Stabilize diskless no-drop replication test#3511
Merged
zuiderkwast merged 7 commits intovalkey-io:unstablefrom Apr 21, 2026
Merged
Stabilize diskless no-drop replication test#3511zuiderkwast merged 7 commits intovalkey-io:unstablefrom
zuiderkwast merged 7 commits intovalkey-io:unstablefrom
Conversation
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## unstable #3511 +/- ##
============================================
- Coverage 76.58% 76.38% -0.20%
============================================
Files 159 159
Lines 80019 80019
============================================
- Hits 61283 61125 -158
- Misses 18736 18894 +158 🚀 New features to boost your workflow:
|
Member
|
Did this test fail recently, I know viktor attempted to fix it in this PR: #3461 Edit: I just saw the comments on the PR, looks like it did fail again recently. |
zuiderkwast
reviewed
Apr 16, 2026
Contributor
zuiderkwast
left a comment
zuiderkwast
left a comment
There was a problem hiding this comment.
Looks pretty good. Only the "no replicas drop" case is covered, so the other cases can still have timing issues, such as "fast" and "slow" cases? I see there are some special cases for the other cases, for example for "timeout" there is pause_process as well. Do you have a full picture?
sarthakaggarwal97
force-pushed
the
daily-repl-rdb-child-timeout-20260414
branch
from
14afd0b to
5a24b23
Compare
Contributor
Author
|
@zuiderkwast 100 runs for all the variants passed: https://github.com/sarthakaggarwal97/valkey/actions/runs/24521848934 |
sarthakaggarwal97
force-pushed
the
daily-repl-rdb-child-timeout-20260414
branch
from
5a24b23 to
17730b5
Compare
Contributor
Author
|
@zuiderkwast I think it still deflakes the tests a lot, but I am afraid out of 500, I still see 1-5 flaky runs. |
sarthakaggarwal97
force-pushed
the
daily-repl-rdb-child-timeout-20260414
branch
from
56ae8dc to
1e467b0
Compare
Contributor
Author
|
This version is quite stable. Not seeing failures anymore. |
madolson
added
the
run-extra-tests
zuiderkwast
approved these changes
Apr 21, 2026
Contributor
zuiderkwast
left a comment
zuiderkwast
left a comment
There was a problem hiding this comment.
Very good! Sorry for the delay.
Signed-off-by: Sarthak Aggarwal <sarthagg@amazon.com>
Signed-off-by: Sarthak Aggarwal <sarthagg@amazon.com>
Signed-off-by: Sarthak Aggarwal <25262500+sarthakaggarwal97@users.noreply.github.com>
Signed-off-by: Sarthak Aggarwal <25262500+sarthakaggarwal97@users.noreply.github.com>
The 'slow' subcase fallback was searching for a log message matching '*Connection with replica client id * lost.*' but the actual server log format is 'Connection with replica <host>:<port> lost.' — there is no 'client id' in the message. Fix the glob to '*Connection with replica * lost.*'. Signed-off-by: Sarthak Aggarwal <sarthagg@amazon.com>
On fast ARM CI runners the RDB transfer completes before both replicas are killed, so the primary logs '2 replicas still up' instead of 'last replica dropped' or '1 replicas still up'. Add a nested catch fallback to accept all three possible outcomes. Signed-off-by: Sarthak Aggarwal <sarthagg@amazon.com>
Signed-off-by: Sarthak Aggarwal <sarthagg@amazon.com>
sarthakaggarwal97
force-pushed
the
daily-repl-rdb-child-timeout-20260414
branch
from
9a86d16 to
fe1d7e6
Compare
Contributor
Author
|
The test failures are unrelated to this change. Related to #2936 |
sarthakaggarwal97
added a commit
to sarthakaggarwal97/valkey
that referenced
this pull request
Apr 23, 2026
This deflakes all variants of `diskless replicas drop during rdb pipe`. The main issue turned out to be that the test was too sensitive to timing and log ordering under TLS, not that the core behavior was wrong. This keeps the same five subcases (no, slow, fast, all, timeout) but makes them much less CI-fragile. CI passes 200 times: https://github.com/sarthakaggarwal97/valkey/actions/runs/24547258515 --------- Signed-off-by: Sarthak Aggarwal <sarthagg@amazon.com> Signed-off-by: Sarthak Aggarwal <25262500+sarthakaggarwal97@users.noreply.github.com> Co-authored-by: Sarthak Aggarwal <25262500+sarthakaggarwal97@users.noreply.github.com>
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 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 deflakes all variants of
diskless replicas drop during rdb pipe.The main issue turned out to be that the test was too sensitive to timing and log ordering under TLS, not that the core behavior was wrong. This keeps the same five subcases (no, slow, fast, all, timeout) but makes them much less CI-fragile.
CI passes 200 times: https://github.com/sarthakaggarwal97/valkey/actions/runs/24547258515