Andy Fiddaman wrote:
On Thu, 8 Mar 2007, Didi Rieder wrote:
; Lucky you, maybe to low message volume....
We have a several Solaris 10 servers running 0.90.1, each processing over
250K messages a day and have seen absolutely no problems. Experimental
code isn't enabled. I wonder what the difference is with your setup?
In my Solaris environment I found that scripted updates would cause
clamd cpu usage to rail. That does not happen when cvd updates occur.
The cpu level goes back down after the changes are read into clamd, but
that doesn't help single cpu systems that are busy with other things.
Disabling scripted updates also removed a number of freshclam errors
that would occur throughout the day when the server failed to deliver
the incremental update. That appears to need more work.
dp
_______________________________________________
Help us build a comprehensive ClamAV guide: visit http://wiki.clamav.net
http://lurker.clamav.net/list/clamav-users.html
