Skip to content

Commit d15d71d

Browse files
JNSimbaclaude
andauthored
[fix](test) Fix flaky assertion in test_streaming_insert_job_fetch_meta_error (apache#62499)
### What problem does this PR solve? PR apache#62023 refactored the filesystem SPI, which changed the error propagation path in `S3SourceOffsetProvider.fetchRemoteMeta()`. Previously (apache#61284), the debug point created a `GlobListResult` with an error status, and the subsequent status check threw an exception with message `"Failed to list S3 files: ..."`. After the refactoring in apache#62023, the debug point was changed to directly throw an `IOException`, which is caught by `StreamingInsertJob.fetchMeta()` and wrapped as `"Failed to fetch meta, ..."`. However, the regression test assertion was not updated accordingly, causing the test to fail intermittently. Co-authored-by: Claude Opus 4.6 (1M context) <[email protected]>
1 parent 4b49f85 commit d15d71d

1 file changed

Lines changed: 1 addition & 1 deletion

File tree

‎regression-test/suites/job_p0/streaming_job/test_streaming_insert_job_fetch_meta_error.groovy‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -80,7 +80,7 @@ suite("test_streaming_insert_job_fetch_meta_error", "nonConcurrent") {
8080

8181
def jobStatus = sql """select Status, ErrorMsg from jobs("type"="insert") where Name='${jobName}'"""
8282
assert jobStatus.get(0).get(0) == "PAUSED"
83-
assert jobStatus.get(0).get(1).contains("Failed to list S3 files")
83+
assert jobStatus.get(0).get(1).contains("simulated S3 auth error")
8484

8585
sql """
8686
DROP JOB IF EXISTS where jobname = '${jobName}'

0 commit comments

Comments
 (0)