02-14-2018 12:32 AM
Hello,
NBU 8.0 8 HP-UX Media Servers sharing 6 LTO5 drives. Master on Windows 2008 R2.
Yesterday there were some unplanned HP-UX restarts when duplication between drives was active.
However since then there is a PEND-TLD state on these drives.
vmoprcmd -crawlreleasebyname <drive name> returns:
Host <host1> returned: SPC-2 RESERVED
Host <host2> returned: SPC-2 RESERVED
...
on all sharing Media Servers.
And Device Config wizard redefines these drives again in PEND-TLD state.
Drives which were not active during restarts are ok (state TLD).
How to get rid of PEND-TLD state? The cause is probably outside NetBackup?
Regards
Michal
Solved! Go to Solution.
02-15-2018 03:41 AM
You may find that is scsi SPC2 reservation is used, you also need to powercycle drives (not some soft reset from the libarry GUI, needs to be real power cycle)
02-14-2018 12:47 AM
Hi Michal,
I haven't had that issue before as my drives all attach to a single host, but have found a couple of links that may help:
https://vox.veritas.com/t5/NetBackup/SSO-drive-status-in-PEND-TLD/td-p/354299
https://www.veritas.com/support/en_US/article.000015193
They might point you in the right direction, but either way I think you're up for restarting NetBackup daemons on Master and all Media servers.
Hope this helps,
Steve
02-14-2018 02:09 AM
OK I have advised the customer to go according to https://www.veritas.com/support/en_US/article.000015193
He'll let me know
Rgds
M.
02-14-2018 03:49 AM
I agree - The cause is probably outside NetBackup.
It seems that there is still a SCSI Reservation at OS-level.
My background is mainly Solaris - When I was still doing onsite support, I would use mt commands to check drive status and see if I could rewind and eject tape in attempt to release the reservation.
My HP-UX skills are less than basic, but probably has similar options with mt command.
If memory serves me right, then
vmoprcmd reset <drive-index>
will attempt to do the same.
02-15-2018 03:41 AM
You may find that is scsi SPC2 reservation is used, you also need to powercycle drives (not some soft reset from the libarry GUI, needs to be real power cycle)
02-15-2018 09:02 AM
Hello,
yes finally the drive(s) restart has resolved the issue.
Thanks/regards
Michal