On Sun, 17 Nov 2002, Ralf G. R. Bergs wrote:
> Now and then I see an Exim son process hanging around for more than half an
> hour. If I send a SIGTERM it I will see the following in mainlog:
>
> SMTP connection from <mailhub> closed after SIGTERM
>
> Is there ANY way to identify which message this process was trying to handle and
> why it blocked? Not even an "strace -p <pid>" did the trick, it just sat there
> doing nothing... :-(
Run the "exiwhat" utility (as root). Or, just send the one process a
SIGUSR1 signal and see what it writes to /var/spool/exim/exim-process.info.
If the client is being very slow, it is perfectly possible for an SMTP
connection to last for more than half an hour.
--
Philip Hazel University of Cambridge Computing Service,
ph10@??? Cambridge, England. Phone: +44 1223 334714.
