I have seen this when you do not use the /clustername:ClusterName 
/clusternode:YES for all the cluster services.  Note fill ClusterName with the 
real name.  Also, make sure you use /clusternode:NO to create the services for 
the non-cluster nodes.  Create the services for the cluster nodes last.  If you 
create the services for the local node last they can remove the password from 
the clusternodes.  This may be due to another problem being reported on the 
list.

Anyway, I had to script our cluster TSM installs and after 2 or 3 days of 
trying everything I found that the following works every time.

1) use cluster.exe to remove the cluster resources. At the very least make sure 
you remove the registry replication (checkpoint) from any resoruces that is 
replicating the TSM password key.  This is important because the checkpoint 
file will only be updated when the resource is online and if you remove the 
services it will bring the resource offline.
2) Remove all TSM services.
3) Create all cluster services using dsmcutil don't start them (/startnow:no)
4) Use dsmcutil updatepw to ensure the password is in the registry correctly.  
Yes I know you just passed it when you created the services but trust me this 
helps
5) Start the services (net start will work)
6) Create the local services don't start them (/startnow:no).  Now I have 
actually had this give me no problem with starting them but I'm following the 
same standard to avoide future problems.
7) use dsmcutil updatepw for this node
8) net start those
9) use cluster.exe to create the cluster resrouces or just at the replication 
(checkpoint) if you only removed it.

"Boileau, Chantal" <Ch.Boileau AT SAQ.QC DOT CA> wrote:
I have a windows 2003 cluster with 2 physicals nodes and 4 virtuals
nodes. I experiencing some problem to backup one of the vitual node.

The problem is

on the dsmsched.log
02/16/2005 23:29:19 Querying server for next scheduled event.
02/16/2005 23:29:19 Node Name: CCDMPSB013
02/16/2005 23:29:19 Please enter your user id : ANS1353E Session
rejected: Unknown or incorrect ID entered
02/16/2005 23:29:19 Scheduler has been stopped.

dsmerror.log
02/16/2005 23:28:36 sessOpen: Error 137 from signon authentication.
02/16/2005 23:28:37 ANS1353E Session rejected: Unknown or incorrect ID
entered
02/16/2005 23:29:09 Trying port number 56582
02/16/2005 23:29:09 Trying port number 56583
02/16/2005 23:29:09 Trying port number 56584
02/16/2005 23:29:09 Trying port number 56585
02/16/2005 23:29:09 Obtained new port number on which to listen.
02/16/2005 23:29:10 sessOpen: Error 137 from signon authentication.
02/16/2005 23:29:10 ANS1353E Session rejected: Unknown or incorrect ID
entered
02/16/2005 23:29:14 Trying port number 56582
02/16/2005 23:29:14 Trying port number 56583
02/16/2005 23:29:14 Trying port number 56584
02/16/2005 23:29:14 Trying port number 56585
02/16/2005 23:29:14 Obtained new port number on which to listen.
02/16/2005 23:29:14 sessOpen: Error 137 from signon authentication.
02/16/2005 23:29:14 ANS1353E Session rejected: Unknown or incorrect ID
entered
02/16/2005 23:29:19 Trying port number 56582
02/16/2005 23:29:19 Trying port number 56583
02/16/2005 23:29:19 Trying port number 56584
02/16/2005 23:29:19 Trying port number 56585
02/16/2005 23:29:19 Obtained new port number on which to listen.
02/16/2005 23:29:19 sessOpen: Error 137 from signon authentication.
02/16/2005 23:29:19 ANS1353E Session rejected: Unknown or incorrect ID
entered

In the cluster registry
(HKEY_LOCAL_MACHINE/software/ibm/adsm/currentversion/backupclient/nodes/ccdm
psb013),

I found 2 password entry one for TSM_PROD that is the name we put at
installation time, and an other that is garbadge caracter ???

If some body have any idea, it will help Thanks.




> Chantal Boileau
> Analyste Informatique Stockage
> Infrastucture de Service
> SAQ
> * : 514-253-6232
> * : 514-864-8471
> * mailto:ch.boileau AT saq.qc DOT ca
> * :http://www.saq.com
>
>
>

                
---------------------------------
Do you Yahoo!?
 Yahoo! Search presents - Jib Jab's 'Second Term'