exporter/prometheusremotewriter: sort Sample by Timestamp to avoid out of order errors #2941
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Codecov Report
@@ Coverage Diff @@
## main #2941 +/- ##
=======================================
Coverage 91.66% 91.66%
=======================================
Files 312 312
Lines 15312 15317 +5
=======================================
+ Hits 14036 14041 +5
Misses 870 870
Partials 406 406
Continue to review full report at Codecov.
|
bogdandrutu
approved these changes
Apr 15, 2021
odeke-em
force-pushed
the
prometheus-remotewriteexporter-fix-out-of-order-samples
branch
from
4c6e148
to
3f40ffc
Compare
bogdandrutu
reviewed
Apr 15, 2021
…t of order errors Ensures that before a prompb.WriteRequest is created, that the TimeSeries contained have their Sample values sorted chronologically by Prometheus to avoid out of order errors reported by Prometheus barfing. Thanks to @rakyll for a reproducer and for diagnosing the problem, which helped distill the issue from a very complex setup that required super expensive Kubernetes clusters with many replicas, but essentially the problem became more apparently when the number of TimeSeries grew. It is important to note that the presence of such a bug signifies that with a large number of replicas are being scraped from, this stalls scraping and takes a long time which means that targets scraped in a round-robin fashion experience staleness when many. This might be even more reasons for setups to adopt a push model as opposed to scrape endpoints. Fixes open-telemetry#2315 Fixes open-telemetry/wg-prometheus#10
odeke-em
force-pushed
the
prometheus-remotewriteexporter-fix-out-of-order-samples
branch
from
3f40ffc
to
8b97422
Compare
|
Thanks much for this PR! I've been testing the exporter with this change and can't reproduce the bug anymore. I'm using relabel_configs to have pod name in the samples to avoid the duplicates that might be potentially coming from the replicas of my app. |
|
I spoke too soon, filed #2949 for a follow up. |
odeke-em
deleted the
prometheus-remotewriteexporter-fix-out-of-order-samples
branch
hughesjj
pushed a commit
to hughesjj/opentelemetry-collector
that referenced
this pull request
Apr 27, 2023
Bumps [peter-evans/create-pull-request](https://github.com/peter-evans/create-pull-request) from 4 to 5. - [Release notes](https://github.com/peter-evans/create-pull-request/releases) - [Commits](peter-evans/create-pull-request@v4...v5) --- updated-dependencies: - dependency-name: peter-evans/create-pull-request dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <[email protected]> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Ensures that before a prompb.WriteRequest is created, that the
TimeSeries contained have their Sample values sorted chronologically
by Prometheus to avoid out of order errors reported by Prometheus
barfing.
Thanks to @rakyll for a reproducer and for diagnosing the problem, which
helped distill the issue from a very complex setup that required super
expensive Kubernetes clusters with many replicas, but essentially the problem
became more apparently when the number of TimeSeries grew.
It is important to note that the presence of such a bug signifies that
with a large number of replicas are being scraped from, this stalls
scraping and takes a long time which means that targets scraped in a
round-robin fashion experience staleness when many. This might be even
more reasons for setups to adopt a push model as opposed to scrape
endpoints.
Fixes #2315
Fixes open-telemetry/wg-prometheus#10