BAB 11.5 SP2 Disk Staging

[TheMisio]

Ladies and Gents,

I have just set up staging backup jobs. The job runs to a staging device and 30 minutes after finish it should copy the data to a tape.
The problem is that the data migration would run for a couple of days and then fail with the error “unable to find this media or blank media”
Sometimes it would happen after I change the tape, sometimes without any reason. The destination target is to a GROUP and to media *, which should backup to any tape in the drive. I have spoken to a CA guy yesterday. He told me to increase buffer size to 512 and disable the drive in Windows Device Manager.
But this morning, I have noticed that the migration job did not even attempt to start.
Does anyone have any idea why my data migration isn’t working? The CA guys know less about BrightStor than me and I don’t think I’ll get anywhere with them.

My set up: Windows 2003 Std SP1, BrightStor 11.5 SP2 build 4234, CERTANCE LTO3 + ADAPTEC 29160, SQL 2000 SP3.

Thanks in advance.

Regards,

Michael

 

[davidmichel]

Well the tech was right about disabling the device. In general tape drives and library should always be disabled from within the OS Device Manager unless running Fibre with the Storport version of the HBA driver.

Try doing a Quick Erase Plus on a tape as a test.

Also install the Tape Engine update.

http://supportconnectw.ca.com/public/storage/downloads/nt/115/basbr115-solpatch.asp

QO82963 ZIP 10.3 MB 10/06/06 NT -DEVICE SUPPORT UDPATE 7 A

Might be database related.
To test
do a cstop
close the Manager
copy the Database directory someplace else
open ARCserve Server Admin
start the Database Engine
Initialize all the databases

Now try it again. If it works correctly you got a DB problem, if it does not, just put the Database directory back.

 

[TheMisio]

davidmichael,

Thank you for your prompt answer.
I shall try doing just that and will let you know.

Thank you again.

Regards,

Michael

 

[TheMisio]

davidmichael,

OK. I have done as you suggested. It kind of seemed to have worked. And here’s what I mean by kind of:
The backup job seemed to have run just fine and the data seemed to have been migrated. However, there is no Parent > Child job available to check how and when the data has been migrated. I mean, when I use disk staging or multiplexing when the job is complete you can branch out the child jobs to see the multiplexing clients (in multiplexing) or backup job > migration job (disk staging). I don’t have that. Also in the job log, there is nothing about migration. However, there are a data migration details in the Activity Log.
Also, when I go to restore > by session and look at the tape where the data has been migrated to, the sessions are there, but when I try to look at the files, it asks me to merge the details to the database.
Any ideas what the heck is going on?
This is the only BAB 11.5 SP2 server I’ve got. My other three servers are still standard 11.5. I just don’t trust SP2. Some weird stuff happening.

Regards,

Michael

 

[TheMisio]

davidmichael,

The data migration fails as soon as I change the tape.
Even the CA guys don’t know what’s going on.
I’m starting to get a little bit annoyed with the Disk Staging.

Regards,

Michael

 

[Severe]

David,

There are a couple of catches to using disk staging as I have recently found.

There is a very poorly documented issue with Disk Staging in that it does not support Media Maximization, so appending media does not work like you may think. If you have multiple jobs running Disk Staging, each of those jobs will attempt to use a new piece of media when they migrate from your FSD to tape. In a single drive, non-library enviroment this could lead to a situation similar to yours regardless if the jobs are using the same media group.

In our case, we created a single job with numerous servers in it. This isn't as problematic as it sounds since we can view each child job seperately from the Job Window. Anyway, this job creates sessions accordingly on the FSD then migrates them all to the same media AND it will append to previously written tapes with the same GFS group.

I'm curious, is the job that is failing a GFS job? I haven't tested how disk staging appends with non-GFS jobs so I'm not sure what the implications are.

Hope some of this helps. It sounds like you're using a single tape drive per server so it may not all be applicable.

 

[TheMisio]

Hi David,

Thank you for your reply.
You message just ensured me in what I suspected. The disk staging in BrightStor is still too immature and unreliable to use it.
Best example is that I have opened an issue with CA at the same time as this thread. I have got two replies on the forum, and only one from CA.
They’ve got absolutely no idea how to fix it. They know there are some issues, but can’t tell me what you said. It doesn’t work under some circumstances.
I do not use GFS just standard weekly backups.
I’ll try you suggestion and include all the servers I need to stage in one backup job.

Thanks again.

Regards,

Michael

 

[AlexChao]

Is disk staging available in version 11.1 ?

 

[Severe]

Alex,

Disk staging in 11.1 one was much more user intensive. Basically you would create your FSD area and use the TapeCopy function to move date.