[
https://issues.apache.org/jira/browse/GEODE-9431?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17384135#comment-17384135
]
ASF GitHub Bot commented on GEODE-9431:
---------------------------------------
pivotal-jbarrett commented on pull request #831:
URL: https://github.com/apache/geode-native/pull/831#issuecomment-882667398
> > This seems like a really big hammer for what strikes me an issue with
maybe a specific RHEL C++ runtime patch or gcc patch. I didn't see any open
tickets with RHEL or GNU regarding these issues in the GEODE ticket. Given that
boost::regex is the basis for std::regex if we are using std::regex incorrectly
resulting in different undefined behavior in RHEL then I would be just as
concerned with our use of boost::regex.
>
> I spent most of last week picking up and discarding various "hammers" as
solutions to this problem - this was the first one that worked. Let me know
what mitigations you think we need for this, I'm fine with most things. If we
need/want a JIRA ticket to revert this back once it's fixed by RedHat or
whomever, that's fine. I would also argue it's not really _that_ big of a
hammer, since the change itself took about an hour to make and verify, and
backing it out is the work of a few minutes. I'll figure out how to file the
bug with RedHat and GNU today, but I don't think waiting for them to provide a
fix is a viable strategy. We need to get our CI pipeline off the floor so we
can proceed with dev work and know we're not further damaging the product on
RHEL-8.
As long as we can get issues opened with RH and and with ourselves to revert
back to `std::regex` once we or RH fixes the underlying issue, I could get on
board with this change.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscr...@geode.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
> CI Failure: SegFault in apache-geode_unittests
> ----------------------------------------------
>
> Key: GEODE-9431
> URL: https://issues.apache.org/jira/browse/GEODE-9431
> Project: Geode
> Issue Type: Bug
> Components: native client
> Reporter: Kamilla Aslami
> Priority: Major
> Labels: pull-request-available
>
> {{Two consequent geode-native-develop/build-rhel-8-release runs failed with
> segmentation fault: }}
> {noformat}
> selected worker: 3eb01fee-0959-4e2d-8c25-44a153962eaa
> Warning: Permanently added '34.67.7.72' (ECDSA) to the list of known hosts.
> Test project /home/build/build/cppcache/test
> Start 1: apache-geode_unittests
> 1/1 Test #1: apache-geode_unittests ...........***Exception: SegFault 0.43
> sec
> Running main() from
> /home/build/build/dependencies/gtest/gtest-extern-prefix/src/gtest-extern/googletest/src/gtest_main.cc
> [==========] Running 374 tests from 42 test suites.
> [----------] Global test environment set-up.
> [----------] 6 tests from durationTest
> [ RUN ] durationTest.ceil
> [ OK ] durationTest.ceil (0 ms)
> [ RUN ] durationTest.toString
> [ OK ] durationTest.toString (0 ms)
> [ RUN ] durationTest.fromString
> [ OK ] durationTest.fromString (0 ms)
> [ RUN ] durationTest.fromStringWithCeil
> [ OK ] durationTest.fromStringWithCeil (0 ms)
> [ RUN ] durationTest.fromStringException
> [ OK ] durationTest.fromStringException (0 ms)
> [ RUN ] durationTest.assertBounds
> [ OK ] durationTest.assertBounds (1 ms)
> [----------] 6 tests from durationTest (1 ms total){noformat}
> ...
> {noformat}
> 0% tests passed, 1 tests failed out of 1
> Total Test time (real) = 0.91 sec
> The following tests FAILED:
> 1 - apache-geode_unittests (SEGFAULT)
> Errors while running CTest{noformat}
>
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
