Kevin Reed <listaccount@???> said:
>
> Marilyn Davis said:
>
> > Hmmm. This would be a process that exim is instructed to use in the
> > configure file? There are only dnslookup, dsearch and lsearch going
> > on in that configure file.
> >
> >>
> >> to fix, restart the processes. to fix permanently, review and correct
> >> your broken startup scripts.
> >
> > However, the startup file, /etc/rc.d/rc.local, looks like:
> >
>
> This is a pretty non-standard way to start processeses...
>
>
> > #!/bin/sh
> > #
> > # This script will be executed *after* all the other init scripts.
> > # You can put your own initialization stuff in here if you don't
> > # want to do the full Sys V style init stuff.
> >
> > touch /var/lock/subsys/local
> > hostname kuna
> > /b/home/exim/bin/exim -bd -q15m <<< should be LAST!
> > apachectl start
> > /b/local/mailman/bin/mailmanctl -s start
> > /usr/lib/cour*/libexec/imapd.rc start
> > /usr/bin/spamd -d -u nobody <<< should be before exim start
> > rsync --daemon
> > su clerk -c /usr/bin/eVote &
>
> You are starting Exim before you start spamd. Should be the other way
> around.
>
> Most of these should be in separate startup scripts with error checking to
> make sure they are actually running.
Thank you! Great advice.
If I fix the order in which I start processes and check that they are
running, am I then "standard"? Can you suggest a document that
addresses this?
>
> It is also showing that you apparently don't have shutdown scripts to
> gracefully shutdown processes...
The shutdown script would stop processes in the opposite order? What
else could I do?
Sander said:
> > I don't know what the PANIC error is about. What broke the pipe?
>
> AFAIK a broken pipe occurs if you try to write on a socket that isn't
> connected anymore. So for instance, the initial connection to spamd
> succeeded, but then spamd died / restarted or the connection got
> terminated otherwise, and when you tried to write to the socket it
> failed because there was nothing to write to.
>
> > I don't think the messages that were temporarily rejected reached as
> > far as the sa-exim code. There were no PANIC's.
>
> There *WAS*. sa-exim PANIC'ed, and because of that, sa-exim didn't know
> what to do with the message and told exim to temp-reject it.
There was the one PANIC, just before everything started working. There
were hundreds of temporary rejections before that one panic.
>
> > I don't know why the first one that made it past RCPT hit a PANIC, but
> > maybe it's a clue of some sort.
>
> sa-exim runs after the DATA section has been closed.
Hundreds of these:
2003-11-16 10:32:16 H=(mailserver.midtown.net) [205.162.100.17] F=<zapa@???> temporarily rejected RCPT <marilyn@???>
2003-11-16 10:32:22 H=(gvqfezf.115.33) [200.55.96.237] F=<fawnmascaro@???> temporarily rejected RCPT <office@???>
2003-11-16 10:32:37 H=(mailserver.midtown.net) [205.162.100.17] F=<zapa@???> temporarily rejected RCPT <marilyn@???>
2003-11-16 10:33:00 H=(gvqfezf.115.33) [200.63.161.50] F=<marquittaazim@???> temporarily rejected RCPT <marilyn@???>
2003-11-16 10:33:20 H=(gvqfezf.115.33) [200.63.161.50] F=<marquittaazim@???> temporarily rejected RCPT <news@???>
2003-11-16 10:33:40 H=(gvqfezf.115.33) [200.63.161.50] F=<marquittaazim@???> temporarily rejected RCPT <owner-list@???>
2003-11-16 10:33:56 H=(MAILMAN.cgam.com) [66.107.38.164] F=<Biglooq@???> temporarily rejected RCPT <charter1@???>
2003-11-16 10:34:00 H=(gvqfezf.115.33) [200.63.161.50] F=<marquittaazim@???> temporarily rejected RCPT <nobody@???>
> > >> ret=write(writefd[1],hl->text,strlen(hl->text));
> > So. I still don't know.
>
> I do.
> I'm 99.9% sure it had to do with spamd / spamc not functioning correctly.
All those RCPT temporary rejections were because of spamd/spamc?
Quoting Tim:
> > Not the earlier tempreject errors in his post (and in the subject line),
> > because they're at RCPT time, so can't be anything to do with SA.
>
> Are we entirely certain that the messages were rejected at RCPT time?
>
> It's just a log-line, it could well be that exim (or sa-exim makes
> exim..) log the reject like 'temporarily rejecting RCPT <...>'. Did
> Marylin actually try to deliver a message by hand, with telnet, and see
> when the message was rejected?
My son could see this SMTP session in his log:
Sun 2003-11-16 11:12:51: ----------
Sun 2003-11-16 11:12:41: [-1:1537:3] Parsing Message <C:\MDAEMON\Remoteq\pd50000001790.msg>
Sun 2003-11-16 11:12:41: [-1:1537:3] From: fishgills@???
Sun 2003-11-16 11:12:41: [-1:1537:3] To: marilyn@???
Sun 2003-11-16 11:12:41: [-1:1537:3] Subject: Testing again
Sun 2003-11-16 11:12:41: [-1:1537:3] Message-ID: <005001c3ac74$84a0a450$0400000a@???>
Sun 2003-11-16 11:12:41: [-1:1537:3] MX-record resolution of [deliberate.com] in progress (DNS Server: 10.0.0.2)...
Sun 2003-11-16 11:12:41: [-1:1537:3] P=010 D=deliberate.com TTL=(37) MX=[rosa.deliberate.com] {209.232.123.53}
Sun 2003-11-16 11:12:41: [-1:1537:3] Attempting MX: P=010 D=deliberate.com TTL=(37) MX=[rosa.deliberate.com] {209.232.123.53}
Sun 2003-11-16 11:12:41: [-1:1537:3] Attempting SMTP connection to [209.232.123.53 : 25]
Sun 2003-11-16 11:12:41: [616:1537:3] Waiting for socket connection...
Sun 2003-11-16 11:12:41: [616:1537:3] Socket connection established (10.0.0.2 : 3611 -> 209.232.123.53 : 25)
Sun 2003-11-16 11:12:41: [616:1537:3] Waiting for protocol initiation...
Sun 2003-11-16 11:12:41: [616:1537:3] <-- 220 deliberate.com ESMTP Exim 4.22 Sun, 16 Nov 2003 11:12:41 -0800
Sun 2003-11-16 11:12:41: [616:1537:3] --> EHLO fishgills.net
Sun 2003-11-16 11:12:41: [616:1537:3] <-- 250-deliberate.com Hello fishgills.net [209.232.123.54]
Sun 2003-11-16 11:12:41: [616:1537:3] <-- 250-SIZE 52428800
Sun 2003-11-16 11:12:41: [616:1537:3] <-- 250-PIPELINING
Sun 2003-11-16 11:12:41: [616:1537:3] <-- 250 HELP
Sun 2003-11-16 11:12:41: [616:1537:3] --> MAIL From:<fishgills@???> SIZE=1688
Sun 2003-11-16 11:12:41: [616:1537:3] <-- 250 OK
Sun 2003-11-16 11:12:41: [616:1537:3] --> RCPT To:<marilyn@???>
Sun 2003-11-16 11:13:01: [616:1537:3] <-- 451 Temporary local problem - please try later
Sun 2003-11-16 11:13:01: [616:1537:3] --> QUIT
Sun 2003-11-16 11:13:01: [616:1537:3] This message is 9 minutes old; it has 51 minutes left in this queue
Sun 2003-11-16 11:13:01: [616:1537:3] SMTP session abnormally terminated, 226 bytes transferred.
Sun 2003-11-16 11:13:01: ----------
>
> If you look close at the logs Marylin sent us, you'll see that the
> messageID that was rejected because sa-exim couldn't feed the message to
> spamd (Broken Pipe) was rejected with the same message: 'temporarily
> rejecting RCPT <...>'.
That's true. But doesn't my son's log indicate that the previous
messages were rejected before sa-exim?
Well, anyway, at least you folks gave me something to fix and I can be
satisfied with that if there are no more ideas.
Thank you all very much.
Marilyn
