Hyper-V environment loses access to volumes after cluster switch restart
Applies to
- ONTAP 9
- Hyper-V
Issue
- After restarting one side of the Cluster Switch, the Hyper-V environment loses access to volumes.
- Other environments, such as vSphere and user PCs, do not encounter this issue.
- The Hyper-V environment uses SMB3 Continuous Availability (CA) shares.
- Observations include:
- Hyper-V connections show increased IdleTime in `
cifs session show`. - IdleTime exceeds 10 minutes and continues to increase until the Hyper-V servers are restarted.
- CIFS session from Hyper-V shows increased idle time and does not reconnect automatically until Hyper-V is rebooted.
- Hyper-V connections show increased IdleTime in `
- Log Output:
Thu Mar 13 13:39:49+0900 [XXX:kernel:csm.createSessionFailed:debug]: ClusterSessionManager(CSM) failed to create session with transport type CT, session tag DEFAULT, record state ACTIVE, CSM error CSM_CONNABORTED, low-level error, socket error 0, and TLS error 0.
Thu Mar 13 13:39:49+0900 [XXX:wafl_exempt15:lmgr.session.failed:info]: Dropped session-scoped locks from session due to failed session.
