-
Notifications
You must be signed in to change notification settings - Fork 44
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Test sharding - take 2 #2832
Test sharding - take 2 #2832
Conversation
35b8d6f
to
dfeb591
Compare
While github.com/pulumi/sharder should be optimal, github.com/pulumi/shard should approach optimality as the number of tests grows. I want to see the degradation in test performance we see when using a simpler solution.
dfeb591
to
3b8c80d
Compare
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## master #2832 +/- ##
==========================================
- Coverage 68.69% 67.48% -1.21%
==========================================
Files 325 325
Lines 41621 41621
==========================================
- Hits 28593 28090 -503
- Misses 11422 11953 +531
+ Partials 1606 1578 -28 ☔ View full report in Codecov by Sentry. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM, thanks for checking this again. I'd wrongly assumed the results would be similar to before the recent refactoring of tests.
Simpler is definately better
This test flakes in the new sharded CI. It wasn't run before (`x/` was ignored in testing), so I'm not worrying about it. It only tests old and unmaintained code at this point.
0fc8d8a
to
bfec06a
Compare
This function has caused repeated test failures with concurrent map mutations, the most recent of which is https://github.com/pulumi/pulumi-terraform-bridge/actions/runs/12772071950/job/35600764628?pr=2832. This commit converts the function to return a copy instead of mutating.
bfec06a
to
a913512
Compare
Follow-up work on #2810.
This PR aimed to validate the more complex test optimization strategy of #2810 (pre-calculated perfect allocation; implemented by github.com/pulumi/sharder) by demonstrating the effects of a simpler strategy (random allocation; implemented by github.com/pulumi/shard).
Initial testing showed that both strategies were bounded by the 15 minute job
Test and Lint / test (1.22.x, windows-latest, DEFAULT, 0)
, which is not effected by sharding.Since both approaches appear identical, we will merge the simpler strategy.