lst_hoe01kwsoft.de
Date: Thu Mar 03 2005 - 02:56:43 CST

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Zitat von Michael Nguyen <michaelntwentyten.org>:

>
> Andreas,
>
> Thanks for writing back. First of all, I haven't made the time to follow
> Wietse's debugging suggestions, but I didn't see an error all day yesterday.
> I am running Postfix 2.1.5 with no TLS. The server that is seeing the
> problem is a test server that has only two users on it. As a result, it
> often goes a full half hour without seeing an email.
>
> When this error happens, typically nothing is happening. It's just running
> quietly, throws the error, and continues to run quietly. No emails have
> been lost as far as I can tell. No connections were made around the time
> when the error is thrown.

Grep for the PID of the exiting process to see what is happening before the
watchdog timeout. For further debbuging follow the instructions in the
DEBUG-Readme.

> I'm stumped. ;-) What I want to know is....is this dangerous? Is it
> related to the idle-ness of the server? The two other Postfix servers that
> we have in production get consistent traffic and they never throw these
> errors.

It is a safety net from postfix. If the master process is not able to get a
answer from the smtpd for a long time it kills it, because it is suspected that
the smtpd process is dead.
Not dangerous but worth to identify if it is happening more than once...

Regards

Andreas

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]