The lock date is set into the future on purpose, so that eventually stale locks can be automatically reclaimed.

Some process died holding a lock. Remove it and drive on if you're sure that process is dead.

I am having a very odd situation with lock files. My entire mailman system suddenly stopped working a few days ago, and I have tracked it down to the lockfiles. In the logs, nothing is able to get a lock. What I have noticed is the timestamp on the lock file is exactly 10 hours in the future from the current time, when it tried to get the lock. This is not a problem with cron or the OS time setting, as I setup another job in cron exactly like qrunner but through bourne shell, and it reports time properly, so something from python on down is causing it to do this oddly. Does anybody have any idea what this could be? Where I can start digging? I am running mailman 2.0.6 on Solaris 8, x86.

-Brandon Gillespie

(BTW, please CC me on any response as I am not on the mailing list)

---

Mailman-Developers mailing list Mailman-Developers@python.org http://mail.python.org/mailman/listinfo/mailman-developers