DB4 ERROR DB_RUN... ?

[Igor Brezac]

On Sat, 13 Dec 2003, Thomas Besold wrote:

> Hello,
>
>
> Igor Brezac schrieb:
>
> >What version of Sleepycat do you use?
> >
> ame        : db4                          Relocations: (not relocateable)
> Version     : 4.0.14                            Vendor: Red Hat, Inc.
> Release     : 14
>

There are reports of problems with berkeley 4.0.14, although some folks
still run this version with no problems.  I do not think many people use
berkeley db for mboxlist.

I suggest that you switch to skiplist for mboxlist.  While you are at it
upgrade cyrus-(imap|sasl) to the latest versions.

> on RedHat 8.0
>
> > Try to run db_recover in
> >/var/imap/db.  This will also reset berkeley db environment.  Backup
> >before you do anything!
> >
> Have i also changed some thinks in cyrus.con?
>
> The strange is, if i have start cyrus and after 30 minutes (run
> ctl_mboxlist -r)
> the runs cyrus fine.
>
> do you think, if i run
>
> db_recover
> then is the next restart from cyrus normal?

It should.

-Igor

> Best regards
> and thanks
>
> Thomas
>
> >
> >-Igor
> >
> >On Fri, 12 Dec 2003, Thomas Besold wrote:
> >
> >
> >
> >>hello,
> >>
> >>i have installed cyrus-imapd-2.1.10
> >>
> >>if i restart cyrus comes this messages
> >>
> >>Dec 11 22:42:13 mx-test master[568]: process started
> >>Dec 11 22:42:13 mx-test master[570]: about to exec
> >>/usr/cyrus/bin/ctl_mboxlist
> >>Dec 11 22:42:13 mx-test ctl_mboxlist[570]: running mboxlist recovery
> >>Dec 11 22:44:38 mx-test deliver[1130]: connect(/var/imap/socket/lmtp)
> >>failed: Connection refused
> >>Dec 11 22:44:38 mx-test deliver[1131]: connect(/var/imap/socket/lmtp)
> >>failed: Connection refused
> >>Dec 11 22:44:39 mx-test deliver[1136]: connect(/var/imap/socket/lmtp)
> >>failed: Connection refused
> >>Dec 11 22:44:39 mx-test deliver[1137]: connect(/var/imap/socket/lmtp)
> >>failed: Connection refused
> >>Dec 11 22:44:40 mx-test deliver[1138]: connect(/var/imap/socket/lmtp)
> >>failed: Connection refused
> >>Dec 11 22:44:40 mx-test deliver[1141]: connect(/var/imap/socket/lmtp)
> >>failed: Connection refused
> >>Dec 11 22:44:40 mx-test deliver[1144]: connect(/var/imap/socket/lmtp)
> >>failed: Connection refused
> >>Dec 11 22:44:40 mx-test deliver[1147]: connect(/var/imap/socket/lmtp)
> >>failed: Connection refused
> >>Dec 11 22:44:41 mx-test deliver[1149]: connect(/var/imap/socket/lmtp)
> >>failed: Connection refused
> >>Dec 11 22:44:41 mx-test deliver[1150]: connect(/var/imap/socket/lmtp)
> >>failed: Connection refused
> >>Dec 11 22:44:42 mx-test deliver[1151]: connect(/var/imap/socket/lmtp)
> >>failed: Connection refused
> >>Dec 11 22:45:11 mx-test deliver[1161]: connect(/var/imap/socket/lmtp)
> >>failed: Connection refused
> >>Dec 11 22:45:33 mx-test master[1186]: process started
> >>Dec 11 22:45:33 mx-test master[1193]: about to exec
> >>/usr/cyrus/bin/ctl_mboxlist
> >>Dec 11 22:45:33 mx-test ctl_mboxlist[1193]: running mboxlist recovery
> >>Dec 11 22:45:33 mx-test ctl_mboxlist[570]: DBERROR db4: Recovery
> >>function for LSN 2 991413 failed
> >>Dec 11 22:45:33 mx-test ctl_mboxlist[570]: DBERROR: dbenv->open
> >>'/var/imap/db' failed: DB_RUNRECOVERY: Fatal error, run database r
> >>ecovery
> >>
> >>
> >>i have also found the page
> >>http://asg.web.cmu.edu/twiki/bin/view/Cyrus/WhyBerkeleyRecoverySoLong
> >>
> >>What you think, i
> >>
> >>The other reason is that your deliver.db may be very large
> >>
> >>I think the deliver.db isn't very large or?
> >>ll /var/imap/deliver.db
> >>-rw-------    1 cyrus    mail      1212416 Dez 12 22:37 /var/imap/deliver.db
> >>
> >>
> >>that's the directory /var/imap/db
> >>
> >>drwxr-xr-x    2 cyrus    mail         4096 Dez 12 01:52 db
> >>-rw-------    1 cyrus    mail         8192 Dez 12 01:52 __db.001
> >>-rw-------    1 cyrus    mail       270336 Dez 12 01:52 __db.002
> >>-rw-------    1 cyrus    mail        98304 Dez 12 01:52 __db.003
> >>-rw-------    1 cyrus    mail     18063360 Dez 12 01:52 __db.004
> >>-rw-------    1 cyrus    mail        32768 Dez 12 01:52 __db.005
> >>
> >>.
> >>.
> >>.
> >>.
> >>-rw-------    1 cyrus    mail     10485734 Dez  1 08:00 log.0000000015
> >>-rw-------    1 cyrus    mail     10485740 Dez 11 13:06 log.0000000016
> >>-rw-------    1 cyrus    mail      1606700 Dez 12 20:05 log.0000000017
> >>
> >>the ctl_mboxlist -r runs also over 30 minutes and i have only 21
> >>mailboxes with 1GB mails
> >>
> >>i have also changed the cyrus.conf entry
> >>from
> >> checkpoint    cmd="ctl_mboxlist -c" period=30
> >>to
> >> checkpoint    cmd="ctl_mboxlist -c" period=5
> >>
> >>
> >>delprune      cmd="ctl_deliver -E 3" period=1440
> >>
> >>
> >>
> >>
> >>
> >>could any help what's wrong
> >>
> >>Best regards
> >>
> >>Thomas
> >>
> >>
> >>
> >>
> >
> >
> >
>
>

-- 
Igor