On Tue, Dec 18, 2012 at 10:10 PM, Felipe Coelho <[email protected]> wrote: > Hi, and sorry for the delay for any answers > > On Dec 11, 2012 3:45 PM, "Mike Naberezny" <[email protected]> wrote: >> >> Can you temporarily run it in the foreground (`supervisord -n`)? If it >> crashes, it will dump a Python traceback before the shell prompt returns. > > > I let supervisord run in the foreground a few times now, but in none of them > there was a traceback. > > 2012/12/11 David Birdsong <[email protected]> >> >> Check dmesg, OOM killer targets supervisord a lot since it forks off many >> child processes. > > > dmesg also shows no messages about it either. > > I'm currently running Supervisor inside Amazon Web Services, and the only > message I got from supervisor was that the program was killed by "User > signal 1". Do you have any idea of how can I get more information on this > particular problem? >
that sounds like a unix log rotate daemon sending supervisord a SIGUSR1 to rotate it's log files. check out any rules under /etc/logrotate.d/* for any rules that might find supevisord logs and check to see if it's sending a signal to processes after moving said processes log files around. > _______________________________________________ > Supervisor-users mailing list > [email protected] > https://lists.agendaless.com/mailman/listinfo/supervisor-users > _______________________________________________ Supervisor-users mailing list [email protected] https://lists.agendaless.com/mailman/listinfo/supervisor-users
