Urgent assistance required

[Paul Raines]

This happened to me all the time with the default DB4 databases
on our RHEL Linux server.

The paths below may be different on your OSX boxes, but to fix
we would do:

service cyrus-imapd stop
/bin/rm /var/lib/imap/db/log.*
su - cyrus -c "/usr/lib64/cyrus-imapd/ctl_cyrusdb -r"
service cyrus-imapd start


NOTE, depending on how you get mail injected into the Cyrus system,
your incoming mail from during the outage may be sitting somewhere
and need assistance.  For example, we run or incoming mail through
procmail which calls 'deliver' to put mail into Cyrus, but if that
fails, put the mail into normal UNIX flat space in /var/spool/mail.
After a failure, I would have to manualy re-inject that mail
from /var/spool/mail into Cyrus.


Eventually we seem to have gotten rid of the problem after we converted the 
databases to 'skiplist' via putting these lines in /etc/imapd.conf:

duplicate_db: skiplist
tlscache_db: skiplist

Then stop cyrus-imapd, delete /var/lib/imap/deliver.db and
/var/lib/imap/tls_sessions.db, and then start cyrus-imapd up



-- 
---------------------------------------------------------------
Paul Raines                email: raines at nmr.mgh.harvard.edu
MGH/MIT/HMS Athinoula A. Martinos Center for Biomedical Imaging
149 (2301) 13th Street     Charlestown, MA 02129	    USA