Wolf Siedler wrote:
Hi!
I am sorry for the vague subject but I don't know a better way to
describe my problem. I am still studying Fedora/389 Directory Server.
I am running a Fedora-DS (1.1) master on CentOS 5.4 which replicates to
several consumers (no master-master replication).
1.1? rpm -qi 389-ds-base
32-bit or 64-bit?
Today, the master suddenly stopped working. I can restart it, but it
never starts listeing on ports 389/636.
"service dirsrv status" claims after several minutes that the service is
dead.
When attempting a start by
/usr/lib/dirsrv/slapd-admin01/start-slapd -d 1
I get after a few seconds a seemingly endless loop of this message:
NSMMReplicationPlugin - agmt="cn=nagios" (rtbkk:389): Consumer failed to
replay change (uniqueid (null), CSN (null)): Bad parameter to an ldap
routine. Will retry later.
Usage of terms nagios and rtbkk point me indeed to one of our
replication agreements.
Can somebody explain what that means?
Looks like you have some bogus ldap operation that it is attempting to
replay.
Use cl-dump to dump your changelog - look for a bogus operation.
Is it possible that a replication agreement fails and then the
server
attempts/repeats over and over until all resources are exhausted?
Can anybody advise how I could overcome this?
I suppose you could use db2ldif to dump your database, then ldif2db to
reinit. Then you'll have to reinit all of your consumers.
Needless to say, I would appreciate any advice/pointers (such
incidents
always seem to happen at the wrong time).
I've never seen this happen before. If you can find a bogus operation
in your changelog, we might be able to work backwards through the access
log to find the source.
Regards,
Wolf
--
389 users mailing list
389-users(a)lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/389-users