Repository navigation
poll BigQuery matchers on missing tables - #40559
Conversation
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## master #40559 +/- ##
=============================================
+ Coverage 56.16% 79.90% +23.74%
=============================================
Files 1124 488 -636
Lines 178336 82397 -95939
Branches 1489 0 -1489
=============================================
- Hits 100170 65843 -34327
+ Misses 75645 16554 -59091
+ Partials 2521 0 -2521
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
|
Assigning reviewers: R: @claudevdm for label python. Note: If you would like to opt out of this review, comment Available commands:
The PR bot will only process comments in the main thread (not review comments). |
|
Thanks for root causing it. There is no change on bigquery_matcher.py recently, is it caused by a offending Beam PR? If not, what time this breakage started happen (which then suggest a BigQuery service breaking change) |
No, bigquery_matcher.py did not change recently and the last edit before this was formatting only It also did not start at one specific time and the same test still passes on some runs and fails on others, so it does not look like a BigQuery breaking change |
Fixes: #30760
PostCommit Python Arm streaming tests were failing because the BigQuery table does not exist until the first insert and the matcher already had a timeout, but a 404 stopped the query after a few retries, so that wait never ran
This keeps polling while the table is missing, until the timeout that was already configured. A 404 with no timeout still fails immediately and I checked the matcher unit tests locally