SL3000 - New Drives Being Seen as Stand-Alone Drive When Trying to Configure in NBU

Asset ID:	1-72-1431624.1
Update Date:	2012-03-08
Keywords:

Solution Type Problem Resolution Sure

Solution 1431624.1 : SL3000 - New Drives Being Seen as Stand-Alone Drive When Trying to Configure in NBU

Applies to:

Sun StorageTek SL3000 Modular Library System - Version: Not Applicable and later [Release: N/A and later ]
Information in this document applies to any platform.

Symptoms

While trying to configure new drives to NBU (Net Backup), it lists them as stand-alone drives, outside the library.
File /usr/openv/volmgr/bin/robtest does not list the new drives in the library, and this program queries the library directly, not the host configuration files.

After checking every thing and nothing else left to do, I asked them to power cycle the library. That solved the issue! Nothing changed in the SLC regarding the drives. They had two sites and two SL3000 libraries, other library they added one drive and the other 3 drives. Same symptom on both sites. After rebooting the other site library as well, that fixed the issue there also. Drives were added on the fly, so this was first reboots after the drive installations.

Changes

- Customer added new tape drives to both libraries (two SL3000 on different sites).

- No partitions involved. New drives got every status in SLC OK. Online, HLI-PRC address, WWN
assigned, port login to fabric. Drives listed normally in all the reports.

- Hosts found the new drives fine (both NBU master and media servers).

Cause

New drives added to the SL3000 library.

Solution

Any drive or capacity change on a non-partitioned FC/SCSI library requires a reboot of the SL3000 to rebuild the element ID table.

After checking every thing and nothing else left to do, had a power cycle done to the library. That solved the issue! Nothing changed in the SLC regarding the drives. They had two sites and two SL3000 libraries, other library they added one drive and the other 3 drives. Same symptom on both sites. After rebooting the other site library as well, that fixed the issue there also. Drives were added on the fly, so this was first reboots after the drive installations.

The behavior experienced is expected. The only way the SCSI element IDs can be created is to boot the SL3000 after a physical change. This applies to added slots too. NBU maps the library slots, drives and CAPs with element IDs. This became a big issue at an account that added drives that were never used. After a routine reboot 6 months later, all the drives remapped and created havoc. The new drives were inserted between the DEM and Base making it worse! HLI does not have this reboot limitation. The new drives or capacity can be used by ACSLS/HSC as soon as the library is audited.

Provided by Richard Kent

Attachments

This solution has no attachment