mudhorn/prometheus
1
0
Fork 0
mirror of https://github.com/prometheus/prometheus.git synced 2025-03-05 20:59:13 -08:00
Code Issues Actions 4 Packages Projects Releases Wiki Activity
12663 commits 313 branches 456 tags 672 MiB
Commit graph

12663 commits

This branch
This branch
All branches
Author SHA1 Message Date
Marco Pracucci a5d218d53d
Configure the 'high eviction' benchmark to always evict 
Signed-off-by: Marco Pracucci <marco@pracucci.com>
 2023-09-28 11:12:40 +02:00
Marco Pracucci d2f7ecbaba
Merge pull request #534 from grafana/label-values-with-matchers-is-bypassing-cache 
`LabelValues()` with matchers should use cache
 2023-09-28 10:47:45 +02:00
Marco Pracucci 8d92930140
Make the benchmark more realistic, returning some postings 
Signed-off-by: Marco Pracucci <marco@pracucci.com>
 2023-09-28 10:46:22 +02:00
Marco Pracucci 9bcca44ac4
Addressed review comments 
Signed-off-by: Marco Pracucci <marco@pracucci.com>
 2023-09-28 10:32:19 +02:00
Oleg Zaytsev a8c31f279f
Test that IndexReader.PostingsForMatchers is called 
Signed-off-by: Oleg Zaytsev <mail@olegzaytsev.com>
 2023-09-28 10:22:37 +02:00
Matthieu MOREL 67dcca5005 ci(lint): enable errorlint linter on cmd 
Signed-off-by: Matthieu MOREL <matthieu.morel35@gmail.com>
 2023-09-28 08:21:01 +00:00
Linas Medziunas 3c047a3518 Expand docs comments 
Signed-off-by: Linas Medziunas <linas.medziunas@gmail.com>
 2023-09-28 09:08:09 +03:00
Linas Medziunas dfb6292600 Compare FloatHistogram.[Zero]Count float values as binary 
Signed-off-by: Linas Medziunas <linas.medziunas@gmail.com>
 2023-09-28 09:06:54 +03:00
Oleg Zaytsev 664c125d87
LabelValues() with matchers should use cache 
When cache was introduced, LabelValues() could never be called with
"concurrent" flag so it didn't make sense to use the cached call through
the Head. However, since the introduction of forced cache, we should use
it, as even with concurrent=false the cache may be used.

Signed-off-by: Oleg Zaytsev <mail@olegzaytsev.com>
 2023-09-27 18:33:18 +02:00
Marco Pracucci 2986a7a4ba
Added BenchmarkPostingsForMatchersCache 
Signed-off-by: Marco Pracucci <marco@pracucci.com>
 2023-09-27 16:17:08 +02:00
Marco Pracucci 0a8b79ad29
Do not measure size of sync.WaitGroup 
Signed-off-by: Marco Pracucci <marco@pracucci.com>
 2023-09-27 15:50:39 +02:00
Marco Pracucci e2c1e7aadc
Fix typo 
Signed-off-by: Marco Pracucci <marco@pracucci.com>
 2023-09-27 15:33:15 +02:00
Marco Pracucci 3c68ce252e
Add PostingsForMatchers cache size by bytes support 
Signed-off-by: Marco Pracucci <marco@pracucci.com>
 2023-09-27 15:25:39 +02:00
Björn Rabenstein 0de7f39e6a
Merge pull request #12894 from linasm/linasm/test-case-for-ValidateHistogram 
Additional test case for ValidateHistogram
 2023-09-27 14:16:57 +02:00
Linas Medziunas 1aad4004c3 Additional test case for ValidateHistogram 
Signed-off-by: Linas Medziunas <linas.medziunas@gmail.com>
 2023-09-27 09:34:43 +03:00
Julien Pivotto a3f424641a
Merge pull request #6724 from peergynt/template_reference 
Added externalURL and pathPrefix functions to Template reference docs
 2023-09-26 13:50:33 +02:00
Julien Pivotto a8cdb7eb6c
Merge branch 'main' into template_reference 
Signed-off-by: Julien Pivotto <roidelapluie@o11y.eu>
 2023-09-26 13:49:58 +02:00
Bryan Boreham 6dcbd653e9
tsdb: register metrics after Head is initialized (#12876) 
This avoids situations where metrics are scraped before the data they
are trying to look at is initialized.

Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
 2023-09-25 21:57:08 +01:00
Alan Protasio a15e884e7a
Prevent Prometheus from overallocating memory on subquery with large amount of steps. (#12734) 
* change initial points slice size

Signed-off-by: Alan Protasio <alanprot@gmail.com>

* refactor on the steps calculation and moving the getXPoint/putXPoint method to the evaluator

Signed-off-by: Alan Protasio <alanprot@gmail.com>

* prevent potential panic

Signed-off-by: Alan Protasio <alanprot@gmail.com>

* Update promql/engine.go

Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
Signed-off-by: Alan Protasio <alanprot@gmail.com>

* Update promql/engine.go

Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
Signed-off-by: Alan Protasio <alanprot@gmail.com>

* Update promql/engine.go

Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
Signed-off-by: Alan Protasio <alanprot@gmail.com>

* Update promql/engine.go

Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
Signed-off-by: Alan Protasio <alanprot@gmail.com>

* Allocating slice with maximum size of 5k

Signed-off-by: Alan Protasio <alanprot@gmail.com>

* adding comments

Signed-off-by: Alan Protasio <alanprot@gmail.com>

---------

Signed-off-by: Alan Protasio <alanprot@gmail.com>
Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
 2023-09-25 20:15:41 +01:00
Linas Medziunas cbd01fc296 Fix NaN sum check in [Float]Histogram.Equals method 
Signed-off-by: Linas Medziunas <linas.medziunas@gmail.com>
 2023-09-25 16:03:55 +03:00
ouyang1204 5d233df7ef
Fix rule check broken (#12715) 
Signed-off-by: DrAuYueng <ouyang1204@gmail.com>
 2023-09-25 17:48:05 +10:00
Goutham Veeramachaneni 86729d4d7b
Update exp package (#12650) 2023-09-21 22:53:51 +02:00
Björn Rabenstein f8dd8770ac
Merge pull request #12757 from bboreham/reuse-bufiter 
TSDB: re-use iterator when moving between series
 2023-09-21 14:08:53 +02:00
Björn Rabenstein ade2ee3fe4
Merge pull request #12716 from bboreham/simplify-seek 
storage: simplify Seek on BufferedSeriesIterator
 2023-09-21 13:56:29 +02:00
Dimitar Dimitrov 1155d736b6
Improve sensitivity of TestQuerierIndexQueriesRace 
Currently, the two goroutines race against each other and it's possible that the main test goroutine finishes way earlier than appendSeries has had a chance to run at all.

I tested this change by breaking the code that X fixed and running the race test 100 times. Without the additional time.Sleep the test failed 11 times. With the sleep it failed 65 out of the 100 runs. Which is still not ideal, but it's a step forward.

Signed-off-by: Dimitar Dimitrov <dimitar.dimitrov@grafana.com>
 2023-09-21 12:30:08 +02:00
Arve Knudsen 320f0c9c4a
Merge pull request #531 from grafana/chore/sync-prometheus 
Sync with latest Prometheus upstream
 2023-09-21 10:11:26 +02:00
Arve Knudsen 85df1d4013 Merge remote-tracking branch 'prometheus/main' into chore/sync-prometheus 
Signed-off-by: Arve Knudsen <arve.knudsen@gmail.com>
 2023-09-21 09:19:23 +02:00
Bryan Boreham 91054875d6
Merge pull request #12732 from bboreham/simplify-rangeeval 
promql: simplify inner loop of rangeEval
 2023-09-20 20:22:05 +00:00
Bryan Boreham d498bb0d8c
Merge pull request #12875 from dimitarvdimitrov/dimitar/TestQuerierIndexQueriesRace-incorrect-exit-condition 
Fix samples generation in TestQuerierIndexQueriesRace

The loop would run forever if the context was cancelled before the goroutine got started.
 2023-09-20 20:20:30 +00:00
Dimitar Dimitrov 6f1284ac93 Fix exit condition of TestQuerierIndexQueriesRace 
The test was introduced in # but was changed during the code review and not reran with the faulty code since then.

Closes #

Signed-off-by: Dimitar Dimitrov <dimitar.dimitrov@grafana.com>
 2023-09-20 20:22:26 +01:00
Bryan Boreham bca9a79bca
Merge pull request #12877 from bboreham/fix-promtool 
promtool: fix compile error from bad merge
 2023-09-20 19:14:30 +00:00
Bryan Boreham 5ecea3c840 promtool: fix compile error from bad merge 
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
 2023-09-20 16:34:20 +00:00
Björn Rabenstein 864da019cd
Merge pull request #12874 from krajorama/outof-order-chunks 
Fix duplicate sample detection at chunk size limit
 2023-09-20 18:01:21 +02:00
Björn Rabenstein 9071913fd9
Merge pull request #12831 from aknuds1/arve/posting-context 
Add context argument to `tsdb.PostingsForMatchers`
 2023-09-20 17:15:15 +02:00
Bryan Boreham 2b77ad4591
Merge pull request #12677 from bboreham/retry-on-500 
remote-write: respect Retry-After header on 5xx errors
 2023-09-20 14:02:34 +00:00
György Krajcsovits 9dbd100a5e Refactor solution to not repeat code 
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
 2023-09-20 15:54:00 +02:00
György Krajcsovits 96d03b6f46 Fix duplicate sample detection at chunks size limit 
Before cutting a new XOR chunk in case the chunk goes over the size
limit, check that the timestamp is in order and not equal or older
than the latest sample in the old chunk.

Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
 2023-09-20 14:49:56 +02:00
György Krajcsovits 56b3a015b6 Add regression test for duplicate detection at chunk size limit 
TestHeadDetectsDuplcateSampleAtSizeLimit tests a regression where a
 duplicate sample,is appended to the head, right when the head chunk is
 at the size limit. The test adds all samples as duplicate, thus
 expecting that the result has exactly half of the samples.

Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
 2023-09-20 14:32:20 +02:00
Ben Ye c78124427e
Support specifying series matchers to analyze tsdb (#12842) 
* support specifying series matchers to analyze tsdb

Signed-off-by: Ben Ye <benye@amazon.com>

* fix cli docs

Signed-off-by: Ben Ye <benye@amazon.com>

---------

Signed-off-by: Ben Ye <benye@amazon.com>
 2023-09-20 11:37:32 +01:00
William Dumont ce6ad15422 remote-write: TestClientRetryAfter status code 500 
and compare the retryAfter values.

Signed-off-by: William Dumont <william.dumont@grafana.com>
 2023-09-20 10:25:43 +00:00
William Dumont febd62a23e remote-write: refactor TestClientRetryAfter 
The new version features a set of test cases that simplify the addition
of new HTTP status codes.

Signed-off-by: William Dumont <william.dumont@grafana.com>
 2023-09-20 10:24:52 +00:00
Bryan Boreham 9b85354acd remote-write: respect Retry-After header on 5xx errors 
If the server sent it to us, we should assume it knows better than we
do and respect it.

Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
 2023-09-20 10:14:38 +00:00
Paschalis Tsilias c173cd57c9
Add a header to count retried remote write requests (#12729) 
Header name is `Retry-Attempt`, only set when >0.

Signed-off-by: Marc Tuduri <marctc@protonmail.com>
Signed-off-by: Paschalis Tsilias <paschalis.tsilias@grafana.com>
 2023-09-20 11:11:03 +01:00
George Krajcsovits 7d713e1fde
Merge pull request #530 from grafana/krajo/fork-sync 
Sync with promtheus/prometheus at 83891135c (2023-09-19)
 2023-09-20 10:09:34 +02:00
György Krajcsovits c9e513f5a3 Merge remote-tracking branch 'upstream/main' into krajo/fork-sync 
# Conflicts:
#	storage/merge_test.go
#	tsdb/chunkenc/histogram_meta.go
 2023-09-19 17:53:25 +02:00
Björn Rabenstein 83891135c6
Merge pull request #12838 from krajorama/fix-disappearing-span-panic 
Fix counterResetInAnyBucket panic
 2023-09-19 17:10:27 +02:00
George Krajcsovits 3512b2d678
storage: make histogram reset handling consistent in chainSampleIterator (#12779) 
storage: make histogram reset handling consistent in chainSampleIterator

---------

Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
 2023-09-19 17:06:46 +02:00
Arve Knudsen b23aa0e34c
Merge pull request #529 from grafana/chore/sync-prometheus 
Sync with latest Prometheus upstream
 2023-09-18 14:53:16 +02:00
Arve Knudsen 3989da3514 Disable GitHub Actions Dependabot updates 
Signed-off-by: Arve Knudsen <arve.knudsen@gmail.com>
 2023-09-18 13:32:59 +02:00
Arve Knudsen e48d4e5835 Merge remote-tracking branch 'prometheus/main' into chore/sync-prometheus 
Signed-off-by: Arve Knudsen <arve.knudsen@gmail.com>
 2023-09-18 09:29:42 +02:00