problem with checkpoint of databases

[Eric.Doutreleau at int-evry.fr]

Hi i m using the following cyrus version

name       : Cyrus IMAPD
version    : v2.1.5 2002/06/24 19:13:27
vendor     : Project Cyrus
support-url: http://asg.web.cmu.edu/cyrus
os         : Linux
os-version : 2.4.18-5
environment: Cyrus SASL 2.1.5
             Sleepycat Software: Berkeley DB 3.3.11: (July 12, 2001)
             OpenSSL 0.9.6b [engine] 9 Jul 2001
             CMU Sieve 2.2
             TCP Wrappers
             lock = flock
             auth = unix
             idle = poll
             mboxlist.db = db3
             subs.db = flat
             seen.db = flat
             duplicate.db = db3-nosync
             tls.db = db3-nosync

and i have from time to time big problems with checkpoint

Indeed for a reason i can't determine the process ctl_cyrusdb -c never
finish.
As this process is launched every 30 minutes i got several dozens 
of that process before becoming aware of that problem.

But the worse happens when i have to restart the server.
Indeed as the checkpoint didn't succeed there were a lot of file
named log.0000000xxx and the server become unavailable a long
time waiting for the command ctl_cyrusdb -r to finish.

it takes around one hour per day of failure of checkpointing to restart.

-Is there a way to speedup that process?

-will the changing the format of the database  help to solve that
problem?

-what could cause the checkpoint to fail?

 
Thanks in advance for any help.



-- 
Eric Doutreleau
I.N.T			| Tel	: +33 (0) 160764687
9 rue Charles Fourier	| Fax	: +33 (0) 160764321
91011 Evry   France	| email	: Eric.Doutreleau at int-evry.fr