ARTEMIS-4197 mitigate races w/various config changes #6080
Merged
Conversation
This file contains hidden or 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
Contributor
Author
|
The full test-suite is green on this. |
brusdev
reviewed
Dec 10, 2025
Member
brusdev
left a comment
brusdev
left a comment
There was a problem hiding this comment.
This solution is very robust, the only drawback I can find is the possibility of multiple records for the same item when the deletion fails. Those leacked records could be detected/removed when the journal is loaded to mitigate this issue.
jbertram
force-pushed
the
ARTEMIS-4197
branch
2 times, most recently
from
4c0bfb6 to
5ba24a1
Compare
Contributor
Author
|
@brusdev, I refactored this a bit to avoid orphaned records. |
Contributor
Author
|
Full test-suite looks good on this. |
Member
|
@jbertram can you rebase your branch to resolve the conflicts? |
jbertram
force-pushed
the
ARTEMIS-4197
branch
from
5ba24a1 to
74a4dc4
Compare
This commit mitigates a number of different potential race conditions involving configuration changes. Consider the storeAddressSetting method in o.a.a.a.c.p.i.j.AbstractJournalStorageManager. It mutates mapPersistedAddressSettings twice - once to delete an entry and once to put an entry. If another thread executes the recoverAddressSettings method in between these two operations it will not contain the record that is being stored. This same race condition exists for a number of different maps. This commit mitigates the race condition by ensuring that the relevant map is mutated only once (i.e. via put). This commit also refactors a number of methods to avoid duplicated code. There are no tests associated with this change since there's no deterministic way to induce the race condition. It relies on static analysis and existing tests to validate the fix and detect regressions.
jbertram
force-pushed
the
ARTEMIS-4197
branch
from
74a4dc4 to
015f5b8
Compare
Contributor
Author
|
@brusdev, done! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
This commit mitigates a number of different potential race conditions involving configuration changes. Consider the storeAddressSetting method in o.a.a.a.c.p.i.j.AbstractJournalStorageManager. It mutates mapPersistedAddressSettings twice - once to delete an entry and once to put an entry. If another thread executes the recoverAddressSettings method in between these two operations it will not contain the record that is being stored. This same race condition exists for a number of different maps.
This commit mitigates the race condition by ensuring that the relevant map is mutated only once (i.e. via put).
This commit also refactors a number of methods to avoid duplicated code.
There are no tests associated with this change since there's no deterministic way to induce the race condition. It relies on static analysis and existing tests to validate the fix and detect regressions.