> > 1. Is there any pattern to this occuring? Time that qmail is up, number of
> > mails delivered? Time the OS has been up?
>
>No pattern
Let me suggest that their might be one though. So it's worth looking for and
is probably the way in which you will discover your problem.
> 2. Does it change if you alter concurrency settings?
>I never changed concurrency settings.
>Is it necessary ?
I was more asking what happens if you *do* alter them. But I suspect that
looking for some sort of pattern is your best first approach.
i start it from a root shell, here goes what i do to start it:
># export PATH=/var/qmail/bin:$PATH <enter>
># qmail-start ./.Mailbox sh -c 'accustamp | cyclog /var/log/qmail' &
Enough has been said on this already. Please remove all modifications to
your system before bringing it in for a warranty claim.
> 7. Can you show the list the log entries which preceed the "lost spawn"
> > message? Any correlation there? Eg, is it always the start of a remote
> > delivery that causes the lost spawn? Have you hit some common concurrency
> > number?
And on the other 5 or 6 occassions per day, is there a pattern to these log
messages that you've tried to discover?
Regards.
