Multiple Issues with dovecot/postfix with mail log errors

I'm having some e-mail accounts that are unable to retrieve e-mail. Some of the accounts are receiving e-mail just fine. Others appear to be receiving mail but it does not show up in their mail folders nor are they able to retrieve it.

This is the message I sent which appears to be delivered via procmail but it does not show in the users mailbox.

Jun 30 21:08:44 server1 postfix/qmgr[29796]: 12A2D2410090: from=jmhug@cox.net, size=1282, nrcpt=1 (queue active) Jun 30 21:08:44 server1 postfix/smtpd[31607]: disconnect from eastrmmtao102.cox.net[68.230.240.8] // Jun 30 21:09:05 server1 postfix/local[31931]: 12A2D2410090: to=nick.paradigmgardens@heliosone.com, orig_to=nick@paradigmgardens.com, relay=local, delay=21, delays=0.36/0.01/0/21, dsn=2.0.0, status=sent (delivered to command: /usr/bin/procmail-wrapper -o -a $DOMAIN -d $LOGNAME) Jun 30 21:09:05 server1 postfix/qmgr[29796]: 12A2D2410090: removed

However others are producing an error in Dovecot. Here is a piece of the log file

Jun 30 20:55:32 server1 dovecot: pop3-login: Login: user=<omaha.paradigmgardens>, method=PLAIN, rip=::ffff:189.115.134.9, lip=::ffff:72.51.44.79 Jun 30 20:55:32 server1 dovecot: POP3(omaha.paradigmgardens): Getting size of message UID=7954 failed Jun 30 20:55:32 server1 dovecot: POP3(omaha.paradigmgardens): Getting size of message UID=7954 failed Jun 30 20:55:32 server1 dovecot: POP3(omaha.paradigmgardens): Couldn't init INBOX: Can't sync mailbox: Messages keep getting expunged

Then, some users are sending/receivng mail just fine. This is occurring on multiple domains on the server. It is also occurring after a rather bizarre spike in processes on the server eating all my memory. I had to reboot the server. I just updated the clamav installation prior to all of this bizarre behavior.

Any help would be great. You know how people hate not getting e-mail.

Status: 
Active

Comments

Howdy -- I'll have to do a little digging regarding those POP3 errors you're seeing.

Regarding the other messages though, that appear to be delivered -- is anything showing up in the procmail log, over in /var/log/procmail.log?

It may offers some hints on what's going awry with those.

OK- thanks for the tip on procmail.log. Tons of errors related to being unable to create tmp directories e.g.

Time:1246349245 From: To: User:omaha.paradigmgardens Size:934 Dest:/dev/null Mode:Spam LibClamAV Error: cli_cvdload(): Can't create temporary directory /tmp/clamav-f33f1047499c7bb8f6204af6c5ab9480 LibClamAV Error: Can't load /var/lib/clamav/main.cvd: Can't create temporary directory ERROR: Can't create temporary directory LibClamAV Error: cli_cvdload(): Can't create temporary directory /tmp/clamav-5a8f017e0f7f5b17c0405b7f43945752 LibClamAV Error: Can't load /var/lib/clamav/main.cvd: Can't create temporary directory ERROR: Can't create temporary directory Subject: [SPAM] Folder: /dev/null

Of course no clamav-f33f1047499c7bb8f6204af6c5ab9480 directory exists in tmp although there are many others. All clamav-* directories in /tmp were created in the last 24 hours

I'm also getting: WARNING: System limit for file size is lower than engine->maxscansize

I changed to the clamdscan and have not seen the WARNING message and now procmail.log is showing this

Time:1246429892 From: To: User:comments.paradigmgardens Size:934 Dest:/dev/null Mode:Spam Subject: [SPAM] Folder: /dev/null

so apparently the spam is being properly filtered now (?)

I'm still having issues getting legitimate mail through to some users. Notably, if a mailbox is set to forward to another e-mail account off the server then the e-mails are passing just fine. They are not showing up in /Maildir though for the forwarded account even though it is set to deliver to the user mailbox and forward as well.

Jul 1 01:44:39 server1 postfix/local[14264]: C330B2410090: to=comments.paradigmgardens@heliosone.com, orig_to=comments@paradigmgardens.com, relay=local, delay=15, delays=0.35/0.01/0/15, dsn=2.0.0, status=sent (forwarded as 149952410092) Jul 1 01:44:39 server1 postfix/qmgr[29796]: 149952410092: from=jmhug@cox.net, size=2267, nrcpt=1 (queue active) Jul 1 01:44:39 server1 postfix/qmgr[29796]: C330B2410090: removed // Jul 1 01:44:44 server1 postfix/smtp[14276]: 149952410092: to=jmhug@cox.net, orig_to=comments@paradigmgardens.com, relay=mx.west.cox.net[68.6.19.3]:25, delay=19, delays=15/0.01/4.3/0.14, dsn=2.0.0, status=sent (250 2.0.0 AWkf1c02b0kngF805Wkkzq mail accepted for delivery) Jul 1 01:44:44 server1 postfix/qmgr[29796]: 149952410092: removed

Well, that's a little strange :-)

What's the output of these commands:

  1. df -h

  2. grep '^[^#]' /etc/security/limits.conf

  3. mount

That should be a good start, we'll work from there ;-)

df -h Filesystem Size Used Avail Use% Mounted on

/dev/sda3 147G 8.8G 131G 7% /

/dev/sda1 99M 9.9M 84M 11% /boot

tmpfs 505M 0 505M 0% /dev/shm

grep '^[^#]' /etc/security/limits.conf

apache hard cpu -1

apache hard nproc -1

apache hard as -1

mount

/dev/sda3 on / type ext3 (rw,grpquota,usrquota)

proc on /proc type proc (rw)

sysfs on /sys type sysfs (rw)

devpts on /dev/pts type devpts (rw,gid=5,mode=620)

/dev/sda1 on /boot type ext3 (rw)

tmpfs on /dev/shm type tmpfs (rw)

none on /proc/sys/fs/binfmt_misc type binfmt_misc (rw)

sunrpc on /var/lib/nfs/rpc_pipefs type rpc_pipefs (rw)

Well I just went back and disabled quotas in the disk quotas link in webmin. I think that may have resolved the issue with the remaining mail accounts. I do know that I had quotas turned off at one time in virtualmin on the specific accounts that were having troubles because they were hitting their quota limits regularly. When I ran checkquotas it showed a number of users over the quota limit.

So in sum switching to clamd and disabling quotas appears to have resolved the issue. I still need to figure out why this happened to begin with.

Were there any updates to webmin/virtualmin that would have turned quotas back on by default at the user level?

I'm going to keep a close eye on the situation today and will follow up with a confirmation if all continues well. Thank you for the help.

As an aside, what would have happened to e-mails sent to boxes that were over the quota limit? Are they deleted by the system?

If an account is at or over their quota, Postfix would generally reject the email with a message back to the sender stating that the account is at/over quota. I think :-)

was this ever resolved? I'm also running into the same quota issues.

mattmm - could you file a separate bug for your issues? They may be un-related to this bug ..