Tripp wrote: > The -qrlst has been around for quite some time, just not a documented feature. When you perform a > -qr, smtp32 will run a -qrlst unless there are the max number of smtp32 processes already running. > This is for list performance.
Scott wrote: > Yes. In not-so-recent versions (such as the 7.06 I believe you > experimented with), only "-qr" was used. Ipswitch added the "-qrlst", > which I would assume is strictly for performance purposes. Sorry about that Scott. I think the first time I tested it on 7.06, my logging tool was overwriting some events or not recording others. It's now recording a -qr followed by a -qrlst every queue run on 7.13. I haven't tested my new smtp32.exe logger on 7.06, but it probably acts the same way as 7.13. My fault for causing that tidbit of confusion. As far as which is more efficient (2 -qr's (A) or 1 of each (B)), I suppose it largely depends on what you consider to be your typical mail server use. My users don't use the IMail list feature. We have a separate app for that. So, for me, it seems option B a little inefficient. Senders essentially have to wait an extra 30 minutes (up to 30? or at least 30?) before they receive an undeliverable message from my server. That being said, my users aren't complaining about mail performance so it's not an issue for me. Whereas users who take advantage of the lists would only see a difference if there were also a lot of other mail in the spool. (smtp32.exe processes that only work on Q*.LST files) There isn't some undocumented optimization flag somewhere, is there? Where you can choose between optimizing for lists vs. optimizing for non-list users? -NOrm -Norm To Unsubscribe: http://www.ipswitch.com/support/mailing-lists.html List Archive: http://www.mail-archive.com/imail_forum%40list.ipswitch.com/ Knowledge Base/FAQ: http://www.ipswitch.com/support/IMail/
