System Restart, Restore, and Sync Tips and FAQs
If you need to stop StorNext services on an Xcellis Workflow Director (by issuing the
systemctl cvfs stop command), you MUST first stop NAS failover services on NFS clusters. Otherwise, your attempt to stop StorNext services may cause StorNext services to hang or fail.
For information about stopping StorNext services on an Xcellis Workflow Director, see CR # 68037 in the Known Issues section of the Xcellis Workflow Director Release Notes.
- From the master node, log in to the console command line. See Access the Console Command Line.
- Verify whether NAS failover is enabled for NFS clusters. See View NAS Cluster Information.
> nascluster show
NAS Cluster IP: 10.65.188.63/eth0, Master: Yes, SNFS Root: /stornext/snfs1, Joined: Yes
Load balancing: Proxy-Disabled
Master IP: 10.65.188.63 snt805549-mdc-s.mdh.quantum.com
VIP: 10.65.163.197 (active, node:master) snt805549-vip1.mdh.quantum.com
1: 10.65.188.63 (Joined, MDC) snt805549-mdc-s.mdh.quantum.com
2: 10.65.173.57 (Joined, MDC) snt805549-mdc-p.mdh.quantum.com
- Disable NAS failover for NFS clusters. See Configure NAS Failover for NAS NFS Clusters.
> nascluster set nfs-ha no
To disable NFS-HA, you must unmount all NFS clients.
Are you sure you want to disable NFS-HA? (yes/No)? yes
- Stop StorNext services on the Xcellis Workflow Director. See CR # 68037 in the Known Issues section of the Xcellis Workflow Director Release Notes.
- After restarting StorNext services on the Xcellis Workflow Director, you can re-enable NAS failover for NFS clusters. See Configure NAS Failover for NAS NFS Clusters.
> nascluster set nfs-ha yes
To enable NFS-HA, you must unmount all NFS clients from all cluster nodes. Those clients must re-mount via the VIP.
Are you sure you want to enable NFS-HA? (yes/No)? yes
- Transfer NAS management services back to the preferred master node, as needed. See Transfer NAS Services to Another Node.
StorNext NAS System processes, such as snnas_controller and smbd, are managed by the following service utilities, which serve as monitoring tools for the init daemon:
- Upstart and System V on CentOS6
- systemd on CentOS7
If StorNext NAS processes are inadvertently interrupted, they have been configured to automatically restart. If you need to stop or restart processes, you can use the following commands from the rootsh shell:
- systemctl on CentOS7
CLI Example Command from rootsh: Restart smbd on CentOS6
# initctl restart smbd
smbd start/running, process 7814
When you issue the
system restart services all command on the master node of an NFS NAS cluster with NAS failover configured, a NAS failover occurs. If NFS clients have shares mounted when the NAS failover occurs, the NFS services that were transferred to the new master node may not allow I/O from the clients. In turn, the clients are unable to mount NFS shares.
To resolve this issue, you must manually restart NFS services.
- From the master node, log in to the console command line as the StorNext admin user. See Access the Console Command Line.
- At the prompt, enter the following two lines to log in to the rootsh shell:
- At the rootsh shell prompt, issue the following commands to restart NFS services on the new master node:
systemctl restart nfs-config
systemctl restart nfs-lock
systemctl restart nfs-server
- Verify that the NFS services have been restarted by issuing the following command: