Video Screencast Help

NDMP restore is failing with block read is not a NetBackup or Backup Exec media header

Created: 01 Dec 2012 | 2 comments

We are trying to validate a recently configured EMC filler backups. We are restoring the data from full backup to the same filler. But all the restores are failing, from differnt tapes and different images. We tried to do image verifcation, that also showing errors. So far, we tried from three different tapes, no luck.

 

Its mounting the tapes, but not even reading

Job detail :-

 

11/30/2012 21:32:05 - begin Restore
11/30/2012 21:32:05 - number of images required: 1
11/30/2012 21:32:05 - media needed: HH0377
11/30/2012 21:32:09 - restoring from image hhccha-vnx-cs0_1353665318
11/30/2012 21:32:12 - Info bpbrm (pid=22507) hhccha-vnx-cs0 is the host to restore to
11/30/2012 21:32:12 - Info bpbrm (pid=22507) telling media manager to start restore on client
11/30/2012 21:32:12 - Info bpbrm (pid=22507) spawning a brm child process
11/30/2012 21:32:12 - Info bpbrm (pid=22507) child pid: 22512
11/30/2012 21:32:13 - Info bpbrm (pid=22512) start tar on client
11/30/2012 21:32:13 - requesting resource HH0377
11/30/2012 21:32:13 - Waiting for scan drive stop IBM.ULT3580-TD5.004, Media server: nbuchasun002
11/30/2012 21:32:14 - Info tar (pid=22523) Restore started.
11/30/2012 21:32:15 - granted resource  HH0377
11/30/2012 21:32:15 - granted resource  IBM.ULT3580-TD5.004
11/30/2012 21:32:17 - Info bptm (pid=22511) Waiting for mount of media id HH0377 (copy 1) on server nbuchasun002.
11/30/2012 21:32:17 - started process bptm (pid=22511)
11/30/2012 21:32:17 - mounting HH0377
11/30/2012 21:32:17 - Info bptm (pid=22511) INF - Waiting for mount of media id HH0377 on server nbuchasun002 for reading.
11/30/2012 21:32:52 - Error bptm (pid=22511) block read is not a NetBackup or Backup Exec media header, len = 0, media id HH0377, drive index 6, data is unknown
11/30/2012 21:32:57 - restored from image hhccha-vnx-cs0_1353665318; restore time: 0:00:48
11/30/2012 21:32:57 - end Restore; elapsed time 0:00:52
11/30/2012 21:32:58 - Error ndmpagent (pid=22523) NDMP restore failed from path UNKNOWN
11/30/2012 21:32:58 - Info bpbrm (pid=22507) got ERROR 172 from media manager
11/30/2012 21:32:58 - Info bpbrm (pid=22507) terminating bpbrm child 22512 jobid=276303
Failed to get status code information (2826)
 
 
 
Image Verification :-
 
11/30/2012 10:36:49 - mounting HH0166
11/30/2012 10:36:49 - Info bptm (pid=22234) INF - Waiting for mount of media id HH0166 on server nbuchasun002 for reading.
11/30/2012 10:37:15 - Error bpverify (pid=16375) from host nbuchasun002, no data in archive
11/30/2012 10:37:15 - Error bpverify (pid=16375) Verify of policy HHC_CHA_DMP_HHCVDM_FS3, schedule PRD_Weekly_Full (hhccha-vnx-cs0_1353060896) failed, tar had an unexpected error.
11/30/2012 10:37:15 - Error bpverify (pid=16375) Status = no images were successfully processed.
11/30/2012 10:37:16 - Error bptm (pid=22234) block read is not a NetBackup or Backup Exec media header, len = 65536, media id HH0166, drive index 1, data is unknown
11/30/2012 10:37:16 - Error bpbrm (pid=22229) from client nbuchasun002: no data in archive
11/30/2012 10:37:16 - Info tar (pid=22233) done. status: 6
11/30/2012 10:37:16 - Info bptm (pid=22234) EXITING with status 172 <----------
11/30/2012 10:37:16 - Info tar (pid=22233) done. status: 172: cannot read media header, may not be NetBackup media or is corrupted
11/30/2012 10:37:16 - end Verify; elapsed time 0:01:03
no images were successfully processed  (191)
 
 
We tried to read the contents of the media, that also is failing :- 
 
Media Contents :-
 
11/30/2012 06:57:49 - requesting resource HH0166
11/30/2012 06:57:50 - Info bptm (pid=16158) start
11/30/2012 06:57:50 - started process bptm (pid=16158)
11/30/2012 06:57:50 - Waiting for scan drive stop IBM.ULT3580-TD5.007, Media server: nbuchasun002
11/30/2012 06:57:50 - granted resource  HH0166
11/30/2012 06:57:50 - granted resource  IBM.ULT3580-TD5.007
11/30/2012 06:57:51 - Info bptm (pid=16158) Waiting for mount of media id HH0166 (copy 1) on server nbuchasun002.
11/30/2012 06:57:51 - started process bptm (pid=16158)
11/30/2012 06:57:51 - mounting HH0166
11/30/2012 06:57:51 - Info bptm (pid=16158) INF - Waiting for mount of media id HH0166 on server nbuchasun002 for reading.
11/30/2012 06:58:14 - mounted HH0166; mount time: 0:00:23
11/30/2012 06:58:14 - Error bptm (pid=16158) tapemark or blank tape encountered reading media header, drive index 5
11/30/2012 06:58:14 - Info bptm (pid=16158) EXITING with status 85 <----------
media read error  (85)
 
Symantec is saying bad medias, but how all the tapes are showing this ?? Anyone faces this issue ??
 
NBU running on Solaris 10 with version 7.1.0.3
 
 

 

Comments 2 CommentsJump to latest comment

sazz.'s picture

Let's keep the filler aside for a moment.
Are you able to restore everthing else without any problem? I mean the normal backups which were not of filer.

If only filer is failing then what kind of tapes you are using for NDMP backup? Is the policy type NDMP? Have you backed them up with the same environment? Can you backup the filer now without any issues?Tape and drive density matches? Is the physical barcode changed on the tapes?

 

I hope you checked
http://www.symantec.com/docs/TECH58631

This error comes when NBU is not able to read the tape header.When you loaded a volume for a backup or restore, the tape manager (bptm), did not find the expected tape header. This error can mean that a robotic device volume is not in the slot number that is in the Media and Device Management volume configuration. It also can mean that a read error (I/O error) occurred. If the tape has been out of the tape library since the backup was written, there is also the possibility that the tape has been used for some other purpose and has been overwritten with non-Netbackup data

Recommended action:
 
•If the volume is in a robot that supports bar codes, verify the robot contents by using a robot inventory option.
•If the volume was mounted on a nonrobotic drive, verify that the correct volume was mounted and assigned.
•Check the NetBackup Problems report. If it shows a fatal read error, attempt the operation again with another drive, if possible.
•If your configuration has multiple servers or HBAs with access to your tape services, make sure the SCSI Reserve or Release is configured correctly. (Most likely, the tape services configuration is an SSO configuration.) For more information, refer to the NetBackup Shared Storage Guide

Nabil polska's picture

could you please try to :

Change the Buffer size on the tape drive to match the Block Size

This issue will be fixed  in the next release of Backup Exec.