cancel
Showing results for 
Search instead for 
Did you mean: 

tpautoconf serial number mismatch

Shekhar_D
Level 4
Certified

In netbackup what are the steps need to follow the if tpautoconf o/p mismatch and how those are communicated in OS side like windows,Sun<Linux,AIX,HP

Contineously we are facing issue of mismatch the drives and drives goes down.Please suggest.

Sometimes intermmitent issue comming on media server 6.5/7.

THanks.

4 REPLIES 4

revarooo
Level 6
Employee

I normally just remove all the drives and re-add them using:

nbemmcmd -deletealldevices -machinename <mediaserver> -machinetype media

especially if you do not have too many drives, then just re-add them after.

 

 

Marianne
Level 6
Partner    VIP    Accredited Certified

When you are sharing tape drives, try to keep them on the same NBU version.

Device mapping mismatch will cause drives to be unable to unload and then DOWN'ed.

If you have VERBOSE entry in vm.conf on media servers, you will probably see errors in syslog/messages/Event Viewer such as:

TLD(0) drive ## (device ##) is being DOWNED, status: Unable to SCSI unload drive.
or
Incorrect media found in drive index #, expected AAAAAA, found BBBBBB, FREEZING AAAAAA
etc...
 
If media server OS allows, try to ensure that Persistent Binding is done between OS and HBA. This will ensure that tape drives keep the same OS device names when system is rebooted.

 

Will_Restore
Level 6

Use -report_disc, -replace_drive, and -replace_robot to reconfigure the

devices in the EMM database to reflect a serial number change that a configured

device replacement caused. After hardware replacement, the correction process

requires that at least one system is available through the operating system. You

may need to re-map, rediscover, and reboot the system.

 

 

Omar_Villa
Level 6
Employee

Unfortunally this happens all the time, mainly when you replace a drive the way we do it is either with the tpautoconf -replace_drive option that will update all the drive in a SSO environment or one by one with the tpconfig -update option, there is no way to alert on this or detected on time, something you can script is the output of the 3 EMM tables related with the drive configuration: configDB (tpconfig), autoconf (tpautoconf) and globDB (vmglob) the output of this 3 needs to match you can write a script to catch any missmatch at Seria Number, Drive Name or Path level.