Re[4]: [Declude.JunkMail] SpamD/SpamC for Declude

2004-01-13 Thread Russ Uhte \(Lists\)
At 05:05 PM 1/12/2004, Sanford Whiteman wrote:
 I guess that was a noble try... but it didn't work.

Well, it probably worked, just not enough. :)
Yeah, I'll buy that! :)


 I'm  going  to try to separate the spamd/spamc processes and see how
 that  goes.
That  will alleviate the utilization issue, for sure. Depending on the
age  of  your  server,  you  should  think  about adding an additional
processor.  I  find that that's one fun part about running mail on old
boxes  with  new disks: as it gives you ability to scale up processing
on  the  cheap  as  needed,  while  still  giving peak performance for
disk-starved  tasks.  A lot of people inadvertently err on the side of
processor power by buying new boxes and ignoring DASD optimization.
Unfortunately, this particular server is out of space for new drives 
internally.  Now realistically, I could rebuild it and do it right, and it 
would probably last a very long time.  When I got the server, it had 4 
drives configured for Raid 5 in a single logical drive with 3 
partitions.  I added two more drives in a mirrored set, and moved the spool 
to this.  That helped drastically.  I may look into external scsi drives...


 I  know  this  server is grossly underpowered for what I'm trying to
 do,  but I inherited it this way, and I don't think I'm gonna get to
 buy a new one here anytime soon. The person before didn't understand
 how to spec out a mailserver.
Gotcha.

One  thing  you should know about that I'm building into SPAMC32 right
now  is  a  SKIPIFWEIGHT  option  that  will return 0 immediately if a
(Declude) weight has already been exceeded, thus saving processing for
way out-of-range spam.
Now that would be awesome.  If there's anything I can do to help, let me 
know.  I don't know a much about VB (I think that's what it's written in?) 
but I'd be willing to help in anyway I can.

Thanks,
Russ
---
[This E-mail scanned for viruses by Declude Virus]
---
[This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)]
---
This E-mail came from the Declude.JunkMail mailing list.  To
unsubscribe, just send an E-mail to [EMAIL PROTECTED], and
type unsubscribe Declude.JunkMail.  The archives can be found
at http://www.mail-archive.com.


Re[4]: [Declude.JunkMail] SpamD/SpamC for Declude

2004-01-12 Thread Sanford Whiteman
 I guess that was a noble try... but it didn't work.

Well, it probably worked, just not enough. :)

 I'm  going  to try to separate the spamd/spamc processes and see how
 that  goes.

That  will alleviate the utilization issue, for sure. Depending on the
age  of  your  server,  you  should  think  about adding an additional
processor.  I  find that that's one fun part about running mail on old
boxes  with  new disks: as it gives you ability to scale up processing
on  the  cheap  as  needed,  while  still  giving peak performance for
disk-starved  tasks.  A lot of people inadvertently err on the side of
processor power by buying new boxes and ignoring DASD optimization.

 I  know  this  server is grossly underpowered for what I'm trying to
 do,  but I inherited it this way, and I don't think I'm gonna get to
 buy a new one here anytime soon. The person before didn't understand
 how to spec out a mailserver.

Gotcha.

One  thing  you should know about that I'm building into SPAMC32 right
now  is  a  SKIPIFWEIGHT  option  that  will return 0 immediately if a
(Declude) weight has already been exceeded, thus saving processing for
way out-of-range spam.

--Sandy



Sanford Whiteman, Chief Technologist
Broadleaf Systems, a division of
Cypress Integrated Systems, Inc.
e-mail: [EMAIL PROTECTED]

SpamAssassin plugs into Declude!
http://www.mailmage.com/download/software/freeutils/SPAMC32/Release/

---
[This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)]

---
This E-mail came from the Declude.JunkMail mailing list.  To
unsubscribe, just send an E-mail to [EMAIL PROTECTED], and
type unsubscribe Declude.JunkMail.  The archives can be found
at http://www.mail-archive.com.