We initially had users procmailrc running spamassassin (v2.54) then
"upgraded" to having them use spamc to talk to spamd. While the
performance was impressive, the machine would frequently hang. The
machine is clearly large enough if it could support them running
spamassassin but we backed off the change.
One of the unique things about our environment is we're running DCE/DFS and
using that to kerb-authenticate imap/ssl ea connections and DFS to push out
/var/spool/mail to perhaps 100 WS st it looks local to them. DCE/DFS is
heavily reliant on RPC calls and this machine had probs with it long ago
until we added a second processor. (For those not familiar with DCE/DFS,
it's very similar to AFS and the moral/functional equivalent of "NIS+/NFS
on steroids")
My theory is that the spamd/spamc interaction involves network connections
that overwhelmed the machine ie doing some tcp/ip tuning might have fixed
this. Any other ideas?
I raise this because some day, we'll likely turn spamd/spamc back on but with
spamd on a remote machine and more specifically, on >1 machine st there's
failover st mail doesn't block. Has anyone done failover spamd on remote
machines? what's involved?
Thanks
=-=-=-=-=-=-=-=-=-=-=-=- generated by /dev/dave -=-=-=-=-=-=-=-=-=-=-=-=-=-=
David Stern University of Maryland
Institute for Advanced Computer Studies
-------------------------------------------------------
This SF.Net email is sponsored by: INetU
Attention Web Developers & Consultants: Become An INetU Hosting Partner.
Refer Dedicated Servers. We Manage Them. You Get 10% Monthly Commission!
INetU Dedicated Managed Hosting http://www.inetu.net/partner/index.php
_______________________________________________
Spamassassin-talk mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
