http://issues.apache.org/SpamAssassin/show_bug.cgi?id=5367





------- Additional Comments From [EMAIL PROTECTED]  2007-03-14 08:40 -------
some more data points:

- it's not just spamd_hup.t, it's also spamd_kill_restart.t and
spamd_kill_restart_rr.t.  all 3 tests share a lot of logic and code.  I suspect
race conditions in test code.

- it's not very frequent -- there are about 5-10 spamd_hup.t failures in 640
buildbot test suite runs.  so quite rare.

I think it's a race condition, where the test script spots a new pid-file and
goes to read it before the spamd has _written_ any spamd PID into it.  I think I
may have fixed it in trunk now -- let's see how it goes.



------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

Reply via email to