|
Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com |
Re: /path/to/smtpd : bad command startup -- throttling
From: David A. Gershman (gershman
dagertech.net)
Date: Thu Dec 09 2004 - 14:35:51 CST
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
> > > Nov 29 01:05:40 {hostname} postfix/smtpd[20750]: fatal: watchdog
timeout
> >
> > It's taking too much time to pass the mail on from smtpd to
> > the cleanup server.
> >
> > Do you have lots of header_checks or body_checks patterns?
No. Any checks like that are done on a per-user basis via procmail.
But I'm wondering if there was a system slow-down causing the time-out.
I'll need to look into that.
> Or a network read was blocked, despite the OS (Solaris?) promising
> available data. My money is on falsely readable sockets. It would
> be helpful if the OP reported the Postfix release and OS name/version:
>
> postconf mail_version
> uname -a
Sorry Viktor, that was partially provided on original email. I just
updated to v2.1.5. The error was from 2.1.4. As for the OS, Linux
Kernel 2.4.27, but just updated.
I'm going to check for signs of a system slowdown else I suppose I'll
just wait and see if it happens again. Thanks.
BTW, mind letting me know what 'OP' stands for?
David
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]