[Raw Msg Headers][Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: help... ZMailer 2.99.54's scheduler is still dying
we have 2 different instalations using .54 on Linux
and .53pre1 on Solaris 2.6 and the scheduler start
dying more than twice a day in the last weeks with messages
in the log similar to those described in this thread.
There is any patch available for this problem?
anything we can do?
Now we are running a cron job wich cheks every 5 minutes,
and reloads scheduler if necesary, but this is not and ideal
And christmas and new year mail load is killing us ;-).
Feliz Año Nuevo,
C novim godom..
(and i don't know finish, so Happy New Year)
> -----Original Message-----
> From: Matti Aarnio [mailto:firstname.lastname@example.org]
> Sent: Wednesday, November 22, 2000 9:50 PM
> To: Eugene Crosser
> Cc: email@example.com
> Subject: Re: help... ZMailer 2.99.54's scheduler is still dying
> On Wed, Nov 22, 2000 at 10:25:12PM +0300, Eugene Crosser wrote:
> > > Hmm.. The "Return-Receipt-To:" handling is entirely within
> > > the local-delivery -- mailbox, and furthermore,
> it requires
> > > that you run the mailbox program with -S option on.
> > > Do you ?
> > Yes I do. At least now I understand why *you* do not see this ;-)
> > Well, even if there is an error in mailbox or in my local changes
> > to that (very unlikely), scheduler shall be more tolerant to such
> > errors...
> > I once sent you a queue file that causes scheduler's
> attempt to unlink
> > directory, do I need to find another one and send it to you again?
> > Just tell me.
> It was most educative to step around the code to see where the
> data corruption happened.
> Driving factor was actually having a recipient which causes error
> condition, but has NOTIFY=NEVER DSN parameter set, AND I think
> the incoming message had to be an error message...
> (Propably could happen with multirecipient message as well, if all
> diagnostics were suppressed.)
> > Eugene
> /Matti Aarnio <firstname.lastname@example.org>