[milters] Archive

Lists Index Date Thread Search

Article: 1085
From: Taylor, Grant
Date: 2006-08-04 19:01:19 -0400
Subject: Re: Milter-ClamC "450 4.7.1 anti-virus session write

Removal...........: milters-request@milter.info?subject=remove
More information..: http://www.milter.info/#Support

Anthony Howe wrote:
> Have you checked if the clamd process has died? I've seen it die 
> sometimes after a freshclam update.

*nod*  I do not have any thing restarting the clamd daemon.  (I have 
started it by hand for now.)  Messages will pass through milter-clamc just 
fine, and then it will fail with said error and then immediately succeed 
again.  Seeing as how nothing is respawning clamd, I don't think that it 
died on me.

> Anyway, I've recently changed this in milter-clamc for the next release:
>    !    Accept the message in the event of an clamd socket error,
>        instead of temporaily rejecting the mail. Better to let
>        the mail keep flowing. Sometimes clamd disappears.

Will this be an option?  We have as a company chosen that it is better to 
TempFail email messages when we can not virus scan them.

> Possibly. Never thought of that. You try setting
>     clamd-max-size
> about 64KB smaller than clamd's max size and see if milter-clamc is more 
> graceful when it knows to limit message size. Note that because of how 
> the sendmail / milter passes date, clamd-max-size is convert into max. 
> number of 64KB chunks rounded up. So if you specify clamd-max-size=32, 
> it would actual do 64KB.

I'm willing to do such, but I'm not sure if that is a milter setting or a 
clamd setting.  Will you throw me a bone please?  If not, I'll do some more 

Grant. . . .

Lists Index Date Thread Search