Hi All,

My server gets stuck and is not able to send mail. It only happens over the weekends and I have to constantly restart (Service Postfix restart) in order for it to start sending again. I've searched forums but have not been able to find any solid answers. I would really appreciate your input please.

Here are some log errors when I run (cat/var/log/maillog | grep warning):

Nov 27 06:24:54 batch-ca4-02 postfix/cleanup[11491]: warning: timeout on cleanup socket while reading input attribute name
Nov 27 06:48:46 batch-ca4-02 postfix/cleanup[11491]: warning: 498D73E3003A: read timeout on cleanup socket

When I run (cat /var/log/maillog | grep error):

Nov 27 04:04:58 batch-ca4-02 postfix/smtpd[4618]: dict_eval: const error
Nov 27 04:04:58 batch-ca4-02 postfix/smtp[4619]: dict_eval: const error
Nov 27 07:11:10 batch-ca4-02 postfix/smtpd[11263]: dict_eval: const error
Nov 27 07:11:10 batch-ca4-02 postfix/smtpd[21990]: dict_eval: const error
Nov 27 08:02:03 batch-ca4-02 postfix/smtpd[26997]: dict_eval: const error
Nov 27 09:02:35 batch-ca4-02 postfix/smtpd[637]: dict_eval: const error
Nov 27 10:02:03 batch-ca4-02 postfix/smtpd[6404]: dict_eval: const error
Nov 27 11:02:26 batch-ca4-02 postfix/smtpd[12311]: dict_eval: const error
Nov 27 12:30:14 batch-ca4-02 postfix/smtpd[20688]: dict_eval: const error
Nov 27 12:30:39 batch-ca4-02 postfix/smtpd[20736]: dict_eval: const error
Nov 27 12:31:38 batch-ca4-02 postfix/smtpd[20834]: dict_eval: const error
Nov 27 13:02:03 batch-ca4-02 postfix/smtpd[24044]: dict_eval: const error

I know that all new mail entering the Postfix queue is written by the cleanup(8) service into the "incoming" queue. When I run (qshape -s incoming) I see that majority of emails are stuck in the incoming queue.

I also read that "If a server is being hammered from multiple directions, consider raising the in_flow_delay to 10 seconds, but only if the incoming queue is growing even while the active queue is not full and the trivial-rewrite service is using a fast transport lookup mechanism." I set the in_flow_delay to 10 seconds and restarted postfix to set it. Now I just have to wait to see if that will solve the issue.

Can anyone that's more experienced with this please tell me what's going on, as this is a little confusing? Thank you for your help and I can provide more info on request.