Re: [exim] av_scanner, ClamAV, defer.

Page principale
Supprimer ce message
Répondre à ce message
Auteur: James Gibbard
Date:  
À: exim-users
Sujet: Re: [exim] av_scanner, ClamAV, defer.
I confess to not know 100% if this is possible with native config options,
indeed the link you seems to suggest no.

If you have some way though, of telling programmatically that clam is
unresponsive then of course there's a way to skip clam checks and therefore
not defer.

I did this for a similar case, where I was using redis to store IP
reputation stuff and I didn't want to defer if network partition or
terrorism or something. I just called an external python script that
checked whether redis was 'accessible', and if not set an ACL var that I
then later used to conditionally skip all checks that relied on redis.

Whether this is the right way to do things depends on your tolerance for
external scripts, resource capacity and how you yourself like to work. In
my situation the cost of such config was justified as it saved me a few
times from deferring mails for over 500k mailboxes. Such a thing should
also be tested extensively of course.




On Friday, January 26, 2018, Marco Gaiarin <gaio@???> wrote:

>
> Today i've hit:
>
>         https://bugzilla.clamav.net/show_bug.cgi?id=11549

>
> Practically, for a broken signature, ClamAV cosumes all fd, after then stop
> responding (but does not die), and so exim start to defer email.
>
> Clearly i'm using one clamav server, on local, eg:
>
>         av_scanner = clamd:/var/run/clamav/clamd.ctl

>
>
> There's some way to tell exim not to defer email? I've read:
>
>         https://www.exim.org/exim-html-current/doc/html/spec_
> html/ch-content_scanning_at_acl_time.html

>
> but seems 'no' to me...
>
>
> Thanks.
>
> --
>   Stiamo arrivando a Beziers. Attento alle curve.
>                         (Claudio, in auto con Igor diretto in Spagna)

>
>
>
> --
> ## List details at https://lists.exim.org/mailman/listinfo/exim-users
> ## Exim details at http://www.exim.org/
> ## Please use the Wiki with this list - http://wiki.exim.org/
>