Highlighted

Volume pools media list

Dear Friends,

We have quite a big infra to take the backups. And my jobs failed now and then with error 96, "no media in the pool". Kindly suggest any script or command, where i can get the free and full media list for each volume pool. We have almost 69 Volume pools.

Thanks in Advance.

netbackup 7.7, win 2008 R2 Ent.

3 Solutions

Accepted Solutions
Highlighted
Accepted Solution!

Re: Volume pools media list

You can run available_media script from cmd on the master server on a daily basis.
The command is in ...netbackup\bin\goodies.
Best to send output to a text file as output in your environment will probably be quite large, e.g:
.... \goodies > available_media > c:\temp\media.txt
(assuming there is a temp folder on c-drive...)

You may want to create a Scratch pool and mark the pool as type scratch.
Over the next couple of weeks, move tapes to Scratch pool as soon as they expire.
You only need to do this once for each tape.
NBU will take tapes from Scratch if no empty media exists in a pool.
Once these tapes that came from Scratch expire, they will automatically be resturned to Scratch.

Next thing to look at is the amount of Volume Pools - why does this environment need 69 pools?

If you understand that media servers do not share media by default (can be enabled), and that retention levels are by default not mixed (should NOT be enabled), it makes sense that media managent in your environment will be a complete nightmare...

 

View solution in original post

Highlighted
Accepted Solution!

Re: Volume pools media list

69 volume pools sound like very bad design. More volumes pool require more media, and more tape drives because netbackup does not mix backup going to different volume pools.  One example, lets assume each volume pool require one additional tape because each pool is 100% used, that will sum up to 69 free media being required. If you had one pool, it would only require one tape. From a tape drive efficiency perspective you will have a similar issue. Many volume poll = many tape drives required.

For one company, you should have no more than 1-3 volume pools. 

For a view of the media you have run avaiable_media script in [install_path]\netbackup\bin\goodies\available_media

 

View solution in original post

Highlighted
Accepted Solution!

Re: Volume pools media list


Saleem04 wrote:


.... but do they all go to separate places when the tape is full? Have you a business requirement that means each type of backup needs to be on separate media (in it's own volume pool)?

Essentially, you could probably run NetBackup straight 'out-of-the-box' with just the one default "NetBackup" volume pool. The application will then utilise the media to the best of its abilities - it's an enterprise product & it's what it's designed for after all.

If we look at the 'default' case of one volume pool "NetBackup", as the application does not mix retention periods on media by default, if you have 2 retention periods you will have a minimum number of TWO partially full media - one for one retention period & one for the other. If you have 4 retention periods, then obviously this increases to a minimum of FOUR partially full media. Now imagine you have 100 volume pools ..... you end up with A LOT of partially full media that NetBackup will not utilise again until such time as the media in question fulfil the requirements of the backup being run.

There is, potentially, the opportunity to be (much) more efficient in how your media is being utilised

View solution in original post

7 Replies
Highlighted
Accepted Solution!

Re: Volume pools media list

You can run available_media script from cmd on the master server on a daily basis.
The command is in ...netbackup\bin\goodies.
Best to send output to a text file as output in your environment will probably be quite large, e.g:
.... \goodies > available_media > c:\temp\media.txt
(assuming there is a temp folder on c-drive...)

You may want to create a Scratch pool and mark the pool as type scratch.
Over the next couple of weeks, move tapes to Scratch pool as soon as they expire.
You only need to do this once for each tape.
NBU will take tapes from Scratch if no empty media exists in a pool.
Once these tapes that came from Scratch expire, they will automatically be resturned to Scratch.

Next thing to look at is the amount of Volume Pools - why does this environment need 69 pools?

If you understand that media servers do not share media by default (can be enabled), and that retention levels are by default not mixed (should NOT be enabled), it makes sense that media managent in your environment will be a complete nightmare...

 

View solution in original post

Highlighted
Accepted Solution!

Re: Volume pools media list

69 volume pools sound like very bad design. More volumes pool require more media, and more tape drives because netbackup does not mix backup going to different volume pools.  One example, lets assume each volume pool require one additional tape because each pool is 100% used, that will sum up to 69 free media being required. If you had one pool, it would only require one tape. From a tape drive efficiency perspective you will have a similar issue. Many volume poll = many tape drives required.

For one company, you should have no more than 1-3 volume pools. 

For a view of the media you have run avaiable_media script in [install_path]\netbackup\bin\goodies\available_media

 

View solution in original post

Highlighted

Re: Volume pools media list

Hi

Highlighted
Accepted Solution!

Re: Volume pools media list


Saleem04 wrote:


.... but do they all go to separate places when the tape is full? Have you a business requirement that means each type of backup needs to be on separate media (in it's own volume pool)?

Essentially, you could probably run NetBackup straight 'out-of-the-box' with just the one default "NetBackup" volume pool. The application will then utilise the media to the best of its abilities - it's an enterprise product & it's what it's designed for after all.

If we look at the 'default' case of one volume pool "NetBackup", as the application does not mix retention periods on media by default, if you have 2 retention periods you will have a minimum number of TWO partially full media - one for one retention period & one for the other. If you have 4 retention periods, then obviously this increases to a minimum of FOUR partially full media. Now imagine you have 100 volume pools ..... you end up with A LOT of partially full media that NetBackup will not utilise again until such time as the media in question fulfil the requirements of the backup being run.

There is, potentially, the opportunity to be (much) more efficient in how your media is being utilised

View solution in original post

Highlighted

Re: Volume pools media list

I agree with what is already posted, but here are answers to your questions 

From your initial post it looks like you are on windows. So that is one strike against what you are trying to do...

You need to be able to run scripting, and I do not know if you have access to a "grep" equivalent.

I am on unix, I have a script that runs periodically that list out scratch pool values.

There was a bug in earlier versions, that allowed scratch pool to be reset, so I actually set my scratch pool every day, to prevent that.

To count volumes in pools 

1. inventory the robot using vmupdate command - you will need to update the robot type (-rt), robot number (-rn), master (-h) and possibly robot host (-rh) if different from master...  "vmupdate -rt acs -rn 3 -use_barcode_rules -h master -rh media"

2. determine scratch pool ( most people use "scratch_pool" )

3. run vmquery command against the scratch pool to see the robot type and number, so you can grep for it

I have 6 robots, some ACS and some TLD, so the output can be like

315611 HCART3 TLD 6 25611 - - 2 09/05/2016 04:07

E06017 HCART ACS 0 - - - 18 04/08/2016 08:51

4. I set a script that runs a command like this against each robot

vmquery -pn scratch_pool -b | grep "ACS 0" | wc -l

and output that to an email 

5. I list what is my scratch pool using this command:

vmpool -list_scratch

6. I reset it with this command:

vmpool -set_scratch scratch_pool

I email myself every morning  - here is an example output

Checking for Scratch tapes in ACS0!

  There are  694 LTO5 scratch tapes in the ACS robot scratch_pool!

  ACS0 used  47  tapes in the last 24 hours

 

 

  Checking for Scratch tapes in ACS3 @ 170DR!

  There are  73 scratch tapes in the ACS3 @ 170DR robot scratch_pool!

  ACS3 used  5  tapes in the last 24 hours

 

  Checking for Scratch tapes in DD1!

  There are  11330 scratch tapes in the DD1 robot scratch_pool!

  DD1 used  284  tapes in the last 24 hours

 

 

  Checking for Scratch tapes in DD2!

  There are  11272 scratch tapes in the DD2 robot scratch_pool!

  DD2 used  275  tapes in the last 24 hours

 

 

  Checking for Scratch tapes in DD3!

  There are  11162 scratch tapes in the DD3 robot scratch_pool!

  DD3 used  384  tapes in the last 24 hours

 

***************************************

Verify scratch_pool is defined as the scratch pool:

Scratch Pools

=============

scratch_pool

 

 ***************************************

 

Setting scratch pool whether it is needed or not!

Verify scratch_pool is defined as the scratch pool:

Scratch Pools

=============

scratch_pool

 

NetBackup 8.1.2 on Solaris 11, writing to DataDomain 9800 6.0.2.50
duplicating via SLP to LTO5 in SL8500 via ACSLS
Tags (2)
Highlighted

Re: Volume pools media list

"From your initial post it looks like you are on windows. So that is one strike against what you are trying to do..."

he he ....

 

Highlighted

Re: Volume pools media list

We do all our windows scripting in Powershell - so we basically have all the same capabilities of Unix shell scripting.  Plus I have downloaded the UNXUTILS for Windows - so if I need to do a grep or gawk or sed I can do so.