Video Screencast Help
Symantec to Separate Into Two Focused, Industry-Leading Technology Companies. Learn more.

Find the drive and status of Media

Created: 24 Oct 2012 • Updated: 03 Jan 2013 | 9 comments
This issue has been solved. See solution.

Hi,

I was trying a restore from a Media. But my restore job is saying that "Required Media" is in use. I believe this media should be free. How can I check if this media is currently mounted? Below are the questions I have based on this scenario.

1. How to find if the media is in Drive or in Library? If it is in a drive, which drive has mounted the drive?

2. Is there a way to find which job or process is using the media?

Env Details:

Master :

NBU version: 7.1

OS Version: HP-UX 11.31

In this above scenario, am looking for the details of VTL media which is of DataDomain VTL 880

Thanks,

RAJ

Comments 9 CommentsJump to latest comment

Nicolai's picture

Look in the activity monitor under device management to see if the media is mounted. If it isn't stop Netbackup activity and issue command :

nbrbutil -releaseMedia MEDIA_ID_NUMBER

Still "in use" use the more powerfull 

nbrbutil -resetAll

http://www.symantec.com/docs/TECH46807

Assumption is the mother of all mess ups.

If this post answered your'e qustion -  Please mark as a soloution.

RonCaplinger's picture

Quickest way to confirm it is in use:

In the NetBackup GUI, under "Media and Device Management", click on "Device Monitor".  You may need to click on the refresh button in the toolbar near the top of that window, or "View" then "Refresh", to update it.  This *should* show everything currently running to your tape drives.  To find out which job is using that media ID, you can view the Activity Monitor for any jobs using the VTL storage unit(s) and determine if there is anything still using that media.  If you don't see the media ID you are looking for in the list, then you should try the next step:

Harder way:

That media ID is most likely allocated in Resource Broker, and may have gotten "stuck" in an allocation for a previous job.  To see what has ahold of it, run this command on your master server:

nbrbutil -dump > /tmp/nbrbutil.txt

Then grep /tmp/nbrbutil.txt for the media ID in question.  If it is found in any of the lines, you can use the nbrbutil command again to manually force release that allocation (search the forum for more info on the nbrbutil usage, there are a few really good ones!). 

Marianne's picture

Something else that will put media to 'In Use' state is duplication job.
If Vault or SLP is used for duplication, all of the media needed gets reserved by the Resource Broker at the start of the job. So, tape may not show up in a drive, but will be 'In Use' because of being reserved.
If memory serves me right, all Media Ids needed for duplication are listed in Job Details.

Supporting Storage Foundation and VCS on Unix and Windows as well as NetBackup on Unix and Windows
Handy NBU Links

raj08's picture

Thanks Nicolai, Ron and Marianne. Before I check Ron method (harder way), my restore job already started. I got no chance to test the nbrbutil -releaseMedia. But it was again a great learning from all your inputs.

@Marianne: 1. If in case a Media ID is "in Use" because of Vault or SLP, is it safer to release media using nbrbutil -releaseMedia <media_id>??

2. And I think that you meant, finding medias needed for duplication can be viewed through Activity Monitor. So there will be only one job duplication runs at a time right? Correct me if I am wrong.

3. And also, I found this weird thing today. I started a restore job which requires 5 medias. All the 5 medias are in onsite pool (vtl) and I could see them in Robot (using GUI). But the restore jobs doesn't proceed further. It just stayed for more then an hour and after it started (logs below). It neither said "Medias are in use" or "Drives are in use". Would you be able to explain this?

10/24/2012 7:53:33 AM - begin Restore
10/24/2012 7:53:33 AM - 1 images required
10/24/2012 7:53:33 AM - media DR0191 required
10/24/2012 7:53:33 AM - media DR0340 required
10/24/2012 7:53:33 AM - media DR0363 required
10/24/2012 7:53:33 AM - media DR0374 required
10/24/2012 7:53:33 AM - media DR0386 required
10/24/2012 7:53:33 AM - media DR0397 required
10/24/2012 7:53:33 AM - media DR0415 required
10/24/2012 9:12:03 AM - restoring image csaptex_1350979214
 

4. Another weird thing, as I wanted to verify if the medias are in Robot, I tried vmcheckxxx -rt tld -rn 2 -list | grep <media_id>. It didn't give any update. Also, vmupdate -rt tld -rn 2 failed with error "robot number and robot host mismatch (61)". but in GUI Robot inventory of that particular robot was successful. Didn't get that, Why? :(

Sorry for posting so many questions in one thread. As the base reason of all these issue is restore not starting due to media and all these steps are checking if the media is available in the robot and/or in ready to use status, I am posting in the same thread.

Regards,

Raj

..Raj

Nicolai's picture

3: I have also seen this happen from time to time. I have always believed it's internal Netbackup processing - some process hammering the Netbackup Catalog (but I may be wrong) . How may files are stored in the image csaptex_1350979214 ?.

I believe if you run vmupdate on a media server that not the control host-  you will see robot host mismatch. Add the -rh robot_control_host option to vmupdate to resolve .

http://www.symantec.com/docs/HOWTO43755

Assumption is the mother of all mess ups.

If this post answered your'e qustion -  Please mark as a soloution.

sabby's picture

"4. Another weird thing, as I wanted to verify if the medias are in Robot, I tried vmcheckxxx -rt tld -rn 2 -list | grep <media_id>. It didn't give any update. Also, vmupdate -rt tld -rn 2 failed with error "robot number and robot host mismatch (61)". but in GUI Robot inventory of that particular robot was successful. Didn't get that, Why? :("

Raj, to answer to the above queston I think you are running the commands like vmcheckxxx and vmupdate on the master server and not the media server. These commands will only run on master when the tape library is attached to master server itself i.e. master is acting media server as well.

You need to run the vmcheckxxx and vmupdate commands on the media server, then you'll get desired result.

Regards,

Sahil Bhatia

raj08's picture

Thanks Sabby and Nicolai. Yes, I had to add -rh <media_server> along with vmupdate command.

Can someone please help me understand question 1 and 2?

..RAJ

..Raj

Marianne's picture

It's 2 months later! Thought you have lost interest...

I have also seen Activity Monitor stuck on 'Media XXXXXX needed'.

When it gets stuck there, I look for this hung bprd process in bpps output:

bprd -dontfork -mpxmain

Every time I have experienced the exact same issue, I have noticed this bprd process.

Workaround is to cancel the restore and kill this bprd process.

Subsequent restart of restore is normally successful.

I have not been able to find the real cause or solution - once logged a call with support, but the engineer needed a level 5 bprd log. Because this issue is not seen too often, it was out of the question to leave logging at level 5 for just 'in case' the problem re-occured. The work-around is much quicker.

Supporting Storage Foundation and VCS on Unix and Windows as well as NetBackup on Unix and Windows
Handy NBU Links

SOLUTION
raj08's picture

Thanks Marianne. WIll try that next time :)

..RAJ

..Raj