Error bptm (pid=PID_NUMBER) error requesting media, TpErrno = Robot operation failed

Problem

Backups to IBM tape drives fail with:

01/17/2012 00:55:44 - Error bptm (pid=PID_NUMBER) error requesting media, TpErrno = Robot operation failed

01/17/2012 00:55:44 - current media MEDIA_ID complete, requesting next media Drive002:NetBackup:MEDIA_ID_2

01/17/2012 00:55:49 - awaiting resource Drive002:NetBackup:MEDIA_ID_2. 

          A pending request has been generated for this resource request. 

          Operator action may be required. Pending Action: Tape is unmountable, 

          Media ID: MEDIA_ID_2, Barcode: MEDIA_ID_2_BARCODE, Density: hcart2, Access Mode: Read, 

          Action Drive Name: N/A, Action Media Server: N/A, Robot Type(Number): 8(0), 

          Volume Group: 000_00000_TLD, Action Acs: N/A, Action Lsm: N/A

Error Message

 01/17/2012 00:55:44 - Error bptm (pid=PID_NUMBER) error requesting media, TpErrno = Robot operation failed

01/17/2012 00:55:44 - current media MEDIA_ID complete, requesting next media Drive002:NetBackup:MEDIA_ID_2

01/17/2012 00:55:49 - awaiting resource Drive002:NetBackup:MEDIA_ID_2. 

          A pending request has been generated for this resource request. 

          Operator action may be required. Pending Action: Tape is unmountable, 

          Media ID: MEDIA_ID_2, Barcode: MEDIA_ID_2_BARCODE, Density: hcart2, Access Mode: Read, 

          Action Drive Name: N/A, Action Media Server: N/A, Robot Type(Number): 8(0), 

          Volume Group: 000_00000_TLD, Action Acs: N/A, Action Lsm: N/A

Solution

- Typical symptoms of this issue is that the drives don't get detected in a later stage of the device config wizard and are often denoted as "unusable".

 - Run robtest and see if the drives get mounted/unmounted properly.

e.g.,

m s1 d1

Initiating MOVE_MEDIUM from address 1025 to 257

MOVE_MEDIUM complete

m d1 s1

Initiating MOVE_MEDIUM from address 257 to 1025

move_medium failed

sense key = 0x5, asc = 0x2c, ascq = 0x0, COMMAND SEQUENCE ERROR

- Check for the installed device drives on Linux using command:

[root@device_host ]# lsmod | grep lin

lin_tape              233180  2

nfnetlink              10713  1 ip_conntrack

scsi_mod              134605  7 lin_tape,st,sg,libata,ips,qla2xxx,sd_mod

- In above case, we see the IBM specific drivers "lin_tape" for tape drives along with the Veritas Provided sg and st drivers.

- Ask the customer to remove these lin_tape drivers (let them/their Sys Admin do it).

- Then delete all the drives (if any configured and malfunctioning) and reboot the server.

- Run /usr/openv/volmgr/bin/scan to see all the drives and robot seen properly and re-run device config wizard to configure the drives properly this time.

Applies To

 NetBackup Enterprise Server 6.5.4 on RHEL (RedHat Enterprise Linux) 5.5 (5.x, for that matter)