Hi

I am using Centos 5.2 32bit Sendmail/Fetchmail (sendmail-8.13.8-2.el5 and fetchmail-6.3.6-1.el5) for our local maildelivery from a common account.I can able to send mails through this server.But while receiving all mails are queued in /var/spool/mqueue directory.

Even if i flush these mails by executing "/usr/sbin/sendmail -q -v ", shows messages are locked. It was working 3 years without any problem, suddenly this problem arises.Another interesting point is if i restart the machine mails are flushed out and users got their mails.

Any tips will be more helpful ?

Regards
Gopal

tail -f sendmail logs whilst flushing.
send output here

Hi

Thanks for the reply,when i executing "/usr/sbin/sendmail -v -q " at the
console as root i got the message as,

Running /var/spool/mqueue/nARCtwIW005034 (sequence 1 of 7)
nARCtwIW005034: locked

Running /var/spool/mqueue/nARCjfPZ004201 (sequence 2 of 7)
nARCjfPZ004201: locked

Running /var/spool/mqueue/nARChReD004019 (sequence 3 of 7)
nARChReD004019: locked

Running /var/spool/mqueue/nARCe1Rq003775 (sequence 4 of 7)
nARCe1Rq003775: locked

Running /var/spool/mqueue/nARD2ar3005629 (sequence 5 of 7)
nARD2ar3005629: locked

Running /var/spool/mqueue/nARCqV8b004702 (sequence 6 of 7)
nARCqV8b004702: locked

Running /var/spool/mqueue/nARCkkgX004268 (sequence 7 of 7)
nARCkkgX004268: locked


and i am not seeing any logs related to this in /var/log/maillog.

Please let me know further.

this error normally means the message is already in the process of being delivered and so the queue process cannot attach to any of these...


please tail -f /var/log/maillog at the same time whilst running the mailq or sendmail -q commands

Hi

Thanks for the info. Upon further deep checking, for each entry in the /var/log/mqueue, while doing flush ( sendmail -q -v ), i got an entry as "timeout waiting for input from local during Draining Input" in the /var/log/maillog.

Hope this help further.

Note : I dont have any spam filter like spamassasin in my server.

well i told you why you get the locked message, now it is up to you to find out why these messages are taking so long to deliver. start by checking the content of the messages..etc

Hi

Thanks for the info. Let me explain my scenario. We are getting support emails from various companies, and it will be fetched to a local user mailbox. In my aliases file i configured such that all the support mails will be copied to some other users also and these mails are not flushed and available in the /var/spool/mqueue.

I setup OTRS locally, and this fetch support mails periodically from the local user mailbox. After googling i find out something like this,

If e-mail is delivered to a program which generates too much output, then sendmail may issue an error:

timeout waiting for input from local during Draining Input

I have sent a mail to OTRS mailing list.Meanwhile is there anything to be
monitored or noticed in the sendmail server.

Cheers
Gopal

ah...if only you gave this info earlier
anyway, simple answer is, i know nothing about OTRS

but i do know that if it gives you more trouble, try using RT.

Hi

After drilling down all the logs, and i found the culprit.Also when i stop the dovecot service,the mails getting flushed and if i start it, then mails are accumulated in the queue.So, it must be a process that sending bulk mails to the server.Upon investing all the logs and procmail running sessions, i found one machine constantly available in the proc sessios delivering mails.I disconnected that one from network,everything seems to start ok... and all mails are flushed immediately.

There must be a worm/virus constantly doing this,(by the way the machine has AVG network edition with latest update),so i am going to erase the Windows XP OS.

It will be helpful for members.

Cheers
Gopal