On 12-4-2016 12:03, Loic Dachary wrote:
Hi,
IIRC there needs to be a pattern defined for where core are to be found. For instance teuthology/task/internal.py has 'sudo', 'sysctl', '-w', 'kernel.core_pattern={adir}/coredump/%t.%p.core'.format(adir=archive_dir),
I hope that helps
Our local Linux Guru is out, so I can't harras him now, but this gives
me a pointer.
Any objections to adding the code below?
(Even if only ifdefed for FreeBSD)
It keep the directory pollution a bit down.
--WjW
On 12/04/2016 11:38, Willem Jan Withagen wrote:
Hi Loic,
You have written a lot of the unittests, so it seems.
I'm having the issue that under FreeBSD the cores are left behind from
the EXPECT_DAETH tests.
But under CentOS7 no cores are left behind.
And 'ulimit -c unlimited' does not help to get cores
Is there anything particular why this is the case...
Google tests suggest that it is the programs responsability to cleanup
after itself.
And would there be any objections to adding code like:
ret = RUN_ALL_TESTS();
if (ret == 0) {
char core [30];
strncpy(core, argv[0], 19);
core[19] = '\0';
strcat(core, ".core");
::unlink(core);
free(core);
}
return ret;
to the tests
Thanx,
--WjW
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
