Help for another issue

Question

Hi,

last week One of drive my LTO3 drives (/dev/rmt/4cbn) in the tape library went to service mode mentioning cleaning required. Afterwards, I did the complete cleaning and manually cleared the error message, and library rebooted.

But now the drive is unable to work anything, whenever it tries to mount or read WITH the following log:

nsrd: /dev/rmt/4cbn Verify label operation in progress
nsrmmd #3: tape_bsf bsf failed: drive status is Media Error, Recorded Entity Not Found

then, it is going to service mode every time, and not able to do any operation.

when i look on the drive, the front panel doesn't give any error. Anything can be done on EMC networker software side ?

some inputs would be helpful.

Thanks

HabibG2 · Answer

Check if you drive SCSI id is the same as before, this could have changed when you rebooted the Library. If so you may have to reconfigure the drive/Library

Have you check the status and the log on the drive via your browser?

ble1 · Answer

when i look on the drive, the front panel doesn't
give any error. Anything can be done on EMC
networker software side ?

No. You can load the tape and play with mt command (you can test bsf) and error should be there.

Hari5 · Answer

You would need to check if the scsi id to device name mapping for the drive is the same as before rebooting the library. You seem to be on Solaris. With Solaris, you might need to reboot the host if the tape library has been rebooted. Have had problems with this earlier.

vijayscsa · Answer

Networker also, giving warning like SCSI id wrong

Jan 27 20:24:39 scsi: [ID 107833 kern.notice] Sense Key: Media Error
Jan 27 20:24:39 scsi: [ID 107833 kern.notice] ASC: 0x14 (recorded entity not found), ASCQ: 0x0, FRU: 0x0
Jan 27 21:15:35 scsi: [ID 107833 kern.warning] WARNING: /pci@1d,700000/scsi@2/st@6,0 (st34):
Jan 27 21:15:35 scsi: [ID 107833 kern.notice] Requested Block: 0 Error Block: 0
Jan 27 21:15:35 scsi: [ID 107833 kern.notice] Vendor: HP Serial Number:

Should reboot of library and Reconfigure library from GUI will help to fix this problem

vijayscsa · Answer

Hi all,Thanks for the answer. now the error starts with the other drive alsoThu 13:46:35 /dev/rmt/5cbn Verify label operation in progress Thu 13:46:54 media warning: /dev/rmt/5cbn moving: tape_bsf bsf failed: drive status is END-OF-DATA NOT FOUND Thu 13:46:54 media warning: /dev/rmt/5cbn reading: no tape label foundI think, rebooting of the solaris networker server is little risky and it is good to hardware reset the library from networker commandline, to have all the scsi ids again checked.Is that okbest regards

HabibG2 · Answer

I don't think so. If as I said earlier your drive SCSI ID has changed by rebooting the library then the only way I know is to reconfigure the drive/library ( delete the drive and add it again). SCSI ID can change buy either rebooting the library or the server itself. This can be avoided by implementing ' Persistent Binding'.

HabibG2 · Answer

As I said, you have to delete the drive from the library and reconfigure it again. This is the only way I know. Delete driveSearch for driveConfigure new foound driveAdd drive to the library.

vijayscsa · Answer

Hi,As you said, the problem couldn't resolve after solution provided by EMC, cleaning /nsr/tmp after stop of nsr service and starting the service and then running nsrjb -HEv command.Again the suspected drive is not able to do any operation:It started with the following:operation source: nsrd jb op; progress: Operation `Verify label' in progress on device `/dev/rmt/4cbn';Verifying label, moving backward 2 filesJan 29 17:00:53 srvblrnwr01 root: [ID 702911 daemon.notice] NetWorker media: (info) write open error: drive status is DRIVE_STATUS_WRITE_PROTECTand then finally,/dev/rmt/4cbn moving:tape_bsf bsf failed: drive status is Media Error, Recorded Entity Not foundWhat solution i could try to fix this drive issue, because i knew this may not be the hardware issue.Need your valuable helpThanks

ble1 · Answer

First of all, you do NOT get error by NW, but rather by system and it is your library that is sending you that signal via OS. ASC code is exactly what is says and all other parties involved are just parsing this (including nsrmmd).

Question here is if this is happening always with same tape volume or it doesn't matter. If it doesn't matter you should check drive (you could test others too). If all drives are giving you the same, then library needs closed look.

Hari5 · Answer

Did you try to recreate your library and tape drive resources on NW? That should fix your problem.

NetWorker

Help for another issue

Was this post helpful?