Recovering from a failed db2iupdt -add or -drop operation in a Db2 pureScale environment

There are a number of reasons why a db2iupdt -add or -drop operation might fail, leaving the Db2 pureScale environment topology in an inconsistent state.

If the problem requires a hardware fix (for example, a host failure), finish that task first. Once the hardware is up and running, you can use the db2iupdt command with the -fixtopology option to automatically either roll back the previous db2iupdt -add operation, or finish the previous db2iupdt -drop operation.

If the operation that failed was an attempt to drop a member host with multiple Db2 members, then the db2iupdt -fixtopology command will only finish the previous db2iupdt -drop of the one specific member that failed. It might be necessary to rerun the original db2iupdt -drop command to finish dropping the other Db2 members on the member host.