|
|
|
|
||||||
| comp.mail.sendmail Configuring and using the BSD sendmail agent. |
![]() |
|
|
LinkBack | Outils de la discussion |
|
|
#1 |
|
Messages: n/a
Hébergeur: |
Hi there,
I have sendmail 8.13.8 running on linux fedora 6. How can I configure how many processes sendmail is spawning since they seem to take up all the memory? Also how can I deal with a 20,000 messages queue since messages do not seem to be delivered, just get stuck in the queue with message from maillog: grew worklist for /var/spool/mqueue to 2000...12000. How can I either break up this spool directory into 2 of 10,000 messages each or increase the sendmail settings to deal with 20,000 messages? Thanks a lot |
|
|
|
#2 |
|
Messages: n/a
Hébergeur: |
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1 On Thu, 25 Apr 2007, Tester wrote: > Hi there, > I have sendmail 8.13.8 running on linux fedora 6. > How can I configure how many processes sendmail is spawning since they > seem to take up all the memory? Also how can I deal with a 20,000 > messages queue since messages do not seem to be delivered, just get > stuck in the queue with message from maillog: > grew worklist for /var/spool/mqueue to 2000...12000. How can I either > break up this spool directory into 2 of 10,000 messages each or > increase the sendmail settings to deal with 20,000 messages? > Thanks a lot in your mc file try adding these and then rehashing... define(`confMAX_QUEUE_CHILDREN',`2000')dnl define(`confQUEUE_SORT_ORDER', `modification')dnl ** on dual xeon 3G's 2G memory we use kiddies of 2000 and sendmail sits there still saying ' bring it on ' you could up it more, but if its lesser spec machine, like only a single cpu P4, with at least 1G memory, set it to 1000, of course also factor in your available bandwith, if you only have this on isdn or dsl, drop it to 200 or 300. The queue sort order, well I've tried 'time' and 'modification', mod.. seems to be the best, but YMMV. Sendmail uses 'host' as the order which is not good with large queues. However the bigger question, is how did your queue get to be THAT large in the first place. 20K messages in a queue, I've seen that occasionally on secondary MX if we drop the front end MTA's for maint or whatever, that takes less than 10 mins to clear out. HOTD: Dont just apply the workaround, find and fix the root cause first. - -- Cheers Res Vote for your favourite MTA at http://polls.ausics.net/v3.php -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFGL+NlsWhAmSIQh7MRAuC4AJ9zcbtaHGXtwVEvJLn0AV PZpM0TZgCeIeZ+ lLYlYWxPRSBpLwZ5gY6wDkw= =psEI -----END PGP SIGNATURE----- |
|
![]() |
| Outils de la discussion | |
|
|