RDB HA configuration UUID does not match cluster UUID after cluster recreate

Applies to

• ONTAP 9.16
• ASA r2

Issue

This error message can occur only after this sequence of events:

1. One system in 2-node cluster fails without a takeover.
2. While node is still done, a system configuration recovery cluster recreate is done bringing the up node into quorum.
3. When the down node is recovery, it is re-joined to the cluster using, system configuration recovery cluster rejoin.
4. Cluster recovery commands disable HA by design. 
5. However, due to a bug, ONTAP may report the following:

::> cluster ha modify -configured true
  
Warning: High Availability (HA) configuration for cluster services requires
         that both SFO storage failover and SFO auto-giveback be enabled. These
         actions will be performed if necessary.
Do you want to continue? \{y|n}: y
   
Error: command failed: RDB dblade agent: RDB HA configuration UUID does not
       match cluster UUID.

6. Mgwd may report the following:

[kern_mgwd:info:4163] config_backup_restore_ops::recreate_cluster - Failed to clear rdb agent uuid in mbox. Reason: invalid operation
[kern_mgwd:info:4163] W [src/rdb/HAM.cc 240 (0x835a50400)]: interpretZapiResult: ZAPI zapi_rdb_ha_mgmt_set_cluster_uuid FAILED. Host: 127.0.0.1, Status: 1, Reason: RDB HA encountered an IO error., Error: 14307
[kern_mgwd:info:4163] 0x835a50400: 80ffff000000008a: ERR: config-backup: config_backup_restore_ops::recreate_cluster - Failed to clear cluster uuid in mbox. Reason: The RDB dblade agent encountered an IO Error in reading/writing to the SFO-based mailbox blobs.