Restart the disk staging process, because of a faulty tape

[faridux]

Hello,

I have a a disk staging storage who works perfectly, the final destination is a LTO2 tape,

A tape was damaged, but the original backup is still on disk,

How can I send these images on a new tape again,

I was thinking to expire the damaged tape and restart manually the relocation to final destination,

it is maybe wrong or is there any other way to duplicate the primary copy

Note It concerns my oldest master (version 5.0) that's why I do not bother the support ...

Thank you

[Nicolai · Accepted solution]

But I have a manual suggestion:

If you know the image id's, list them in a text file or use this command to list all images on the disk stu

bpimmedia -l -mediaid {DISK_STU_PATH} -d 01/01/2010 | grep IMAGE |awk '{ print $4 }' > /tmp/to_duplicate

Do not add a / to the path for the diskstu. Then do a:

bpduplicate -dstunit {storageunit} -Bidfile /tmp/to_duplicate  -primary -dp  {DESTINATION_VOLUME_POOL} -L /tmp/progress.log

bpduplicate will then start duplicating the images listed in "to_duplicate".  The duplication process will delete the /tmp/to_duplicate files when finished. The duplicated images will be the primary version (-primary qualifier to bpduplicate) and you can now safe expire the disk images (optional of cause).

e.g.

bpexpdate -backupid {backupid} -d 0 -copy 1

Before starting to duplicating the images, pls check Netbackup is configured to allow more than two copies. Check Host properties - master server - global attributes - Maximum backup copies.

[alazanowski]

Use the GUI - go into the catalog and search for Copy #2 of that backup (assuming that the destroyed tape was set as the primary after duplication) and then do a duplicate job using the option from right-click. If duplicate isn't available make sure you are set to duplicate on the catalog view and not verify. command line always works, but guis make it nice to find and perform what you're looking for. Set the new duplication to primary and expire the broken tape copy.

[marekkedzierski]

Outbacker - it should be copy #1

[faridux]

Thank you for your help, with your advices I was able to write a script to replace the broken tape
I let the primary copy on disk as requested by my customer
The tape broken hosts the copy number2 of 66 backupids,

1. With bpimmedia command I have listed the backupid present on the bad tape
2. for each backupid I have expired the copy on the bad tape and I have duplicated the primary copy on a new tape

I provide an example for one backupid I did manually,maybe it could help someone else

A. I check the primary copy is on disk (at least not in the broken tape)
server1:/tmp# bpverify -backupid server2_1281456137 -p -primary
 Path = "/staging_path"  Server = server1
 Bid = server2_1281456137  Kbytes = 25952  Fragment = 1

B. I check the copy number on the broken tape (always the number 2 for my part)
server1:/tmp# bpverify -backupid server2_1281456137 -p -cn 2
 Media id = S02146  Server = server1
 Bid = server2_1281456137  Kbytes = 25952  Filenum = 64  Fragment = 1

C. I expire the copy number on the broken tape
server1:/tmp# bpexpdate -d 0 -backupid server2_1281456137 -copy 2 -force
server1:/tmp# bpverify -backupid server2_1281456137 -p -cn 2
INF - Status = found no images or media matching the selection criteria.

D. Then I duplicate the primary copy, the duplicated image becomes the number 2
server1:/tmp# bpduplicate -dstunit server1-hcart2-robot-tld-0 -primary -backupid server2_1281456137 -dp archiMR_copy
Duplicate started Wed Sep 22 2010 15:43:29
Activity monitor job id = 308177
INF - Destination storage unit server1-hcart2-robot-tld-0 on host server1
INF - Duplicating policy staging_path schedule staging_path backup id server2_1281456137 copy 1 created on 08/10/2010 18:02:17 on source path /staging_path
INF - Duplicate of backupid server2_1281456137 successful.
INF - Status = successfully duplicated 1 of 1 images.

E. I check the copy number2
server1:/tmp# bpverify -backupid server2_1281456137 -p -cn 2
 Media id = S02199  Server = server1
 Bid = server2_1281456137  Kbytes = 25952  Filenum = 4  Fragment = 1

F. Hereunder the log of the job
server1:/tmp# bperror -jobid 308177 -U
    TIME            SERVER/CLIENT                      TEXT
09/22/2010 15:43:30 server1 server2  begin reading backup id server2_1281456137
                    (duplicate), copy 1, fragment 1
09/22/2010 15:44:28 server1 server2  begin writing backup id server2_1281456137,
                    copy 2, fragment 1, to media id S02199 on drive index 4
09/22/2010 15:44:28 server1 server2  successfully read (duplicate) backup id
                    server2_1281456137, copy 1, fragment 1, 25952 Kbytes at
                    448.461 Kbytes/sec
09/22/2010 15:44:33 server1 server2  successfully wrote backup id
                    server2_1281456137, copy 2, fragment 1, 25952 Kbytes at
                    43038.143 Kbytes/sec

Thanks again

[Will_Restore]

thanks for the follow-up

[Nicolai]

Thumps up from me. Thanks for marking my post as a solution.