[midPoint] Duplicate node in the Cluster

[Dick Muller]

Hi Pavol,


I deleted the two nodes, recreated the Quartz tables and only started the node that was having a problem.

Also the same thing.


Strangest thing is that in the log it states that it notices an existing node with that name and a specific OID, but can't find it.

And then stops the Task manager.


This is the peace of the log from the latest attempt:


2016-07-01 08:41:30,481 [] [ClusterManagerThread] ERROR (com.evolveum.midpoint.task.quartzimpl.cluster.NodeRegistrar): The record of this node cannot be read (OID 42745a2a-6010-4716-9718-b0de1c5aa436 not found), but another node record with the name 'NA1-IDM-PRD01' exists. It seems that in this cluster there are two or more nodes with the same name 'NA1-IDM-PRD01'. Stopping the scheduler to minimize the damage., reason: Object of type 'NodeType' with oid '42745a2a-6010-4716-9718-b0de1c5aa436' was not found. (class com.evolveum.midpoint.util.exception.ObjectNotFoundException)


Thanks,


Dick

________________________________
Van: midPoint <midpoint-bounces at lists.evolveum.com> namens Pavol Mederly <mederly at evolveum.com>
Verzonden: vrijdag 1 juli 2016 10:26:39
Aan: midpoint at lists.evolveum.com
Onderwerp: Re: [midPoint] Duplicate node in the Cluster


Hello Dick,


this is really strange. If - after your cleanup procedure - you start only that one node (NA1-IDM-PRD01), does the error still occur?


Pavol

On 01.07.2016 10:15, Dick Muller wrote:

Hi,


I run two MidPoint servers in a cluster and that worked fine, but since the last two weeks the first server is complianing that there is a duplicate nodeId.

That is strange because the server is not renewed. I deleted the nodes in the Task manager, rebuild the quartz database in MySQL and restarted the tomcat instances.

But still no luck, the same error comes back.


In the idm.log the following error is raised:


2016-07-01 08:07:05,974 [] [ClusterManagerThread] ERROR (com.evolveum.midpoint.task.quartzimpl.cluster.NodeRegistrar): The record of this node cannot be read (OID 9eca6710-ec8f-4380-a3e4-62de4c97deb4 not found), but another node record with the name 'NA1-IDM-PRD01' exists. It seems that in this cluster there are two or more nodes with the same name 'NA1-IDM-PRD01'. Stopping the scheduler to minimize the damage., reason: Object of type 'NodeType' with oid '9eca6710-ec8f-4380-a3e4-62de4c97deb4' was not found. (class com.evolveum.midpoint.util.exception.ObjectNotFoundException)


Regards,

________________________________

Dick Muller

Systems Engineer

Delftechpark 37i
2628 XJ Delft
d: +31 88 2682586
m: +31 6 46477690

[http://client.tahzoo.com/tahzoo/logo_blue_100w.png]<http://www.tahzoo.com/>




_______________________________________________
midPoint mailing list
midPoint at lists.evolveum.com<mailto:midPoint at lists.evolveum.com>
http://lists.evolveum.com/mailman/listinfo/midpoint


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.evolveum.com/pipermail/midpoint/attachments/20160701/a0ccdd7a/attachment.htm>