Ans1312e - Server Media mount not possible - Problem solving approach

[adsm04]

Hi we are getting ANS1312E Server media mount not possible errors during incremental backups.

First we thought its due to lack of scratch tapes and loaded enough, still getting the same error.

Now is boils down to lack of tape drives since all the tape drives are busy writing for other clients.

In this kind of a scenario:

1)Can we lock few tape drives for a particular client during the backups?

or

2)Can we create a disk pool so that if no tape or tape drives, backups should be done on disk instead of tape pool and later on when the tape/tape drives are avbl the data should be moved back to tapes from the disk automatically for offsite storage purpose.

How can we acheive this? Thanks in advance!!

 

[maciak]

A lot of questions

Let's get them one by one



1. Are you getting the error ALWAYS or just sometimes?

If it's always, you might have a problem with maxscratch parameter. Each stgpool has a maxscratch parameter which says how many scratch volumes can this stgpool acquire from the scratch pool. If you reached the number, no more scratch tapes are allowed for this stgpool.

Try to check it by issuing this command:

select stgpool_name, maxscratch, numscratchused from stgpools where devclass<>'DISK'

If that's the case, you should increase the maxscratch parameter:

update stgpool <stgpool_name> maxscratch=<some_bigger_number>



2. As far as I know, it's NOT possible to reserve some drives just for one or more nodes. (Someone correct me please if I'm wrong)



3. Of course it's possible to use a diskpool for this purpose. In fact it's usually implemented with DISK stgpool being the first and TAPE stgpool it's next stgpool where the data go through the MIGRATION process (automatic or manual).

It's quite complex topic, I recommend to consult the Administrator's guide first and then ask what's not clear.

 

[adsm04]

Hello Maciac,

Thanks for your suggestions. I checked for the things that you had asked for....

1)This problem occurs most of the time but not always.

2)MaxScratch allowed for this storage pool=1000, which i think is a pretty good number. On a daily basis this storage pool wont backup more than 1TB of data.



In the meanwhile i'm also checking to create a diskpool to see if this problem is resolved.

 

[icewalker]

ANS1312E and Virtual Tape Libraries

I've recently started having these errors pop up occasionally with a few clients because we have moved over to a Virtual Tape Library.

I have enough virtual tape drives to handle the number of clients that can connect (or I thought I did). Our MAXSESSIONS is 25 so I have 12 tape drives. Since there are 2 sessions for each backup, this seemed like it would work the majority of the time (I know, the math is not exact).

Anyway, I was wondering if there is an option in the client to say "just relax, wait about 5 minutes, and try again" instead of the client spamming the log with ANS1312E messages and just plain quitting with a FAILED 12.

 

[rore]

Icewalker,
Increase the number of virtual drives. If you don't have a diskpool (probably not because you are using VTL) 12 drives seems to be not enough. This is subject to your environment and backup load which is not described in the post.

Maciak,
I think you can resolve your problem configuring a disk pool. Tipically, incremental backups doesn't provide enough sustained data transfer to feed a tape drive. (I can be wrong, it depends on your environment not described in the post). Send to the disk pool the slowest incremental clients. Send directly to tape the big chunks of data like big databases, image backup. Then configure a daily migration job.

Rudy