netraid upgrade urgent issues

[sergio tagliapietra]

Hi,
i did the bad mistake of upgrading netraid-4m software on my netserver lc2000r 2nodes cluster.
I followed all the instructions given by hp: i backed up both of the nodes, moved all services to first node, then on second node i removed cluster component (prepare for upgrade option set), removed FAST, updated the driver, updated the firmware, reinstalled FAST and cluster component.
Everything seemed ok so i wanted to move all services back to updated node and start with the other.
I successfully moved first group of services (mainly file shares) than tried to move second group.
This second group contain a netraid4m diskset resource that mounts two partitions of the same mirrored disk, Q (quorum) and S (for dhcp, wins and spool dir).
Q comes online without problems while S cannot mount, so wins, shcp and print spooling don't work.
This is the error HPNvdisk puts in system log...

Tipo evento: Avviso
Origine evento: HPNvdisk
Categoria evento: Nessuno
ID evento: 8000
Data: 30/07/2002
Ora: 19.31.10
Utente: N/D
Computer: NODO-B
Descrizione:
Resource Error : Cluster Resource Error. Controller Error String: Not all drive letters are restored. : There was no file system found on the specified container.
Dati:
0000: 06 00 00 00 ....

while there are no errors reported by HPNvdisk in cluster.log.
Is it a "common" issue?
Thank you for any suggestion.

[Marino Meloni_1]

it appear like if a ressource cannot go on line for some dependences, check in cluster administrator if all the dependencies of this ressources are available.
it it was a netraid 4m or a disk problem, all disk set would be unavailable and phisical disk will be in detached in the Fast utility
bye

marino

[sergio tagliapietra]

i really think the problem is not in the services configuration, 'cause all the services stops with the same error (invalid path) and drive "S" of the diskset is not mounted.

[Marino Meloni_1]

sergio,
did you see it in on the first node
did you see the partition in disk administrator, or did you see unalocated space?
I never see it before.
bye
marino

[sergio tagliapietra]

Thank you for the help, but i just fixed the problem.
I had to give a 'controller rescan' command in fast cli after moving the diskset to second node... after this the HPNvdisk resource mounted disk S without problems.
Thanks :-)

[Marino Meloni_1]

Nice if you have fixed
and Thanks to post us the resolution, this will help some of us to solve future problems

ciao

marino