Aggregates degraded due to incorrect data partition ownership after switchback
Applies to
- ONTAP 9
- MetroCluster IP
- Switchover/switchback
- Advanced Disk Partitioning (ADP)
Issue
- After swtichback aggregates are in degraded state with local plex offline:
Aggregate aggr1 (online, raid_dp, mirror degraded, fast zeroed) (block checksums Plex /aggr1/plex0 (offline, failed, inactive)
- Disk/container ownership shows that P1/P2 was not updated properly and they show as owned by the DR cluster:
Cluster_B::*> disk show -partition-ownershipDisk Partition Home Owner -------- --------- ----------------- ---------1.10.0 Container node_B_2 node_B_2 Root node_B_2 node_B_2 Data1 node_B_1 node_A_1 Data2 node_B_2 node_A_2
- EMS shows messages for left behind disks/aggregates:
Thu Dec 13 [NodeA1: config_thread: raid.mcc.leftBehindAggr:notice]: Aggregate aggrB2 was left behind during switchback.Thu Dec 13 [NodeA1: config_thread: raid.mcc.leftBehindSpare:notice]: Disk 0m.i1.2L16 was left behind during switchback.
