bvIssue : In Oracle Applications post cut over we could see some of the nodes getting abandoned.
ERROR: Following abandoned nodes are available in the configuration
Node-Name1
Node-Name2
Above (Node-Name1 and 2 have been abandoned from the cluster).
adstartall/adstpall.sh will also fails in that abandoned node with below error
ERROR: This is an abandoned/lagging node. Services cannot be started up on this node.
RCA : This is part of the adop patching guidelines for making things consistent across the nodes. This happens if any patch cannot be applied for some issue in some nodes of the cluster and still we are proceeding with cutover and then if prompt to proceed with option “Y” will definitely abandon the nodes where it is not completed.
Solution :
Make sure the patch has been applied across the nodes.
You can cross verify the information for the adop -status and the status column of ad_adop_session_patches and make sure it is consistent across the nodes.
Remedy will be to delete the abandoned node and add them back to cluster using normal cloning method.
After adding the nodes we need to run fs_clone/prepare then only the status gets updated in the seeded table on the abandoned node information.
From Oracle Document :
In a Multi node ADOP cycle, if any operation (prepare or apply or cutover) fails on any non-primary node (code issue or configuration issues), and the user ignores the failure, that node becomes abandoned at the end (After cutover on primary node).Once a node is marked as abandoned, user has to delete that node from the configuration, to move forward with other available nodes
Reference :
How to Restore An Abandoned Node in Oracle E-Business Suite Release 12.2 (Doc ID 1677498.1)
Through... my glasses :) 