Hello,
Our openldap is sometimes long to stop and init scripts failed to stop.
Apr 20 09:10:47 ldap1 slapd[15105]: daemon: shutdown requested and initiated. Apr 20 09:10:47 ldap1 slapd[15105]: slapd shutdown: *waiting for 0 operations/tasks to finish* Apr 20 09:*12:46 *ldap1 slapd[15105]: slapd stopped.
It tooks 2 minutes to stop. So the init scrit failed and restart is not safe. An auto update on ubuntu leaves slapd in a strange state once and the cause was that long wait for 0 task to finish. Normaly, zero means immediate. Here two minutes! We have 5 glued bdb databases. Is it due to the bdb close for these databases? Is there a way to get things more reliable
OpenLDAP: slapd 2.4.23 (Mar 30 2011 16:20:41) $ ^Ibuildd@crested :/build/buildd/openldap-2.4.23/debian/build/servers/slapd Description: Ubuntu 10.10 2.6.35-27-server #48-Ubuntu SMP Tue Feb 22 21:53:16 UTC 2011 x86_64 GNU/Linux libdb-4.8.so
Thanks
Dom
On Wednesday, 20 April 2011 09:28:26 LALOT Dominique wrote:
Hello,
Our openldap is sometimes long to stop and init scripts failed to stop.
Apr 20 09:10:47 ldap1 slapd[15105]: daemon: shutdown requested and initiated. Apr 20 09:10:47 ldap1 slapd[15105]: slapd shutdown: *waiting for 0 operations/tasks to finish* Apr 20 09:*12:46 *ldap1 slapd[15105]: slapd stopped.
It tooks 2 minutes to stop. So the init scrit failed
Then the init script is partly at fault.
and restart is not safe. An auto update on ubuntu leaves slapd in a strange state once and the cause was that long wait for 0 task to finish. Normaly, zero means immediate. Here two minutes! We have 5 glued bdb databases. Is it due to the bdb close for these databases? Is there a way to get things more reliable
Supply your slapd configuration. It could be that you aren't checkpointing frequently enough.
Regards, Buchan
openldap-technical@openldap.org
-
Buchan Milne -
LALOT Dominique