When one of the slices that configure the netmirror volume is in ACTIVE state and the other slice is in other than ACTIVE state, do not stop or restart the node that is connected to the disk where the ACTIVE slice exists. If the node is stopped or restarted, the ACTIVE slice does not exist in the netmirror volume. In this case, the service will stop.
To stop or restart the node, restore the status of the slice on the other node.
See below for how to restore the slice status.
If the slice of the other node is in NOUSE state, execute the sdxswap -I command or use [Operation]:[Restore Physical Disk] of GDS Management View to restore the disk of the other node.
If the slice of the other node is in INVALID state, refer "(1) The netmirror slice that configures the netmirror volume is in INVALID state." in "D.1.6 Errors in the Mirroring Among Servers" to restore the slice status.
Even if all the slices that configure the net mirror volume are in ACTIVE or STOP state, the other slice becomes INVALID state when both nodes are restarted. This state is restored when the cluster application is started.
