Email stays in /var/spool/exim.in/input

T

terrabel

Verified User
Joined
May 6, 2008
Messages
20
The server is running on CentOS, we use installscript from web for host to install / update clamav / spamassassin and mailscanner.
The mails are being scanned by the spamassassin according to us see:
074 X-Spam-Checker-Version: SpamAssassin 3.3.0 (2010-01-18) on server.domein.nl
015 X-Spam-Level:
132 X-Spam-Status: No, score=-1.2 required=5.0 tests=BAYES_00,HTML_MESSAGE,
RCVD_IN_DNSWL_NONE,SPF_SOFTFAIL autolearn=no version=3.3.0
Click to expand...
but all the mails are stuck in a queue in the /var/spool/exim.in/input directory.
I cannot get them to go further in the progress.

what actions can I take?

Also there are a lot of mails in /var/spool/exim/input
Is that normal? and some are very old (over 1 year).
 
Last edited:
Hi,

Have you tried to force delivery? What are the errors?

some helpful commands in your case:
Code: 
exim -bp
	mailq --- The mailq is relevant as it gives your the email IDs.
exim -M emailID
	force delivery of one message
exim -qf
	Force another queue run
exim -qff
	Force another queue run and attempt to flush frozen messages
exim -Mvl messageID
	View Log for message
exim -Mvb messageID
	View Body for message
exim -Mvh messageID
	View Header for message
exim -Mrm messageID
	ReMove message (no errors sent)
exim -Mg messageID
	Give up and fail message, message bounces to sender

Have you restarted the ClamAV service?
 
restarted server

Hello Ranz,
I even restarted the server.
I don't know where to start in debugging this.
I tried your -qf again and got this:
1O3djU-0005ua-Mc User 0 set for local_delivery transport is on the never_users list
But for the rest the mails stay where they are (in the queue).
 
Update

I have reinstated a safe version of exim.conf and now all new mail is going just fine, but their are still a lot of mails in the /var/spool/exim.in/input directory.
How do I get those to be delivered?
 
1) See my response to your post in the other thread.

2) There's no reason for most of us to ever use mailscanner with exim; the exim.conf file will do, can do, and should do, everything mailscanner will do, and we support the exim.conf file :).

3) What do you mean by a safe version of exim.conf? Why do you consider the one you were using as unsafe? Where did you get it?

4) Are they frozen? Are your queue runners trying to deliver them (if so you'll see evidence in your exim mainlog)? You can learn how to manage the queue from a very good page on Brad The Mad's site.

Jeff
 
You could also try to force delivery of that specific message, check errors, if it fails - perhaps even do an "exim -mg MESSAGID" on it - so that your queue can proceed.
 
Exim changes

I changed yesterday (3.00 am for me) the exim.conf file and restarted the machine. Immediately after that the new mails did come through. So after testing I looked in the mail queue's and found that the old queue's were not being handled any more. So I posted my question and went to bed....
This morning I can conclude that the new mails kept going as they should.
Thanks for the replies, I will try to answer them here. (and make new questions)...
1) See my response to your post in the other thread.
Click to expand...
I have made my apologies, I know it is not good to double post, but after so many hours of searching, I wanted to get a solution...
2) There's no reason for most of us to ever use mailscanner with exim; the exim.conf file will do, can do, and should do, everything mailscanner will do, and we support the exim.conf file .
Click to expand...
I installed the server many years ago and for some reason installed the mailscanner at that time. It seemed to work for me then. I recall a problem with it and probably I have changed a few elements in the config file without remembering (and documenting) that change. But three days ago I had the problem with ClamAV being outdated, after repairing that (eventually with the update script) I didn't have any mail going out of the queue. So that is why I looked for problems with MailScanner again.
3) What do you mean by a safe version of exim.conf? Why do you consider the one you were using as unsafe? Where did you get it?
Click to expand...
Probably just wrong choice of words. I don't exactly mean safe version, but I had an old config in the /etc directory called exim.safe. I took that version and commented the mailscanner queue out and the spamassassin and clamav options. So the other wasn't unsafe, I just reused an older version of the conf. After that the new mail just runs fine.
4) Are they frozen? You can learn how to manage the queue from a very good page (http://bradthemad.org/tech/notes/exim_cheatsheet.php) on Brad The Mad's site.
Click to expand...
The problem is that the exim commands will not show the files that are in my queue in the /var/spool/exim.in directory. I think it only shows the files that are in the /var/spool/exim. How do I manage the messages in the exim.in directory?
A PM from John also states the following:
"exim.in" isn't a default exim path (it's not used).. it must be related to ClamAV.

Likely, you'll need to copy the messages from /var/spool/exim.in/* into /etc/spool/exim/*, in their same respective directories, eg:
/var/spool/exim/input/*
/var/spool/exim/msglog/*

Thank you,

John
Click to expand...
I just don't know if I can cp the files without any problem. Just kill exim and start copying???

thanks in advance for a reply.
 
mail queue copied

I have just killed exim and then copied the complete mail queue from the exim.in/input and exim.in/msglog to the exim/input and exim/msglog. Chown to give the user mail access to the files and restarted exim.
It seems to work fine. Mail is coming in piece by piece.
It might take some time to completely empty the queue, but it seems to be ok.
Thanks everyone for the hints and answers.
 
You must log in or register to reply here.
Back
Top