[ https://issues.apache.org/jira/browse/IMPALA-7595?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16627345#comment-16627345 ]
Csaba Ringhofer commented on IMPALA-7595: ----------------------------------------- Sorry for not addressing this sooner, I did not realize that this leads to frequent test failures. I have played a bit with Parquet files that contained similar corrupt time values. The good news: - I couldn't crash a debug Impala (3.x without "IMPALA-7521) even with extreme values The bad news: - these corrupt values are written verbatim to new Parquet files (I tried with create table as select), so if there is such a corrupt value, it can spread to new data - if read with Hive, some functions (e.g. from_utc_timestamp) throw an exception, and the whole query is aborted even if there were valid values > Check failed: IsValidTime(time_) at timestamp-value.h:322 > ---------------------------------------------------------- > > Key: IMPALA-7595 > URL: https://issues.apache.org/jira/browse/IMPALA-7595 > Project: IMPALA > Issue Type: Bug > Components: Backend > Affects Versions: Impala 3.1.0 > Reporter: Tim Armstrong > Assignee: Csaba Ringhofer > Priority: Blocker > Labels: broken-build, crash > > See https://jenkins.impala.io/job/ubuntu-16.04-from-scratch/3197/. hash is > 23c7d7e57b7868eedbf5a9a4bc4aafd6066a04fb > Some of the fuzz tests stand out amongst the tests that were running at the > same time as the crash, particularly: > 19:12:17 [gw4] PASSED > query_test/test_scanners_fuzz.py::TestScannersFuzzing::test_fuzz_alltypes[exec_option: > {'debug_action': '-1:OPEN:SET_DENY_RESERVATION_PROBABILITY@1.0', > 'abort_on_error': False, 'mem_limit': '512m', 'num_nodes': 0} | table_format: > parquet/none] -- This message was sent by Atlassian JIRA (v7.6.3#76005) --------------------------------------------------------------------- To unsubscribe, e-mail: issues-all-unsubscr...@impala.apache.org For additional commands, e-mail: issues-all-h...@impala.apache.org