CORE-8686 crash_tracker: remove old crash files #24972
Merged
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
pgellert
requested review from
aanthony-rp and
michael-redpanda
and removed request for
a team
CI test resultstest results on build#61347
test results on build#61447
test results on build#61516
|
src/v/crash_tracker/recorder.cc
Outdated
| continue; | ||
| } | ||
|
|
||
| crash_files.emplace_back(entry.path(), entry.last_write_time()); |
There was a problem hiding this comment.
can we trust the timestamp on the file? Or should we read the file and check its timestamp? I'm wondering if you could just reuse record::get_recorded_crash().
There was a problem hiding this comment.
I have refactored this now to reuse get_recorded_crashes(). To be able to handle malformed crash files, I am using sorting the crash files by recorded_crash::timestamp() which uses the timestamp in the file if the file is valid and falls back to the last_write_time if the file could not be parsed.
| ss::future<> recorder::start() { | ||
| co_await ensure_crashdir_exists(); | ||
| co_await remove_old_crashfiles(); |
There was a problem hiding this comment.
I was going to make a comment that maybe we should clean up on a certain frequency rather than whenever RP reboots... but these files only get created when RP crashes so yeah this is fine :)
pgellert
force-pushed
the
crashlog/remove-old-files
branch
from
4c98bbb to
97aaece
Compare
|
force-push: address code review feedback (reuse record::get_recorded_crash().) |
michael-redpanda
previously approved these changes
Feb 3, 2025
Extract some functions before adding another to `start()`. Pure refactor, no behaviour change.
The broker will retain the 50 most recent crash files and delete all older crashes on startup. This is to ensure that crash report files do not fill up the disk. Note that we remove files before creating them to ensure cleanup succeeds before we create new files. This means that technically we will have up to 51 crash reports on disk (50 kept + 1 created). This is acceptable since 50 is a loose limit.
Implement a simple test to verify that crash reports are getting removed on `start()` when their number exceeds `crash_files_to_keep`.
pgellert
force-pushed
the
crashlog/remove-old-files
branch
from
97aaece to
e7d5780
Compare
|
michael-redpanda
approved these changes
Feb 3, 2025
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.
The broker will retain the 50 most recent crash files and delete all older crashes on startup. This is to ensure that crash report files do not fill up the disk.
Note that we remove files before creating them to ensure cleanup succeeds before we create new files. This means that technically we will have up to 51 crash reports on disk (50 kept + 1 created). This is acceptable since 50 is a loose limit.
A fixture test is added to verify this behaviour.
Fixes https://redpandadata.atlassian.net/browse/CORE-8686
Backports Required
Release Notes