Video Screencast Help

New server and upgrade to 7.5.0.5

Created: 03 May 2013 | 17 comments

So I have a new 2008 server. 

 

Same name, same IP etc as the old 2003 6.5.5 server.

 

Moved everything over as 6.5.5 and seemed to test fine.

 

Upgrade to 7.5.0.5 and had some issue.  catalog not working, image policys etc.  Fixed all them

 

Now I can't restore a back.  Clients all see the backup Backup<Archive, and restore

 

But if I run a backup I get this.

 

 1370,2,5,2850,,,uns40-22.americas.ad.nsk.com,uns40-88,1367603486,0000000668,0000000000,,1,,,,/I/data/IST/IST/NiemanN/OneNote/,0,1520,root,,,,90000,root,uns40-88,,,,0,47,1,/I/data/IST/IST/NiemanN/OneNote/,1,1520,,,1367603488,0000000666,0000000000,,,31,5/3/2013 1:51:26 PM - begin Restore operation,5/3/2013 1:51:28 PM - 1 images required,5/3/2013 1:51:28 PM - media 7216L5 required,5/3/2013 1:51:28 PM - media 7211L5 required,5/3/2013 1:51:28 PM - media 7214L5 required,5/3/2013 1:51:29 PM - restoring image uns40-22_1364596650,5/3/2013 1:51:31 PM - Info bpbrm(pid=2676) uns40-22 is the host to restore to,5/3/2013 1:51:31 PM - Info bpbrm(pid=2676) telling media manager to start restore on client,5/3/2013 1:51:33 PM - Info bpbrm(pid=2108) uns40-22 is the host to restore to,5/3/2013 1:51:33 PM - connecting,5/3/2013 1:51:34 PM - Info bpbrm(pid=2108) start tar32 on client,5/3/2013 1:51:36 PM - Error bptm(pid=4728) The following files/folders were not restored:,5/3/2013 1:51:36 PM - Error bptm(pid=4728) UTF - /I/data/IST/IST/NiemanN/OneNote/,5/3/2013 1:51:36 PM - Error bptm(pid=4728) UTF - /I/data/IST/IST/NiemanN/OneNote/Server Login.one,5/3/2013 1:51:36 PM - Error bptm(pid=4728) UTF - /I/data/IST/IST/NiemanN/OneNote/Unfiled Notes.one,5/3/2013 1:51:36 PM - Error bptm(pid=4728) UTF - /I/data/IST/IST/NiemanN/OneNote/Lotus Notes/,5/3/2013 1:51:36 PM - Error bptm(pid=4728) UTF - /I/data/IST/IST/NiemanN/OneNote/Lotus Notes/New Section 1.one,5/3/2013 1:51:36 PM - Error bptm(pid=4728) UTF - /I/data/IST/IST/NiemanN/OneNote/Lotus Notes/Open Notebook.onetoc2,5/3/2013 1:51:36 PM - Error bptm(pid=4728) UTF - /I/data/IST/IST/NiemanN/OneNote/OneNote 2010 Guide/,5/3/2013 1:51:36 PM - Error bptm(pid=4728) UTF - /I/data/IST/IST/NiemanN/OneNote/OneNote 2010 Guide/Abbreviations.one,5/3/2013 1:51:36 PM - Error bptm(pid=4728) UTF - /I/data/IST/IST/NiemanN/OneNote/OneNote 2010 Guide/EEFF.one,5/3/2013 1:51:36 PM - Error bptm(pid=4728) UTF - /I/data/IST/IST/NiemanN/OneNote/OneNote 2010 Guide/EndPoint Encryption For PC.one,5/3/2013 1:51:36 PM - Error bptm(pid=4728) more than 10 files were not restored\, remaining ones are shown in the progress log.,5/3/2013 1:51:36 PM - Info bpbrm(pid=2676) got

 

ERROR 95 from media manager,5/3/2013 1:51:36 PM - Info bpbrm(pid=2676) terminating bpbrm child 2108 jobid=1370,5/3/2013 1:51:36 PM - restored image uns40-22_1364596650 - (requested media id is not assigned to this host in the EMM database(95)),5/3/2013 1:51:39 PM - Warning bprd(pid=1520) Restore must be resumed prior to first image expiration on INFINITY      ,5/3/2013 1:51:40 PM - end Restore operation; operation time: 000:00:14,5/3/2013 1:51:34 PM - requesting resource 7211L5,5/3/2013 1:51:34 PM - Error nbjm(pid=3248) NBU status: 95\, EMM status: Media does not exist,5/3/2013 1:51:34 PM - Error nbjm(pid=3248) NBU status: 95\, EMM status: Media does not exist,0,0,,,,,,,,,,uns40-88,7211L5,,,1,1,,,,,1,,,, ,

Ok it is like the media is expired even though it is set to infinity.  We can't expire our backups due to legal reason.
 
Did I miss a step when builing or upgrading the server
 
Server is a 2008 windows server 64Bit
Running 7.5.0.5
Media, and Master are the same server.
Library is a IBM 3310
Operating Systems:

Comments 17 CommentsJump to latest comment

NathanNieman's picture

Also see this when looking in the logs\user_ops

 

 

 

14:59:37 5/3/2013: Restore Started
 
14:59:38 (1371.xxx) Restore job id 1371 will require 1 image.
14:59:38 (1371.xxx) Media id 7216L5 is needed for the restore.
14:59:38 (1371.xxx) Media id 7211L5 is needed for the restore.
14:59:38 (1371.xxx) Media id 7214L5 is needed for the restore.
 
14:59:39 (1371.001) Restoring from copy 1 of image created 3/29/2013 6:37:30 PM
14:59:41 (1371.001) INF - If Media id 7211L5 is not in a robotic library administrative interaction may be required to satisfy this mount request.
14:59:46 (1371.001) Status of restore from copy 1 of image created 3/29/2013 6:37:30 PM = requested media id is not assigned to this host in the EMM database
 
14:59:47 (1371.xxx) INF - Status = Restore error.
 
15:03:33 (1371.001) INF - TAR STARTED 5096
15:08:34 (1371.001) INF - EXIT STATUS 36: failed trying to allocate memory
 
 
Not sure what this means
media id is not assigned to this host in the EMM database
NathanNieman's picture

Not in the robot but this happens for any restore jobs.  I have ran ones with tapes that are in the Libaray.  I try one from last night I get the same error.  Sorry I picked a bad backup to display.

Netbackup use to just say tape is not in Libaray, and let you load it so I didn't thing about that.

Inventory is up to date and works fine.  

 

Am I missing something?

 

LucSkywalker1957's picture

Go back and check all the steps in the install guide too just to cover all your bases. http://www.symantec.com/business/support/index?pag...

Your master and EMM server both the same server correct?

When you say you "fixed all them" in reference to your catalogs and images, what did you fix?

Hate to ask, did you reboot the server after your upgrade?

What does your application and system event logs say if anything?

It's almost like your EMM database isn't right or connected.

Maybe your device configs or media types got changed in Netbackup from what they actually are. Check that your inventory is listing all the media and devices are correct type and that they match. Can you do successful backups and do a successful test restore from the new backup? Is it only old data from the 6.5 environment that you can't restore from?

Also, does the new server have at least as much memory as the old server? Have you configured any virtual swap?

Marianne's picture

Moved everything over as 6.5.5 and seemed to test fine. 

How exactly was everything moved over?  What do you consider to be everything?

requested media id is not assigned to this host in the EMM database

The above says to me the image database was moved/copied but not EMM database.

The supported and recommended way is catalog recovery: http://www.symantec.com/docs/TECH77447

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

NathanNieman's picture

When I say moved we did a full hot catalog.

 

Built a new server with the same name and installed 6.5.5 just like the old.

Restore the catalog "Along with setting up all drives storage units etc"

Backups run fine, and Catalogs look fine.  We can see all the backups but when we try to do a restore on any date tape in libaray or not we get the above errror.

Marianne's picture

Hopefully the catalog recovery log file still exists?

We need to see this log file to find out if EMM database was recovered successfully. 
 <install-path>\VERITAS\NetBackup\Logs\user_ops\<user>\logs\Recover<date-time>.log 

Big problem if images and config (e.g. policies) were recovered and not EMM - all tapes would be unassigned and overwritten by new backups.
Images still know on which tape the backup should be and which media server wrote the backup.

Please show us output of:
nbemmcmd -listmedia -mediaid 7211L5

You may want to run a 'Tape Contents' report on 7211L5 to see what is actually written on the tape. 

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

NathanNieman's picture

Ok I can do this.  We still have the old Backup sever up it is just off the network.

 

All old tapes are no longer in the libaray and are stored off site so they shouldn't be writen on.

 

Ok I am seeing nothing in the LOG files under any user maybe the issues?

 

I runs the nbemmcmd and I get

 

NBEMMCMD, Version:7.5.0.5

The fuction returned the following failure status:

volume does not exist in database (35)

command did not complete successfully.

 

That doesn't sound good.

Marianne's picture

It depends on how long ago you did the catalog recovery.

I think restore logs are only kept for 3 days and not 28 like other logs.

Did you open the Media section of the GUI after you performed the catalog recovery? 
You should've been able to see all existing media with correct Assign Times and Pool info.

It seems as if EMM was not restored.

Big problem now is that you have already upgraded to 7.5 and started doing backups - so we cannot recover missing EMM entries from the old catalog backup.

The only solution that I can see is to run nbcc and submit to Symantec Support.
They will provide scripts to re-add media to EMM database.

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

NathanNieman's picture

I have a ticket open with them now.  This was done last week so it has been over 3 days.  I am in the GUI and everything looked good.  I swear we tested a restore but it looks like I was wrong.

 

Thank you for all your help I will update with status when we are fixed.

 

Mark_Solutions's picture

When you did your catalog recovery did you install NetBackup before changing the name of the new Server or has it always been the same name?

Does the new one and old one both have the same NetBackup name (i.e. server and server rather than server and server.domain.com)?

In relation to this line:

"Upgrade to 7.5.0.5 and had some issue. catalog not working, image policys etc. Fixed all them"

What exactly did you do to fix then all?

Authorised Symantec Consultant

Don't forget to "Mark as Solution" if someones advice has solved your issue - and please bring back the Thumbs Up!!.

NathanNieman's picture

Server have the same name and IP.  None of that was change.

 

catalog issue was the files were not brough over when moving to the new server.  There was a folder on the root of C the last admin made that the catalogs were stored, and saving to.  This was not moved.  Once I moved this catalogs started to work again.

Marianne's picture

Seems moving of catalogs to another location always create issues with catalog migration.

I'm still wondering what happened to your EMM database. Seems we will never know because the Recover log no longer exists.

All I can suggest now is to download nbcc from http://www.symantec.com/docs/TECH156730 and submit the output to Symantec Support.

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

Mark_Solutions's picture

Sounds like you have just copied the catalog files (images etc) across from the old server but have not restored the EMM Database.

This is why you now have no media, just a list of backup images.

Does the old server still exist along with its files?

Did you duplkicate the disk based catalogs to tape?

If not then i think you are quite possibly in a mess i am afraid.

Authorised Symantec Consultant

Don't forget to "Mark as Solution" if someones advice has solved your issue - and please bring back the Thumbs Up!!.

NathanNieman's picture

Mark

 

I didn't copy anything but the catalog.  I guess I missed this step some where

 

Yes the old server is still up and running just not on the network.  I can connect to it and copy what is needed I just need to know what is needed and where it is.

 

Yes all catalogs, jobs end up on tape.

 

Also before we upgraded to 6.5.5 I pulled the mirrored drive so if I have to go back to 6.5.5. I can swap the drive in and start over.  I just need to do this soon since it sounds like we are going to lose a week of backsup.

 

I have a call into the help desk also sent the NBCC abd NBSU files to them been a day since I have heard back.

 

So what needs to be copied from where I guess is my issues.  If I lose a week I can deal with that.  I can't lose 4 years worth of backups.

Marianne's picture

You cannot copy EMM database from the old server:

  1. The EMM db version is 6.5.x. You have in the meantime upgraded to 7.5.x. copying the database will break EMM on the new server.
  2. You have in the meantime started doing backups, so, you cannot go back and start all over.

The only solution is to wait for Support to get back with nbcc analysis and scripts.

Seems your tapes are at least safe as they are out of the robot.

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

NathanNieman's picture

Thank you Marianne.  Like i said if I have to I can put in the old Mirrored drive and start fress.

 

Yes the tape should be fine since 90% are off site.

 

I will wait to see what Supports says to do.