HELP!! - No e-mail being delivered!

[Crazy Pete]

One of my users today just mentioned that she didn't receive any e-mail even though she knows people have sent her some. I also noticed the same thing, I got no mail on any of my 3 domains. Checking through exim_mainlog shows a lot of this since last night...

2003-08-10 15:23:30 19loFZ-0002lY-3z Berkeley DB error: /var/spool/exim_incoming
/db/retry: hash version 5 requires a version upgrade
2003-08-10 15:23:30 19loFZ-0002lY-3z failed to open DB file /var/spool/exim_inco
ming/db/retry: File exists
2003-08-10 15:23:30 19loGa-0002m6-FH Berkeley DB error: /var/spool/exim_incoming
/db/retry: hash version 5 requires a version upgrade
2003-08-10 15:23:30 19loGa-0002m6-FH == krazikid99@hotmail.com R=defer_router de
fer (-1): All deliveries are handled by MailScanner
2003-08-10 15:23:30 19loGa-0002m6-FH Berkeley DB error: /var/spool/exim_incoming
/db/retry: hash version 5 requires a version upgrade
2003-08-10 15:23:30 19loGa-0002m6-FH failed to open DB file /var/spool/exim_inco
ming/db/retry: File exists

I already did a /scripts/upcp and /scriptx/exim4 thinking maybe something was out of date. The only change I made was to the mail.c file adding some additional headers to track spam better. That didn't seem to add anything but mail was still coming in/out just fine. Just to make sure, I also rebuilt apache and PHP again, with no success. What can I do?? My mail stats for today show this...

Exim statistics from 2003-08-10 04:02:01 to 2003-08-10 15:14:03

Grand total summary
-------------------
At least one address
TOTAL Volume Messages Hosts Delayed Failed
Received 497KB 46 14 0 0.0% 0 0.0%
Delivered 0 0 0

Deliveries by transport
-----------------------
Volume Messages

cPanel.net Support Ticket Number:

[Crazy Pete]

Ok, apparently it's an issue with MailScanner now. Not sure what happened between last night and today but now when I follow Mahmoud's instructions for disabling MailScanner...

cp /etc/exim.conf /etc/exim.conf-mailscanner
cp /etc/exim.conf-BeforeMailScanner /etc/exim.conf
mv /etc/exim_outgoing.conf /etc/exim_outgoing.disable
Service MailScanner stop
Service exim restart

Now mail delivers fine apparently, at least it did in my test message. I'm afraid that any mail which was supposed to be delivered and wasn't has disappeared, though. Am I right about that?

cPanel.net Support Ticket Number:

[mmkassem]

The emails did not come out of the queue so they should not be lost.. R U sure that you started MailScanner and restart Exim after finishing all the steps in my How-To?

[Crazy Pete]

Originally posted by mmkassem
The emails did not come out of the queue so they should not be lost.. R U sure that you started MailScanner and restart Exim after finishing all the steps in my How-To?

I did, it was running just fine for days until last night when someone said they weren't getting e-mail. Then I realized I wasn't either. I tried restarting MailScanner and Exim a couple times with no difference. I was watching the logs as I sent myself a test e-mail and saw it come in, but then just sit there.

I even tried the auto-installer that Nick made with no difference, as soon as MailScanner started again I got the same problem. Once I uninstalled it, all the mail came through and went out that'd been queued.

It's really strange, nothing changed from when it was working til when it stopped.

cPanel.net Support Ticket Number:

[Crazy Pete]

Looks like the problem started happening today at 3:23PM. I checked the logs for the error that occured and its first instance is...

2003-08-10 15:23:23 19lDdA-0008Um-93 Berkeley DB error: /var/spool/exim/db/retry
: hash version 5 requires a version upgrade

Here are the lines around that, in case one of them indicates some cause for the problem...

2003-08-10 15:23:23 exim 4.20 daemon started: pid=3584, -q1h, listening for SMTP
on port 25 (IPv4)
2003-08-10 15:23:23 Start queue run: pid=3585
2003-08-10 15:23:23 19lW4x-0005DD-T3 Message is frozen
2003-08-10 15:23:23 19lDdA-0008Um-93 Unfrozen by auto-thaw
2003-08-10 15:23:23 exim 4.20 daemon started: pid=3595, -q1h, not listening for
SMTP
2003-08-10 15:23:23 exim 4.20 daemon started: pid=3601, no queue runs, listening
for SMTP on port 465 (IPv4)
2003-08-10 15:23:23 Start queue run: pid=3596
2003-08-10 15:23:23 19lW4x-0005DD-T3 Message is frozen
2003-08-10 15:23:23 19lDdA-0008Um-93 Spool file is locked (another process is ha
ndling this message)
2003-08-10 15:23:23 19lDdA-0008Um-93 Berkeley DB error: /var/spool/exim/db/retry
: hash version 5 requires a version upgrade
2003-08-10 15:23:23 19kYTL-00063L-17 Berkeley DB error: /var/spool/exim/db/retry
: hash version 5 requires a version upgrade
2003-08-10 15:23:23 19lDdA-0008Um-93 Berkeley DB error: /var/spool/exim/db/retry
: hash version 5 requires a version upgrade
2003-08-10 15:23:23 19kYTL-00063L-17 Berkeley DB error: /var/spool/exim/db/retry
: hash version 5 requires a version upgrade
2003-08-10 15:23:24 19lDdA-0008Um-93 ** nedpul@yahoo.com.hk R=lookuphost T=remot
e_smtp: SMTP error from remote mailer after end of data: host mx1.mail.yahoo.com
[64.156.215.6]: 554 delivery error: dd This user doesn't have a yahoo.com.hk ac
count (nedpul@yahoo.com.hk) [0] - mta193.mail.scd.yahoo.com
2003-08-10 15:23:24 19lDdA-0008Um-93 Frozen (delivery error message)
2003-08-10 15:23:24 19kYTL-00063L-17 Spool file is locked (another process is ha
ndling this message)
2003-08-10 15:23:24 19lQAF-0003L7-TL Unfrozen by auto-thaw
2003-08-10 15:23:24 19lQAF-0003L7-TL Berkeley DB error: /var/spool/exim/db/retry
: hash version 5 requires a version upgrade
2003-08-10 15:23:24 19lQAF-0003L7-TL Berkeley DB error: /var/spool/exim/db/retry
: hash version 5 requires a version upgrade
2003-08-10 15:23:25 19lQAF-0003L7-TL ** cd16682rom@yahoo.com R=lookuphost T=remo
te_smtp: SMTP error from remote mailer after end of data: host mx1.mail.yahoo.co
m [64.157.4.78]: 554 delivery error: dd This user doesn't have a yahoo.com accou
nt (cd16682rom@yahoo.com) [0] - mta109.mail.sc5.yahoo.com
2003-08-10 15:23:25 19lQAF-0003L7-TL Frozen (delivery error message)
2003-08-10 15:23:25 19lTEj-0003sP-2H Unfrozen by auto-thaw
2003-08-10 15:23:25 19lTEj-0003sP-2H Berkeley DB error: /var/spool/exim/db/retry
: hash version 5 requires a version upgrade
2003-08-10 15:23:25 19lTEj-0003sP-2H Berkeley DB error: /var/spool/exim/db/retry
: hash version 5 requires a version upgrade

cPanel.net Support Ticket Number:

[mmkassem]

Follow the How To again
(make sure you have exim_outgoing.conf)
then run /scripts/exim4

then make the modifications I asked for in /etc/init.d/exim and tell me how it goes..

cPanel.net Support Ticket Number:

[mmkassem]

make sure to /scripts/updatenow before running /scripts/exim4

what's your cpanel version?

cPanel.net Support Ticket Number:

[Crazy Pete]

Originally posted by mmkassem

what's your cpanel version?

cPanel 7.4.0-R29

cPanel.net Support Ticket Number:

[Crazy Pete]

Originally posted by mmkassem
Follow the How To again
(make sure you have exim_outgoing.conf)
then run /scripts/exim4

then make the modifications I asked for in /etc/init.d/exim and tell me how it goes..

I suppose I could try that again. I didn't update anything though prior to this error occuring so I'm not sure what exactly could've happened.

cPanel.net Support Ticket Number:

[mmkassem]

The fix is:

/usr/sbin/exim_dbmbuild /dev/null /var/spool/exim_incoming/db/retry
chmod 640 /var/spool/exim_incoming/db/*
chown -R mailnull:mail /var/spool/exim_incoming
service exim restart

cPanel.net Support Ticket Number:

[goodmove]

Mahmoud, could you put those commands inside to avoid any misunderstanding... thnx.

[mmkassem]

The fix for this problem is:

/usr/sbin/exim_dbmbuild /dev/null /var/spool/exim_incoming/db/retry
chmod 640 /var/spool/exim_incoming/db/*
chown -R mailnull:mail /var/spool/exim_incoming
service exim restart

>> a request from goodmove.

[pavemen]

I know this is pretty old, but this issue arises for me, even now. The solution posted above seems to correct it, but I am unsure exactly what is altering the settings so that internally generated mail is no longer sent (defer_router used)

Can someone put the above solutoin into a script that can be setup to test for the problem and then apply the corrections or is it safe to simply run the above steps as a cron job every morning (after cpanel updates)