This can be seen if the OS no longer has a reliable stable connection to the tape drive or if the tape drive and the connection to it is unstable or has problems.
To check further, disable the device within networker then try to run the following command:
mt -f 'device' status
E.g.:- mt -f .\Tape0 .\Tape0> status
By disabling the device in NetWorker, we know that we are working with only the OS.
If it errors or shows an incorrect output then the OS, the tape drive or the connection needs to be investigated further. Rebooting the affected systems may clear the error for some time and may be used to further demonstrate what component is unstable.
Check to ensure the OS is seeing the device correctly – that is, does it see it as an Ultrium LTO (for example), or just a generic or incorrect drive?
If there is an error after the command is issued, and the drive was working fine previously, then the first question I would have be is “What changed?” It could be an OS patch, a NetWorker update, a switch update. It might even be something as simple as a cable being closed in a door and damaging it.
Remember to re-enable the device after the issue is resolved.
If your mt command is hanging (mt -f \\.\Tape0) then it means it hangs on I/O which means something wrong with SAN or SCSI usually. This sort of problem can't be resolved with application. You may wish to reboot the server to check if issue is gone. if not, have someone to check switch logs and library logs Eventually, you may wish to shutdown server, reboot library and then start server (if issue would port init from library side).
What bug is that? Picture shows that system does not give you tape path as, along with earlier message, driver is missing. You can check that from Windows side too.
Ok, so path disappears after cleaning (or request for cleaning). Since this path is not something given by NW, but rather by system utilizing drivers (system or vendor ones) - you should really look elsewhere. LTO usually likes to claim no cleaning is needed, but cleaning cartdriges exist nevertheless. Did you do the cleaning? If yes, and if drives requires cleaning again (from HW perspective) then you have issue with drive and should be replaced.
How is your environment set to do cleaning via NetWorker or the hardware?
If a cleaning tape is loaded into the drive and is busy trying to clean the tape device, running inquire will not report the tape name as the scsi probe command sent to the device will be ignored as it is busy, so you may not see the name etc reported.
Inquire should only be run when all drives are free and available, otherwise you can actually rewind the tape volume and loose data.
If the hardware is making repeated attempts to clean the drive then I would contact the HW vendor to get the tape drive checked/replaced. Either way this really points to OS level access not NW being the issue.
Mark_Bellows
240 Posts
0
October 25th, 2011 12:00
This can be seen if the OS no longer has a reliable stable connection to the tape drive or if the tape drive and the connection to it is unstable or has problems.
To check further, disable the device within networker then try to run the following command:
mt -f 'device' status
E.g.:- mt -f
.\Tape0
.\Tape0> status
By disabling the device in NetWorker, we know that we are working with only the OS.
If it errors or shows an incorrect output then the OS, the tape drive or the connection needs to be investigated further. Rebooting the affected systems may clear the error for some time and may be used to further demonstrate what component is unstable.
Check to ensure the OS is seeing the device correctly – that is, does it see it as an Ultrium LTO (for example), or just a generic or incorrect drive?
If there is an error after the command is issued, and the drive was working fine previously, then the first question I would have be is “What changed?” It could be an OS patch, a NetWorker update, a switch update. It might even be something as simple as a cable being closed in a door and damaging it.
Remember to re-enable the device after the issue is resolved.
Mark Bellows
batucito
7 Posts
0
October 26th, 2011 08:00
The process is 90% and does not work, go through all the tapes
Mark_Bellows
240 Posts
0
October 26th, 2011 11:00
What is the 10% you feel is missing and why do you say it does not work?
I would like to know so I can be sure the information is correct.
Mark
ble1
4 Operator
•
14.4K Posts
0
October 26th, 2011 12:00
If your mt command is hanging (mt -f \\.\Tape0) then it means it hangs on I/O which means something wrong with SAN or SCSI usually. This sort of problem can't be resolved with application. You may wish to reboot the server to check if issue is gone. if not, have someone to check switch logs and library logs Eventually, you may wish to shutdown server, reboot library and then start server (if issue would port init from library side).
batucito
7 Posts
0
October 27th, 2011 12:00
mount a tape reaches up to 90%, when I try to format remains 90%. Any process that tape is running at 90% and throws the error above
batucito
7 Posts
0
October 28th, 2011 08:00
ble1
4 Operator
•
14.4K Posts
0
October 31st, 2011 06:00
The error on picture seems to suggest that driver for tape drive is missing on the box.
batucito
7 Posts
0
November 2nd, 2011 04:00
ble1
4 Operator
•
14.4K Posts
0
November 2nd, 2011 05:00
What bug is that? Picture shows that system does not give you tape path as, along with earlier message, driver is missing. You can check that from Windows side too.
batucito
7 Posts
0
November 2nd, 2011 14:00
batucito
7 Posts
0
November 2nd, 2011 14:00
The problem is when the drive requests head cleaning
ble1
4 Operator
•
14.4K Posts
0
November 3rd, 2011 02:00
Ok, so path disappears after cleaning (or request for cleaning). Since this path is not something given by NW, but rather by system utilizing drivers (system or vendor ones) - you should really look elsewhere. LTO usually likes to claim no cleaning is needed, but cleaning cartdriges exist nevertheless. Did you do the cleaning? If yes, and if drives requires cleaning again (from HW perspective) then you have issue with drive and should be replaced.
masonb
445 Posts
0
November 3rd, 2011 03:00
Batucito,
How is your environment set to do cleaning via NetWorker or the hardware?
If a cleaning tape is loaded into the drive and is busy trying to clean the tape device, running inquire will not report the tape name as the scsi probe command sent to the device will be ignored as it is busy, so you may not see the name etc reported.
Inquire should only be run when all drives are free and available, otherwise you can actually rewind the tape volume and loose data.
If the hardware is making repeated attempts to clean the drive then I would contact the HW vendor to get the tape drive checked/replaced. Either way this really points to OS level access not NW being the issue.
Regards,
Bill Mason