Uh - well THAT is good information. So it is your sense that now they are ready to take on customers with dual-processor Dells without Declude falling asleep?
That means I can finally upgrade (thanks to your persistence). Best Regards Andy Schmidt Phone: +1 201 934-3414 x20 (Business) Fax: +1 201 934-9206 -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Darrell ([EMAIL PROTECTED]) Sent: Tuesday, October 18, 2005 09:49 PM To: [email protected] Subject: Re: [Declude.JunkMail] Wondering about Declude 3.x Matt, I have been running 3.0.5.7 on a single and multi proc machine both of which process over 100K messages a day with one getting close to 150K with no issues. Everything is processing fine. Under 3.0.5.7 I am not seeing the country and mailfrom issue I seen in the 3.0.5.5 version. Also, the sleep issue I seen in earlier versions does not occur in 3.0.5.7. Depending on your volume you will need to tweak the settings for how the threads operate. If you have a bursty flow and a high thread count you can see pretty good prolonged cpu spikes because you will/could have a bunch of external processes launched at the exact same time. Under the old versions <2.x this was not an issue since processes were launched as the volume came in balanced over time - as you know with 3.x it batches processes messages for the most part. With some tweaking you can mitigate it fairly well. Darrell ------------------------------------------- Check out http://www.invariantsystems.com for utilities for Declude And Imail. IMail Queue Monitoring, Declude Overflow Queue Monitoring, SURBL/URI integration, MRTG Integration, and Log Parsers. ----- Original Message ----- From: "Matt" <[EMAIL PROTECTED]> To: <[email protected]> Sent: Tuesday, October 18, 2005 5:44 PM Subject: [Declude.JunkMail] Wondering about Declude 3.x > Since things have been quite for some time, I just wanted to check up > on > what is happening with 3.x. The last that I heard, there were several > people having issues with multi-processor systems. The thread settings > also concern me in the way that they are being implemented. It appears > from reports that these can greatly affect the performance of a system > (and therefore it's stability/ability), and knowing how variable E-mail > can be, I'm not sure that this is something that I would want to have hard > coded on my system. I would hope that there would be another way to go > about this. > > Right now I'm on 2.0.6.16 and have been for some time, but as bugs > arise, > and bug fixes are released, I would like to have the peace of mind to > upgrade to the latest code, but I'm not sure that I have that yet. > > I know that the folks at Declude have been working long and hard at > this > for sometime now, and I don't want to disrespect that hard work, but I > would appreciate an update on where things are, and where they are going > as far as the service issues go. > > Thanks, > > Matt > --- > 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. > --- 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. --- 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.
