksandr Valialkin
600490131f lib/promscrape: print all the labels for the target on error message for failed scrape 
This should improve debuggability.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/420
 2020-04-16 23:35:05 +03:00
Aliaksandr Valialkin
bd4c6d21dd lib/promscrape: retry target scraping when the target closes previously established keep-alive connection to it 
This should fix the following error:

the server closed connection before returning the first response byte. Make sure the server returns 'Connection: close' response header before closing the connection
 2020-04-16 23:25:29 +03:00
Aliaksandr Valialkin
426a0567c4 lib/promscrape: code cleanup in runScraper func 2020-04-15 11:36:24 +03:00
Aliaksandr Valialkin
067c7afebc lib/promscrape: show information on improperly configured scrape targets at the bottom of /targets page 
This is a common error whith improperly configured target autodiscovery and/or relabeling.
This error leads to duplicate scraping of the same targets with the same set of labels, which leads
to duplicate samples in time series.
 2020-04-14 14:55:05 +03:00
Aliaksandr Valialkin
ac35635b71 lib/promscrape/discovery/kubernetes: remove only unused client for API server during cleaning 2020-04-14 14:19:21 +03:00
Aliaksandr Valialkin
78863d7066 lib/promscrape: add promrelabel.GetLabelValueByName helper function 2020-04-14 14:12:01 +03:00
Aliaksandr Valialkin
c64f003cfb lib/promscrape: mention job name in error messages when target cannot be scraped 
This should improve debuggability
 2020-04-14 13:33:13 +03:00
Aliaksandr Valialkin
4718a5d951 lib/promscrape: reset ScrapeWork.ID in tests 2020-04-14 13:31:31 +03:00
Aliaksandr Valialkin
257521a634 lib/promscrape: properly expose statuses for targets with duplicate scrape urls at /targets page 
Previously targets with duplicate scrape urls were merged into a single line on the page.
Now each target with duplicate scrape url is displayed on a separate line.
 2020-04-14 13:10:01 +03:00
Aliaksandr Valialkin
6a75c95194 lib/promscrape: remove labels starting with __meta_ after applying relabel_configs as Prometheus does 
This should reduce CPU load during scraping when target discovery generates
big number of `__meta_*` labels (for instance, k8s discovery).

See https://www.robustperception.io/life-of-a-label for details.
 2020-04-14 12:23:22 +03:00
Aliaksandr Valialkin
01d7d799dc lib/promscrape: rename 'scrape_config->scrape_limit' to 'scrape_config->sample_limit' 
`scrape_config` block from Prometheus config contains `sample_limit` field,
while in `vmagent` this field was mistakenly named as `scrape_limit`.
 2020-04-14 11:59:57 +03:00
Aliaksandr Valialkin
2e4e202c2b lib/promscrape: add initial support for kubernetes_sd_config 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/334
 2020-04-13 21:03:28 +03:00
Aliaksandr Valialkin
2814b1490f lib/promscrape: add -promscrape.config.strictParse flag for detecting errors in -promscrape.config file 2020-04-13 13:15:44 +03:00
Aliaksandr Valialkin
90b4a6dd12 lib/promscrape: extract common auth code to lib/promauth 2020-04-13 12:59:10 +03:00
Aliaksandr Valialkin
3e55c7e069 lib/promscrape: reduce timestamp jitter when scraping targets 
This should improve compression for timestamps
 2020-04-01 16:11:35 +03:00
Aliaksandr Valialkin
499594f421 lib/promscrape: allow overriding external_labels as Prometheus does 
Prometheus docs at https://prometheus.io/docs/prometheus/latest/configuration/configuration/#scrape_config say:

> In communication with external systems, they are always applied only
> when a time series does not have a given label yet and are ignored otherwise.

Though this may result in consistency chaos when scrape targets override `external_labels`,
let's stick with Prometheus behavior for the sake of backwards compatibility.

There is last resort in vmagent with `-remoteWrite.label`, which consistently
sets the configured labels to all the metrics before sending them to remote storage.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/366
 2020-03-12 20:24:42 +02:00
Aliaksandr Valialkin
d4beb17ebe lib/promscrape: remove possible races when registering and de-registering scrape workers for /targets page 2020-03-11 16:30:21 +02:00
Aliaksandr Valialkin
cdf70b7944 lib/promscrape: consistently update /targets page after SIGHUP 2020-03-11 03:20:03 +02:00
Aliaksandr Valialkin
49d7cb1a3f all: fix golangci-lint issues 2020-03-10 19:41:46 +02:00
Aliaksandr Valialkin
7c432da788 lib/promscrape: do not retry idempotent requests when scraping targets 
This should prevent from the following unexpected side-effects of idempotent request retries:
- increased actual timeout when scraping the target comparing to the configured scrape_timeout
- increased load on the target

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/357
 2020-03-09 13:31:52 +02:00
Aliaksandr Valialkin
986dba5ab3 app/vmagent: do not allow non-supported fields in -remoteWrite.relabelConfig and file_sd_configs 
This should reduce possible confusion like in the https://github.com/VictoriaMetrics/VictoriaMetrics/issues/363
 2020-03-06 20:19:13 +02:00
Aliaksandr Valialkin
9a944fd169 lib/promscrape: consistency renaming: stopCh -> globalStopCh 2020-03-03 20:08:08 +02:00
Aliaksandr Valialkin
b63e4464f4 lib/promscrape: properly reload new configs on SIGHUP 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/335
 2020-02-26 13:54:00 +02:00
Aliaksandr Valialkin
6739c2749d lib/promscrape: go fmt 2020-02-25 20:56:44 +02:00
Aliaksandr Valialkin
7a33da8fea lib/promscrape: do not add missing port to __address__ label in order to be consistent with Prometheus behavior 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/331
 2020-02-25 20:49:50 +02:00
Aliaksandr Valialkin
04762344c6 app/vmagent: initial implementation for vmagent 2020-02-23 13:36:03 +02:00