Github user dongjoon-hyun commented on the issue:
https://github.com/apache/spark/pull/21093
`BucketedWriteWithoutHiveSupportSuite` is testing for `Seq("parquet",
"json")`, and the testsuite fails after `insertInto` and at `write bucketed
data`. So, the flakiness seems to be irrelevant to this patch. Also, the final
one passed 8 hours ago.
```
[info] BucketedWriteWithoutHiveSupportSuite:
[info] - bucketed by non-existing column (28 milliseconds)
[info] - numBuckets be greater than 0 but less than 100000 (10 milliseconds)
[info] - specify sorting columns without bucketing columns (8 milliseconds)
[info] - sorting by non-orderable column (34 milliseconds)
[info] - write bucketed data using save() (9 milliseconds)
[info] - write bucketed data using insertInto() (9 milliseconds)
... Error starts
[info] - write bucketed data *** FAILED *** (4 seconds, 601 milliseconds)
```
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
