[
https://issues.apache.org/jira/browse/IMPALA-5525?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Pranay Singh resolved IMPALA-5525.
----------------------------------
Resolution: Resolved
The issue reported in the JIRA is fixed by the below commit
-------------------------------------------------------------------------------
test_scanners_fuzz.py currently tests compressed parquet but
does not test uncompressed parquet. This fix adds a new test
case for uncompressed parquet.
Testing
-------
Ran the query_test/test_scanners_fuzz.py in a loop (5 times)
and there was no impalad crash seen.
Change-Id: I760de7203a51cf82b16016fa8043cadc7c8325bc
Reviewed-on: http://gerrit.cloudera.org:8080/8056
Reviewed-by: Tim Armstrong <[email protected]>
Tested-by: Impala Public Jenkins
> Extend TestScannersFuzzing to test uncompressed parquet
> -------------------------------------------------------
>
> Key: IMPALA-5525
> URL: https://issues.apache.org/jira/browse/IMPALA-5525
> Project: IMPALA
> Issue Type: Improvement
> Components: Infrastructure
> Reporter: Tim Armstrong
> Assignee: Pranay Singh
> Priority: Minor
> Labels: ramp-up
>
> TestScannersFuzzing currently tests compressed parquet but does not test
> uncompressed parquet. Uncompressed parquet would help with test coverage
> because there's more potential to corrupt the actual data in interesting
> ways, not just the headers.
> To do this we'd probably need to set COMPRESSION_CODEC=none, do a create
> table as select to write out some parquet data, then do the fuzz testing on
> that.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
