timeouts and sigpipe

[Gerry Doris]

On Sat, 17 Jan 2004, Chris Yuzik wrote:

> Hi everyone.
>
> Last night my system let some spam through, again with spamassassin
> timeouts. I've checked through the maillog and see a few things that
> seem a bit unusual.
>
> Firstly, I see a SIGPIPE message in the log quite regularly. Not sure
> what has caused this to start.
> Jan 17 02:47:41 ns1 MailScanner[6138]: New Batch: Found 4 messages waiting
> Jan 17 02:47:41 ns1 MailScanner[6138]: SIGPIPE received - trying new log
> socket
> Jan 17 02:47:41 ns1 MailScanner[6138]: New Batch: Found 4 messages waiting
> Jan 17 02:47:41 ns1 MailScanner[6138]: New Batch: Scanning 1 messages,
> 3305 bytes
>
> I also see that MailScanner had some trouble with ORDB last night:
> Jan 17 02:48:14 ns1 MailScanner[30540]: RBL Check ORDB-RBL timed out and
> was killed, consecutive failure 1 of 7
>
> And, of course, spamassassin was having trouble also:
> Jan 17 02:49:16 ns1 MailScanner[30540]: SpamAssassin timed out and was
> killed, consecutive failure 1 of 20
>
> Any help or suggestions would be greatly appreciated.
>
> Thanks,
> Chris
>

I believe the SIGPIPE error was removed several releases back.  What level
of MailScanner are you running?

--
Gerry

"The lyfe so short, the craft so long to learne"  Chaucer