When you’re building a brand new OES 2 cluster environment or when you’re upgrading the environment from OES 1 to OES 2, in some cases the following problems will appear: cluster resources won’t migrate anymore (will go comatose everytime when trying to migrate to another cluster node) and/or NCP shares will disappear regularly.
When you’re experiencing these problems, check the cluster load and unload scripts.
If you created the cluster resources with OES 1 iManager, a line will be added which isn’t needed anymore with OES 2. You’ll have to delete this line out of the cluster load and unload scripts. This is how it’s done:
Open iManager and go to Cluster Manager / Cluster Options:
Click on the Object Selector button:
Browse the tree to the location of the cluster object and click on it:
The cluster resources are displayed.
Mark the selection box for the cluster resource of your choice and click on ‘details’:
Now you can view the details of this cluster resource. When you click on the tab ‘scripts’ you’ll see the load- and unload scripts.
In the unload script, we find the line we were looking for:
This line was needed on OES 1 but on OES 2 you’ll need to delete it.
Even if you built a brand new OES 2 cluster without upgrading from OES 1, but you built it with the OES 1 iManager, you can still find this line in the unload scripts.
Put a hash before the line so it won’t be executed anymore and click on ‘apply’ at the bottom of the screen. Now the cluster resources will migrate without any problems and the NCP shares won’t disappear anymore.