[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: current state of spamass-milter
From: |
Chad Whitten |
Subject: |
Re: current state of spamass-milter |
Date: |
Thu, 30 Sep 2004 14:36:31 -0500 |
User-agent: |
KMail/1.5.4 |
Im handling about 10k messages per day per node that get past the rbl checks
so I guess I would be okay. would you mind sharing the script that checks
for the socket existence?
On Thursday 30 September 2004 14:31, Chris Crowley wrote:
> I remember that problem. The outward appearance would be that sendmail
> would hang when you connected to it. But, inspecting the problem showed
> the milter had failed.
>
> In my experience ( 3 nodes, 75k messages / day on each ) the milter has
> been very stable. I am running the 0.2 branch.
>
> I do have a script which checks every few minutes for the existence of
> the socket to make sure that the milter is actually running. I also
> check sendmail response time using a simple netcat ( nc ) script, and
> take the node out of the cluster and restart everything in case of
> problems.
>
> Chris Crowley
>
> Dan Nelson wrote:
> > In the last episode (Sep 30), Chad Whitten said:
> >>ive got 4 mx servers running mimedefang + spamassassin & sendmail
> >>8.13. i tried out spamass-milter about a year ago but it would
> >>randomly die. Is this fixed now? Ive downloaded via cvs and
> >>compiled and plan to test it out on one of my mx servers but was just
> >>looking for some feedback before giving it a go.
> >
> > I haven't had my milters die on me in months, but they are relatively
> > light-duty (10k messages/day).
> >
> > Whenever you get a crash, report it so I know there's a problem :) The
> > only thing that I know will cause the current CVS code to crash is an
> > ipv6 connection. There might be a simple fix but I haven't looked at
> > it.
--
Chad Whitten
Network Administrator
neXband Communications
address@hidden
601-944-4801 Phone
601-944-4803 Fax