Re: [exim] Host-based retrying problem

On 2006-03-03 at 10:12 -0800, Christian Gregoire wrote:
> The problem is that when A gets a host error from B for some reason,
> all messages destined to B are queued. And when retry time has come, A
> sends all messages to B at once, whose load average rises, so leading
> Sendmail to refuse connections. So emails get queued on A ....

> When Sendmail runs on A, everything is fine. I guess that's because it
> uses message-based retrying so that messages are re-sent 'on the
> flow'.

> Any idea ?

retry_data_expire option, lower from 7 days to however many seconds.

If the retry data is always going to be useless for your situation (if A
only sends mail onto B) and you have frequent problems with B, then if
you don't want to have Exim doing db updates every minute, a crude but
effective method is to nuke the retry hints file regularly.

crontab for exim runtime user:
* * * * * rm -f /var/spool/exim/db/retry

where the path /var/spool/exim should be replaced by whatever
"exim -bP spool_directory" gives you.

-Phil