cancel
Showing results for 
Search instead for 
Did you mean: 

drive up but not taking backup.

mathur_gm
Level 3

Dear all,

We have veritas netbackup [6.5 V] setup with 2 sun Solaris master server & other media server for solaris/HP-UX/windows client. Many of our servers have huge database so nearly 20 servers are play the role of SAN client also. Many of the time when we see the output of vmorpcmd  –d ds. It shows that some of the drives are up but not taking any media for backup. We have try to make drive downà up.
Cleaning  the driveà crawlreleasebyname command for releasing the drive.
But these drive could take the backup. Whenever we restart the NBU services those started working fine but after some time show the same problem. These problem is not happening for specific drive it is happening with any of the drive.
Note:- I think this happen because any SAN client & media server where drive is shared, is not releasing  the drive. There is command which release the drives without restarting NBU services or any other solution.
Below is status of vmoprcmd -d ds command output showing drive 15 in same state:-

root@ggnbkp24 # ./vmoprcmd -d ds

                                  DRIVE STATUS

Drv Type   Control  User      Label  RecMID  ExtMID  Ready   Wr.Enbl.  ReqId
  0 hcart    TLD               Yes   DW0004  DW0004   Yes     Yes        0
  1 hcart    TLD               Yes   DW0012  DW0012   Yes     Yes        0
  2 hcart    TLD               Yes   DW1733  DW1733   Yes     Yes        0
  3 hcart    TLD               Yes   DW1914  DW1914   Yes     Yes        0
  4 hcart    TLD               Yes   DW1664  DW1664   Yes     Yes        0
  5 hcart    TLD               Yes   DW1498  DW1498   Yes     Yes        0
  6 hcart    TLD               Yes   DW0368  DW0368   Yes     Yes        0
  7 hcart    TLD               Yes   DW0324  DW0324   Yes     Yes        0
  8 hcart    TLD               Yes   DW1891  DW1891   Yes     Yes        0
  9 hcart    TLD               Yes   DW1882  DW1882   Yes     Yes        0
 10 hcart    TLD               Yes   DW0224  DW0224   Yes     Yes        0
 11 hcart    TLD               Yes   DW1427  DW1427   Yes     Yes        0
 12 hcart    TLD               Yes   DW1493  DW1493   Yes     Yes        0
 13 hcart    TLD               Yes   DW0453  DW0453   Yes     Yes        0
 14 hcart    TLD               Yes   DW1608  DW1608   Yes     Yes        0
 15 hcart    TLD                -                     No       -         0
 16 hcart    TLD               Yes   DW1509  DW1509   Yes     Yes        0
 17 hcart    TLD               Yes   DW1839  DW1839   Yes     Yes        0
 18 hcart    TLD               Yes   DW0121  DW0121   Yes     Yes        0
 19 hcart    TLD               Yes   DW1535  DW1535   Yes     Yes        0
 20 hcart    TLD               Yes   DW1128  DW1128   Yes     Yes        0
 21 hcart    TLD               Yes   DW0591  DW0591   Yes     Yes        0
 22 hcart    TLD               Yes   DW0464  DW0464   Yes     Yes        0
 23 hcart    TLD               Yes   DW1313  DW1313   Yes     Yes        0

thanks in advance

with regards,
Gaurav_monu
 

1 ACCEPTED SOLUTION

Accepted Solutions

Anonymous
Not applicable
Run robtest command and query the drives in your library(s) and confirm what vmoprcmd is seeing first.

Then maybe have a look at how the drive config for SCSI Reserve is placed.

tpconfig -dl will show you the setting

If no backup is running then my troubleshoot for a drive with tape in drive is Reset Drive.
Last resort is 'vmoprcmd -crawlreleasebyname ...' command

View solution in original post

13 REPLIES 13

Anonymous
Not applicable
Run robtest command and query the drives in your library(s) and confirm what vmoprcmd is seeing first.

Then maybe have a look at how the drive config for SCSI Reserve is placed.

tpconfig -dl will show you the setting

If no backup is running then my troubleshoot for a drive with tape in drive is Reset Drive.
Last resort is 'vmoprcmd -crawlreleasebyname ...' command

Stumpr2
Level 6

Why do all the drives show mounted tapes?
Are the tapes being dismounted and returned to their proper slot when backups finish?
Doesn't look like it.

mathur_gm
Level 3
hI SIR,

SEE RECENT vmoprcmd -d ds output.many of the drive are  up & these don't have any media              [   shown in robtest & physical checking of  drive ]. 
see below drive are up but not taking any media for backup:-5,7,11,13,15,21
I have already  run crawlreleaseby name comamnd for theese drivesbut problem is same.

root@ggnbkp24 # bash
root@ggnbkp24 # vmoprcmd -d ds

                                  DRIVE STATUS

Drv Type   Control  User      Label  RecMID  ExtMID  Ready   Wr.Enbl.  ReqId
  0 hcart    TLD               Yes   DW0019  DW0019   Yes     Yes        0
  1 hcart    TLD               Yes   DW1042  DW1042   Yes     Yes        0
  2 hcart    TLD               Yes   DW1084  DW1084   Yes     Yes        0
  3 hcart    TLD               Yes   DW1466  DW1466   Yes     Yes        0
  4 hcart    TLD               Yes   DW1042  DW1042   Yes     Yes        0
  5 hcart    TLD                -                     No       -         0
  6 hcart    TLD               Yes   DW1695  DW1695   Yes     Yes        0
  7 hcart    TLD                -                     No       -         0
  8 hcart    TLD               Yes   DW1781  DW1781   Yes     Yes        0
  9 hcart    TLD               Yes   DW0014  DW0014   Yes     Yes        0
 10 hcart    TLD               Yes   DW0121  DW0121   Yes     Yes        0
 11 hcart    TLD                -                     No       -         0
 12 hcart    TLD               Yes   DW1729  DW1729   Yes     Yes        0
 13 hcart    TLD                -                     No       -         0
 14 hcart    TLD               Yes   DW1329  DW1329   Yes     Yes        0
 15 hcart    TLD                -                     No       -         0
 16 hcart    TLD               Yes   DW0475  DW0475   Yes     Yes        0
 17 hcart    TLD                -                     No       -         0
 18 hcart    TLD               Yes   DW1691  DW1691   Yes     Yes        0
 19 hcart    TLD               Yes   DW1928  DW1928   Yes     Yes        0
 20 hcart    TLD               Yes   DW0010  DW0010   Yes     Yes        0
 21 hcart    TLD                -                     No       -         0
 22 hcart    TLD                -                     No       -         0
 23 hcart    TLD               Yes   DW0742  DW0742   Yes     Yes        0
root@ggnbkp24 #


with regd

gaurav

Omar_Villa
Level 6
Employee
Have you try using reseting the drives with vmoprcmd -reset command? Im sure this is a kind of bug between the drive assignment from the Media Server to the FT Media Server, remeber that always the FT Medias will give priority to the SAN clients and if there is no Drive available for the SAN Client the backup will go through the IP Network, but still that client haves priority. Try disableling the SAN Clients and see how that runs, this is only to isolate the posibility that the SAN Client is related with the issue.


Please let us know.
best regards.

Omar_Villa
Level 6
Employee

Also remember that for SAN Client or FT Medias there is a special service under the EMM DB who controls all the FC Transports, maybe checkin the VxUL logs on the master for Originator 199 and see for any hanged process that is holding your drives to be release.

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified
Hi Gaurav,
It is possible that you have 'orphaned' resource allocation for these drives. Compare vmdareq  & vmoprcmd -d  output with nbrbutil -dump output. Save nbrbutil output to a file and look for MDS allocations. If any of the drives that are not in use have an MDS allocation, release the allocation with 'nbrbutil -releaseMDS <allocation#>'.
Also confirm Storage Unit configuration - how many drives per STU? What is MPX level for each drive?
What does Job Details tab display wrt reason for queued jobs?

Mike_Gavrilov
Moderator
Moderator
Partner    VIP    Accredited Certified
If you realy shure that this is reservation issue check if you use SPC-3 (your library should support it):
http://seer.entsupport.symantec.com/docs/314787.htm
In this http://seer.entsupport.symantec.com/docs/images/316341/NBU_p127i041.png you should select SCSI persis. reserv.
But if you using SAN-client check if you using  6.5.4 version  (SAN client on it is more stable and will be better on 6.5.5)

Mike_Gavrilov
Moderator
Moderator
Partner    VIP    Accredited Certified
Good idea.
Before Gaurav collect set of logs and traces of the jobs we can just play "Who is a best clairvoyant" :)

Srikanth_Gubbal
Level 6
Certified
i agree with Marianne, 'nbrbutil -releaseMDS <allocation#>' should work or else  in a final go use nbrbutil -resetAll

Zahid_Haseeb
Moderator
Moderator
Partner    VIP    Accredited
resrart the Robot/drive
restart library
see this command also tpconfig -d and then try to take backup

pawan_nagra
Not applicable
Certified
Hi Gaurav,

Please restart the services of NBU on all master and media servers.
Thanks
Pawan

NavA
Level 4
Gaurav,
  On the drives that are not taking backups run the mt command and check the status. I have come across this problem quite a few times and the only thing that worked for me is to recreate the drives from scratch. [backup tpautoconf -t, scan -tape, sgscan tape outputs]

mt -f drive-path status
EX:
# mt -f /dev/rmt/20cbn status
/dev/rmt/20cbn: no tape loaded or drive offline

If you get write protected or reserved and also if vmoprcmd status goes into PEND-ACS then it is SCSI reservation conflict. [vmoprcmd -crawlrelease does help in some situations]

THe fix that worked for me is to rescan drives from scratch.

rm /dev/rmt/*
devfsadm -Cc tape -v [ or cfgadm to force configure hba that is connecting to tape drives].
then do sg.build & sg.install stuff.
tpautoconf -a

Try ot on media server first.


EdsonFernandez
Not applicable
 Hi,

Please, see this document:

http://seer.entsupport.symantec.com/docs/325265.htm

It helped me to solve a similar problem...


[]'s