So I am running debian etch and I am to be made aware that my mailserver issues are all stemming from clamd. Basically, the mail is held for a long time before it hits my inbox, I am running virtual domains off postfix with a mysql backend. Every once in awhile mail gets stuck for a long time and clamd spikes the server load to 1.00 , 0.99, 0.91.
Running debian etch.
Solutions?
Debian Etch + Clamd
-
bad_brain
- Site Owner
- Posts: 11636
- Joined: 06 Apr 2005, 16:00
- 19
- Location: In your eye floaters.
- Contact:
well, I have used clamd too at my early server days (on Sarge), but I have removed it after a while, simply because it's useless imo (I mean, if you suspect a user to send infected mails from your server he should be kicked off anyway, and incoming mails can be filtered by RBLs and the other anti-spam features to keep malware out). and clamd is also known to be a resource hog.
first spots to look at are of course the logs (syslog and maillogs), but because you are using postfix (which is a great mail service, beats exim by far) I have another suspect: is maybe greylisting used? check if postgrey is installed.
another reason could be a dead RBL entry, in such a case the server would try to lookup the RBL again and again which causes a delay in mail delivery (check the smtpd_recipient_restrictions block in main.cf for RBL entries)
if in doubt feel free to post the smtpd_recipient_restrictions block, I can supply you with a good config if you want, using postgrey and RBLs without any need for clamav (works nice since months on a system with about 40 virtual hosts and maybe 150 email boxes)....
p.s. check the cronjobs for an eventual system scan by clamscan, would also cause a high load and eventually a delay of clamd activity for mails.
first spots to look at are of course the logs (syslog and maillogs), but because you are using postfix (which is a great mail service, beats exim by far) I have another suspect: is maybe greylisting used? check if postgrey is installed.
another reason could be a dead RBL entry, in such a case the server would try to lookup the RBL again and again which causes a delay in mail delivery (check the smtpd_recipient_restrictions block in main.cf for RBL entries)
if in doubt feel free to post the smtpd_recipient_restrictions block, I can supply you with a good config if you want, using postgrey and RBLs without any need for clamav (works nice since months on a system with about 40 virtual hosts and maybe 150 email boxes)....
p.s. check the cronjobs for an eventual system scan by clamscan, would also cause a high load and eventually a delay of clamd activity for mails.