If the system is using encryption it may be that the encryption devices are not all started properly.
If the drives are shared between the systems, a normal reboot may not reset the disks completely as you may expect with stand-alone solutions.
To correct this symptom you will need to log into the system CLI and run the following command as root:
zpool clear (PoolName)
The "cryptopenall" may takes some time to run so check status with this command:
It is safe to run these commands at any time. Note there may be a brief i/o delay while the zpool clear is run, but it is usually safe to do this with client access.
If this does not work you may want to shut down both nodes and then booting them from a cold state.
Additional details can be found at this link: