Verify all images by volume group or volume pool

[Seth_E]

Hi all,

This may be far fetched, but, is there a way that netbackup can validate all primary copies (or others) within a given period. So, lets say I backup 1 TB to disk and duplicate to tape. I'd like it so that Netbackup can check to ensure that all images on the tapes are good before I send them offsite.

I've recently run into an issue with this error on a few tapes:

bptm read error on media id A01157, drive index 4, reading header block, More data is available.

I than recalled the offsite copy and checked it into the library. I got the same error on the offsite tape as well. Didn't matter what drive read the tape.

I've seen this on 6 different tapes so far and found that they all belong to a specific policy. This was also around the time I got "Storage Full" errors on our DSU (recently increased the size to compensate).

Anyway, I'll open a support call about these tapes and was wondering if any of you out there regularily validate images and if so, how do you go about doing it.

Thanks in advance.

[watsons · Accepted solution]

Image verify for that particular copy. 

bpverify can do for either image (copy), on disk or pool:

bpverify: [-l] [-p] [-pb] [-v] [-local] [-client <name>]
    [-st <sched_type>] [-sl <sched_label>] [-L <output_file> [-en]]
    [-policy <name>] [-s mm/dd/yyyy HH:MM:SS] [-e mm/dd/yyyy HH:MM:SS]
    [-M master_server] [-dml]
    [-Bidfile <file_name>]
    [-pt <policy_type>] [-hoursago <hours>]
    [[-cn <copy_number>] | [-primary]]
    [-backupid <backup_id>] [[-id <media_id or path>] |
    [-stype server_type]
    [-dp disk_pool_name [-dv disk_volume]]] [-priority <number>]

If needed, run bpmedialist -mcontents to read into the tape content, this can take some time but will ensure tape can be mounted and read.

[Seth_E]

Thanks. This is a good start. I ran the command

It appears that bpverify doesn't work with volume pools. I have volumes that remain in the library (onsite) and I'd like to run bpverify against them. I see that this command works with disk (which I'll use to check as well)

Am I incorrect?

[Seth_E]

I found this post to be helpful:

https://www-secure.symantec.com/connect/forums/how-can-i-check-if-my-images-storage-unit-are-corrupt...

Seems the -hoursago works well. Thanks for your help.

[mph999]

bptm read error on media id A01157, drive index 4, reading header block, More data is available.

... looks like the tape is either mis-positioned or overwritten.

When NBU reads a header on a tape(media header or backup header) which is 1k in size it allocates a buffer to read the header into.  I cannot remember the size of the buffer, it's either 32k or 64k, but this does not matter, all we need to know is the header is 1k, and the buffer is >1k.

The message is suggesting that whatever we have read didn't fit into the buffer, so whatever was read must be > 1k, and the only thing >1k on a NBU tape are the data blocks - so we have tried to read a header, but read data.