cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
Showing results for 
Search instead for 
Did you mean: 

Community Tip - You can change your system assigned username to something more personal in your community settings. X

Windchill 10.2 Identical Node Cluster

ptc-6679044
1-Visitor

Windchill 10.2 Identical Node Cluster

Hi

I am trying to configure a 10.2 identical node type cluster with two nodes obviously one of which will be the active master.

As per the Advanced Config Guide (http://support.ptc.com/WCMS/files/156979/en/WCAdvDeployGuide.pdf), the 2 properties that need to be set are:

wt.rmi.server.hostname=A

wt.cache.master.slaveHosts=B, C, D

Both these properties are set. But when I start Windchill, the MethodServer will start only on that node which is started first. For Ex. if I start Windchill on Node # 1 first, the Server Manager and the MethodServer both come up on it. But on node # 2 which is started after node # 1, only the Server Manager starts and never goes beyond the following line:

INFO : wt.manager.CacheMasterNegotiator - Master stub retrieved: ServerManagerImpl$MasterServerManagerImpl_Stub[UnicastRef [liveRef: [endpoint:[xxxxxxx.xxxx.xxx:5002](remote),objID:[-2dd23186:148e7679245:-7ffd, -7282938796370177322]]]]

No MethodServer on node # 2. If I go back to node # 1 and kill the Server Manager and MethodServer processes there, a Method Server process is initiated on node # 2 almost immediately. I believe whats happening is that the ServerManager on node # 1 starts and registers itself as the master (by putting the entry in the RMISTUBS database table). Then as per the expected functioning of a 10.2 cluster, the Server Manager on node # 2 appears to be waiting for the Server Manager on node # 1 to go down by checking the RMISTUBS table every few seconds and as soon as the SM on N#1 goes down it springs into action and starts a MS.

While this logic seems to be correct, I actually need the MS to start on N#2 as well while its SM is on the lookout for the SM on N#1 to go down. Isnt't that how it is supposed to be? Am I misssing something in the configuration here? Maybe something other than the 2 properties mentioned before need to be set?

3 REPLIES 3

Hey, did you understand the problem and got any solution? I'm also facing a similar issue of Server manger hang-up in an identical cluster node setup.

Please let me know if any fix you found. Thanks

samahali
5-Regular Member
(To:psethia-3)

We are encountering the same issue for a WC11 M030 installation. Appreciate if anyone can share the solution that solved the issue. 

MikeP.
10-Marble
(To:samahali)

Same. @samahali , any chance you remember the solution 6 years back?

- Mike P.
Announcements


Top Tags