Search Docs...
⌘ K
After performing the upgrade operation, it is necessary to update I/O rerouting and SCVM auto-restart scripts.
Information:
If you upgrade the cluster using the upgrade center, the I/O rerouting and SCVM auto-restart scripts will be updated automatically during the upgrade process—no manual update is required.
Execute the following command in any SCVM of the cluster, and the system will automatically check and upgrade the I/O rerouting script on each ESXi node in the cluster.
zbs-deploy-manage update_reroute_version
The I/O rerouting scripts for all ESXi nodes are upgraded to the latest version: The page displays the message reroute scripts is already the latest version in esxi xxx.xxx.xxx.xxx for each node, where xxx.xxx.xxx.xxx represents the management IP of the node in the cluster. After verifying the I/O rerouting scripts on all nodes, the page will display the message Success update latest vmware reroute script to all esxi done, meaning you are allowed to proceed to the next step.
Some ESXi nodes' I/O rerouting scripts are pending upgrade to the latest version: The page displays after upgrade zbs, generate latest scripts esxi need, which indicates that this node's I/O rerouting script is not the latest version, and the system immediately initiates automatic upgrade. After all nodes automatically upgrade the I/O rerouting scripts, the page will display the message Success update latest vmware reroute script to all esxi done. To ensure all nodes in the cluster have completed the I/O rerouting script upgrade, it is recommended to re-run the command zbs-deploy-manage update_reroute_version. If the expected result matches the output All ESXi nodes have upgraded the I/O reroute script to the latest version, you can proceed to the next step.
Run the command ps -c | grep scvm_failure | grep -v grep on all ESXi nodes. If the page includes the reroute.py script, it indicates a successful I/O rerouting script upgrade.
Execute the command zbs-deploy-manage update-scvm-autostart on any SCVM in the cluster. If the output meets the following requirements, it indicates that all ESXi nodes' SCVM auto-restart scripts have been updated.
The page does not display esxi scvm uuid: xxx sn miss, please check scvm info, where xxxx represents the scvm uuid;
The page contains the message Success update latest vmware scvm autostart script to all esxi done.
On all ESXi nodes, run the command cat /etc/rc.local.d/local.sh | grep scvm. If the page contains sh /vmfs/volumes/xxxx/vmware_scvm_autostart/scvm_autostart.sh & where xxxx indicates the scvm uuid, continue to the next procedure. If not, it means the SCVM auto-restart script update has failed, and in this case, you need to contact Arcfra technical support for assistance.
Execute the command ls -lh /vmfs/volumes/*/vmware_scvm_autostart/scvm_autostart.sh | grep -v datastore. If the output includes the scvm_autostart.sh file, it indicates that the SCVM autostart script is installed successfully. If not, it means the SCVM auto-restart script update has failed, and in this case, you need to contact Arcfra technical support for assistance.