Giveback aggregates failed

[weiliu]

Hello.
When I upgraded the ONTAP version, the aggregation failed after "Giveback". Later, I learned that it was because there were connected sessions in CIFS. After running storage failover giveback-ofnode node_name - overline-vetoes true, the toes are successfully pooled and returned. However, the system prompts that this command will cause data loss.
How can I avoid aggregation failure in the next upgrade?

[ChLokesh · Accepted solution]

• CIFS is a stateful protocol because it maintains session information and context between client and server during a connection.
• During a giveback operation if there are any active CIFS sessions, giveback operation for the associated aggregates wont happen.
• We might see an error similar to the below one while running storage failover show on storage CLI.

Giveback of one or more SFO aggregates failed

• To prevent this from happening, we must make sure that the ONTAP upgrade is being performed during a maintenance window where none of the CIFS clients are connect to the storage.
• Before the start of the upgrade activity run a cifs session show to check if there are any connect clients. if yes, run a cifs session close against those clients to close active sessions.

Related links: 

• https://docs.netapp.com/us-en/ontap-cli/vserver-cifs-session-show.html
• https://docs.netapp.com/us-en/ontap-cli-9131/vserver-cifs-session-close.html#description