#HA netapp for volumes
1 messages · Page 1 of 1 (latest)
HA netapp for volumes
not this way, no. You have two possiblilities:
- create a backend that is already HA-protected on the storage system (e.g. MetroCluster, SVM-DR or SnapMirror) and use that for your StorageClass
- create two volumes on both backends you have, and have the application write to both simultaneously (depends on the application obviously)
Ok.. Thanks @open ocean ... Now i understood
Hi @open ocean , for PVC migration between one netapp storage to another netapp storage as existing storage is end of support... What is the recommended procedure?
my suggestion would be to keep the cluster as is and just replace the nodes with new ones... then you don't need to do anything
i.e. join new nodes to the existing cluster, move all volumes and LIFs to the new nodes, then remove the old nodes from the cluster
if scaling the existing cluster is not possible, need to figure out who has the responsibility to migrate those volumes.
Also an important part of the procedure directly depends on the Trident driver type.
some ideas:
- SVM DR (all volumes at once, but lots of caveats)
- snapmirror managed by the storage team + trident volume import
- trident integration with snapmirror (migration volume by volume managed by the K8S users)
- Backup&Restore (with Trident Protect for instance)
note that if you run ONTAP-NAS-ECO, snapmirror is out of the picture