On Thu, 28 Aug 2003, D.Monroe wrote:

> Mark Simon Powell wrote:
>
> > Hi,
> >   I managed to fill my disk quite easily with a test zip:
> >
> > $ mkfile -n 10g fileofdeath
> > $ zip -9 zipofdeeath fileofdeath
>
> try /var/qmail/control/databytes ?

  42.zip is 42K in size. It will consume vast amounts of cpu time,
bringing your relay to it's knees unless you apply some sort of cpu
resource limit. However, then it is simply soft failed and will keep
retrying and retrying constantly hogging cpu every time qmail scanner
processes it.
  With regards to the zipofdeath; Unfortunately our
/var/qmail/control/databytes is set to over 10MB.  Users want this, but
this makes us susceptible to this. Anyway, why should I have to set the
databytes limit abitrarily low just to avoid this? Surely the scanner
should be able to cope with this?
  Currently any kind of resource limiting causes unzip/uvscan to fail and
qmail scanner to cause the file to be held at the remote end and then
retried again.
  Cheers.

-- 
Mark Powell - UNIX System Administrator - The University of Salford
Information Services Division, Clifford Whitworth Building,
Salford University, Manchester, M5 4WT, UK.
Tel: +44 161 295 5936  Fax: +44 161 295 5888  www.pgp.com for PGP key


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
Qmail-scanner-general mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/qmail-scanner-general

Reply via email to