On Wed, 20 Nov 2002 10:18:47 +0000 (GMT), Philip Hazel wrote:
[...]
>> Anyway, to make a long story short, I thought "smtp_receive_timeout = 1m"
>> might help, but it didn't. :-(
>
>No, that will bite only when the remote host doesn't answer quickly
>enough.
In fact this is *exactly* what I was suspecting -- but obviously (since the
above statement didn't help) the problem is somewhere else.
> It suggests that indeed SpamAssassin is where the delay is.
Thought so. :-(
>> Is there any option (I don't think so) that would allow me to set a maximum
>> lifetime for Exim son processes? And if processes haven't died after that
time
>> Exim would automatically kill them? Do you think that such an option would
be
>> useful in case it's not already there?
>
>If you are calling SA via the local_scan() function, there is a
This is exactly how I'm calling SA.
>local_scan_timeout option, whose default is 5 minutes. However, since
>you are seeing 30m delays, it can't be that (unless you are using a
>version of Exim earlier than 4.03, because that's when
>local_scan_timeout was introduced).
No, I'm using 4.10 under Debian/GNU Linux.
Still any ideas how to debug the problem? Especially since "kill -USR1 <pid>"
isn't very helpful either in these cases:
2002-11-19 13:34:36 29832 handling incoming connection from my.mailhub.de
[aaa.bbb.c.dd]:51389
Thanks,
Ralf
--
L I N U X .~.
The Choice /V\
of a GNU /( )\
Generation ^^-^^
