Sometimes the cmd_db_exec_test:format_cell__null testcase fails with SIGSEGV (I've seen this more than once on recent copies of CURRENT): From https://jenkins.freebsd.org/job/FreeBSD_HEAD/101/console : [192.168.10.2] out: local/kyua/cli/cmd_db_exec_test:format_cell__blob -> passed [0.012s] [192.168.10.2] out: local/kyua/cli/cmd_db_exec_test:format_cell__float -> passed [0.012s] [192.168.10.2] out: local/kyua/cli/cmd_db_exec_test:format_cell__integer -> passed [0.011s] [192.168.10.2] out: local/kyua/cli/cmd_db_exec_test:format_cell__null -> *** Fatal signal 11 received [192.168.10.2] out: *** Log file is /root/.kyua/logs/kyua.20160201-233110.log [192.168.10.2] out: *** Please report this problem to kyua-discuss@googlegroups.com detailing what you were doing before the crash happened; if possible, include the log file mentioned above [192.168.10.2] out: Segmentation fault (core dumped) [192.168.10.2] out: Warning: run() received nonzero return code 139 while executing 'kyua test'!
Is it possible to get the log file referenced in the crash or, better, a core dump?
lwhsu/rodrigc?
Something else seems to be off on CURRENT. I'm seeing on and off failures with Jenkins: - https://jenkins.freebsd.org/job/FreeBSD_HEAD/101/ - https://jenkins.freebsd.org/job/FreeBSD_HEAD/114/ - https://jenkins.freebsd.org/job/FreeBSD_HEAD/118/ - https://jenkins.freebsd.org/job/FreeBSD_HEAD/126/ 101 seems to be the first failing instance. I need to look around that revision for potential issues that might be causing this failure... As part of this, I need to upgrade my VM and rerun the tests. Craig/Li-Wen: could you please gather the results of the next few Jenkins run(s) so we can diagnose the issue further?
I can reproduce this bug readily; it has nothing to do with FreeBSD's Jenkins system. It also doesn't seem to be related to CURRENT. I'm seeing it on a custom version of FreeBSD that was forked from stable/10 at revision 277174. I've opened an upstream issue here, with more details of the crash https://github.com/jmmv/kyua/issues/156
This was caused by the upgrade to Kyua 0.12 and the port was reverted to 0.11 to mitigate the Jenkins crashes. The underlying issue was reported and diagnosed in https://github.com/jmmv/kyua/issues/156 . The fix will be in 0.13, which I'll try to release soon. This bug can be considered fixed so closing. If we experience problems after the 0.13 upgrade, they'll need a separate diagnostic.