Unable to backup Exchange 2010 DAG with Symantec Netbackup 7.x

[IT_GCOSI]

Hi Guys,

Here's our problem :

Backup server
Windows Server 2008 R2
Netbackup 7.1.0.3

Exchange Server
Windows Server 2008 R2
Exchange 2010 Server w/o SP
Netbackup Agent 7.1.0.3
Server01 -> Hub 01
Server02 -> Mailbox 01
Server03 -> Hub 02
Server04 -> Simultaneous Copy of Server02's store

Unfortunately, we are unable to backup exchange through the DAG method. We got this error message :

2012-05-09 08:09:07 - Info nbjm(pid=6708) starting backup job (jobid=11914) for client DAG1, policy Exchange_DAG_Backup, schedule Full 
2012-05-09 08:09:07 - Info nbjm(pid=6708) requesting MEDIA_SERVER_ONLY resources from RB for backup job (jobid=11914, request id:{43FF186F-197A-4F57-B551-9B529F66F8B4}) 
2012-05-09 08:09:07 - requesting resource SERVERBKP01-hcart-robot-tld-0
2012-05-09 08:09:07 - requesting resource SERVERBKP01.NBU_CLIENT.MAXJOBS.DAG1
2012-05-09 08:09:07 - requesting resource SERVERBKP01.NBU_POLICY.MAXJOBS.Exchange_DAG_Backup
2012-05-09 08:09:07 - requesting resource EXCHANGE_RESOLVER.SERVERBKP01.Exchange_DAG_Backup.DAG1
2012-05-09 08:09:07 - granted resource SERVERBKP01.NBU_CLIENT.MAXJOBS.DAG1
2012-05-09 08:09:07 - granted resource SERVERBKP01.NBU_POLICY.MAXJOBS.Exchange_DAG_Backup
2012-05-09 08:09:07 - granted resource EXCHANGE_RESOLVER.SERVERBKP01.Exchange_DAG_Backup.DAG1
2012-05-09 08:09:07 - estimated 0 Kbytes needed
2012-05-09 08:09:07 - begin Parent Job
2012-05-09 08:09:07 - begin Unknown Type, Start Notify Script
2012-05-09 08:09:07 - Info RUNCMD(pid=4436) started           
2012-05-09 08:09:07 - Info RUNCMD(pid=4436) exiting with status: 0        
Status 0
2012-05-09 08:09:07 - end Unknown Type, Start Notify Script; elapsed time: 00:00:00
2012-05-09 08:09:07 - begin Unknown Type, Step By Condition
Status 0
2012-05-09 08:09:07 - end Unknown Type, Step By Condition; elapsed time: 00:00:00
2012-05-09 08:09:07 - begin Unknown Type, Read File List
Status 0
2012-05-09 08:09:07 - end Unknown Type, Read File List; elapsed time: 00:00:00
2012-05-09 08:09:07 - begin Unknown Type, Resolver Discovery
2012-05-09 08:09:08 - started process bpbrm (3040)
2012-05-09 08:09:09 - Info bpbrm(pid=3040) DAG1 is the host to restore to     
2012-05-09 08:09:09 - Info bpbrm(pid=3040) reading file list from client       
2012-05-09 08:09:14 - Info bpbrm(pid=3040) client_pid=3052           
2012-05-09 08:09:15 - Info bpbrm(pid=3040) from client DAG1: TRV - BPRESOLVER has executed on server (SERVER01) 
2012-05-09 08:09:19 - Info bpresolver(pid=3052) done.  status: 2        
2012-05-09 08:09:19 - Error bpbrm(pid=3040) You cannot use the backup directive for Database Availability Groups for an Exchange Standalone backup
2012-05-09 08:09:24 - Info bpresolver(pid=3052) done. status: 72: the client type is incorrect in the configuration database
Status 2
2012-05-09 08:09:24 - end Unknown Type, Resolver Discovery; elapsed time: 00:00:17
2012-05-09 08:09:24 - begin Unknown Type, Stop On Error
Status 0
2012-05-09 08:09:24 - end Unknown Type, Stop On Error; elapsed time: 00:00:00
2012-05-09 08:09:24 - begin Unknown Type, End Notify Script
2012-05-09 08:09:24 - Info RUNCMD(pid=5584) started           
2012-05-09 08:09:24 - Info RUNCMD(pid=5584) exiting with status: 0        
Status 0
2012-05-09 08:09:24 - end Unknown Type, End Notify Script; elapsed time: 00:00:00
Status 2
2012-05-09 08:09:24 - end Parent Job; elapsed time: 00:00:17
none of the requested files were backed up(2)

We applied a lot of changes based on what we found in many similar post, but the result is the same. No Backup...

We have upgraded the Netbackup version to 7.1.0.3 (Servers and Agents). We do not have license for 7.5.

We are wondering if upgrading our Exchange 2010 Farm to SP2 could help in this case.

Any Q for this problem?

Best regards,

Johnny

[Nicolai]

Do you have both DAG and non-DAG (2010 Standalone) on the same server ?

2012-05-09 08:09:19 - Error bpbrm(pid=3040) You cannot use the backup directive for Database Availability Groups for an Exchange Standalone backup

Somthing indicate it.

[IT_GCOSI]

Both SERVER02 and SERVER04 are DAG

[Nicolai]

But do you have information stores on one of the servers that are not DAG ?

[IT_GCOSI]

How can I valid this information?

[Gerald_W__Gitau]

Not sure if this will help but check this:

https://www-secure.symantec.com/connect/forums/netbackup-7-and-exchange-2010-dag-backups

[J_H_Is_gone]

can you post your policy so we can get a look at how you have it setup

bppllist <policyname> -L

[IT_GCOSI]

Ok, here what I found.

I've followed this case : http://www.symantec.com/business/support/index?page=content&id=TECH177360

And

Get-StorageGroupCopy
Get-StorageGroupCopyStatus

Those are unknown commands from the powershell of my Exchange 2010 Servers. What should I do in this case?

Regards,

Johnny

[rawilde]

Looking at your DAG layout of:

Server01 -> Hub 01
Server02 -> Mailbox 01
Server03 -> Hub 02
Server04 -> Simultaneous Copy of Server02's store

It would appear that servers 01 and 03 only have the hub transport roles and do not have the mailbox role.  Is this correct? 

The resolver job ran on Server01 (must currently be active for the DAG virtual name) and complained that its not the right server type for a DAG backup. Can you move the active virtual name to a server that has a milbox role and try the backup again?

I am guessing that NetBackup cannot detect any Exchange store on teh node (since it does not have the mailbox role) and makes the assumption that its not really part of a DAG.

[IT_GCOSI]

Hi rawilde,

Yes, you're right. Server01 and Server03 have both Client-Access and Hub Transport roles.

Server02 and Server04 have both Malbox Role and are part of the DAG named DAG1.

In the dag configuration windows, there's the Witness Server Field. This is where Server01 appear. Is that the one you want me to change for Server02 ou Server04?

Thanks!

Johnny

[rawilde]

That is probably just used for defineind the witness share host.  I would try to use the "failover cluster manager" on one of those nodes and failover the virtual DAG name.  I don't know if you can explicitly move it, but if you suspend a given node, it will move to another node.  If you can get it moved to a mailbox node, try the backup again.

[IT_GCOSI]

Hi Rawilde,

You were right,  something was wrong about the Exchange Cluster but even after fixing the problem... Backing Up DAG do not work...

I guess updating my Exchange 2010 farm to SP2 with Rollup 2 is the only option for us. But before  that, I would prefer be able to do a full backup of my Exchange Farm.

So is there anything else?

Just a remimber, I still got this error message when trying to backup Exchange DAG :

2012-05-10 10:22:32 - Info nbjm(pid=6708) starting backup job (jobid=11986) for client DAG1, policy Exchange_DAG_Backup, schedule Full 
2012-05-10 10:22:32 - Info nbjm(pid=6708) requesting MEDIA_SERVER_ONLY resources from RB for backup job (jobid=11986, request id:{F0C169EC-E952-4043-A83C-87171D5EC766}) 
2012-05-10 10:22:32 - requesting resource SERVERBKP01-hcart-robot-tld-0
2012-05-10 10:22:32 - requesting resource SERVERBKP01.NBU_CLIENT.MAXJOBS.DAG1
2012-05-10 10:22:32 - requesting resource SERVERBKP01.NBU_POLICY.MAXJOBS.Exchange_DAG_Backup
2012-05-10 10:22:32 - requesting resource EXCHANGE_RESOLVER.SERVERBKP01.Exchange_DAG_Backup.DAG1
2012-05-10 10:22:32 - granted resource SERVERBKP01.NBU_CLIENT.MAXJOBS.DAG1
2012-05-10 10:22:32 - granted resource SERVERBKP01.NBU_POLICY.MAXJOBS.Exchange_DAG_Backup
2012-05-10 10:22:32 - granted resource EXCHANGE_RESOLVER.SERVERBKP01.Exchange_DAG_Backup.DAG1
2012-05-10 10:22:32 - estimated 0 Kbytes needed
2012-05-10 10:22:32 - begin Parent Job
2012-05-10 10:22:32 - begin Unknown Type, Start Notify Script
2012-05-10 10:22:32 - Info RUNCMD(pid=3520) started           
2012-05-10 10:22:33 - Info RUNCMD(pid=3520) exiting with status: 0        
Status 0
2012-05-10 10:22:33 - end Unknown Type, Start Notify Script; elapsed time: 00:00:01
2012-05-10 10:22:33 - begin Unknown Type, Step By Condition
Status 0
2012-05-10 10:22:33 - end Unknown Type, Step By Condition; elapsed time: 00:00:00
2012-05-10 10:22:33 - begin Unknown Type, Read File List
Status 0
2012-05-10 10:22:33 - end Unknown Type, Read File List; elapsed time: 00:00:00
2012-05-10 10:22:33 - begin Unknown Type, Resolver Discovery
2012-05-10 10:22:33 - started process bpbrm (1484)
2012-05-10 10:22:34 - Info bpbrm(pid=1484) DAG1 is the host to restore to     
2012-05-10 10:22:34 - Info bpbrm(pid=1484) reading file list from client       
2012-05-10 10:22:40 - Info bpbrm(pid=1484) client_pid=4584           
2012-05-10 10:22:40 - Info bpbrm(pid=1484) from client DAG1: TRV - BPRESOLVER has executed on server (SERVERC01) <-- ?????
2012-05-10 10:22:42 - Info bpresolver(pid=4584) done.  status: 2        
2012-05-10 10:22:42 - Error bpbrm(pid=1484) You cannot use the backup directive for Database Availability Groups for an Exchange Standalone backup
2012-05-10 10:22:47 - Info bpresolver(pid=4584) done. status: 72: the client type is incorrect in the configuration database
Status 2
2012-05-10 10:22:47 - end Unknown Type, Resolver Discovery; elapsed time: 00:00:14
2012-05-10 10:22:47 - begin Unknown Type, Stop On Error
Status 0
2012-05-10 10:22:47 - end Unknown Type, Stop On Error; elapsed time: 00:00:00
2012-05-10 10:22:47 - begin Unknown Type, End Notify Script
2012-05-10 10:22:48 - Info RUNCMD(pid=1856) started           
2012-05-10 10:22:48 - Info RUNCMD(pid=1856) exiting with status: 0        
Status 0
2012-05-10 10:22:48 - end Unknown Type, End Notify Script; elapsed time: 00:00:01
Status 2
2012-05-10 10:22:48 - end Parent Job; elapsed time: 00:00:16
none of the requested files were backed up(2)

Thanks!

Johnny

[rawilde]

Johny

We usually get a gethostname function in bpresolver on the host that is running resolution and return that to the server.  Could the active cluster node have another name assigned to it?

I would find out what node of your DAG actually executed bpresolver at the time of the backup attempt.  If you enabled NetBackup logging, it should appear under netbackup\logs\bpresolver.

Lets start there.

[IT_GCOSI]

Ok, about the cluster.
Cluster Name = DAG1
Nodes Names : SERVER02 and SERVER04 (Ther have Mailbox Roles Installed on them, that's it)

SERVER01 have Client-Access and Hub Transport roles installed.

And about enabling logging, are you talking on the Netbackup Master Server or on each NetBackup client? There is a lot of logging actually but when I go to Netbackup\Logs\ there is no bpresolver directory...

[rawilde]

bpresolver will run on the client.  At the time of backup, it should run on the node that is active for the DAG virtual name.

under logs you can eitehr create bpresolver folder or run the mklogdir.cmd script.

[Mick_Scott]

Looks like something wrong with host names as NBU communication with DAG is going to SERVERC01

Everywhere else you refer to the hosts as server01 and server02  is the "C"  and even in the first log out put at top of post.

Double check IP addresses and name resolution

also have a look at the following

http://www.symantec.com/docs/TECH144855

[IT_GCOSI]

rawilde, I've created the bpresolver dir on each Exchange Server but when I ran a backup job, no logs appeared in that new dir.

I wonder how to enable logging for bpresolver...

[Yogesh9881 · Accepted solution]

Just for a testing ...File Level Backup 

can you create another backup policy & check file level backup status for all DAG members servers ?

it will help us to isolate the issue ...

we need o/p as asked by J.Hinchcliffe  bppllist <policyname> -L

i will suggest you update host file on Master / media & every DAG member server  as shown below..

10.xx.xx.xx   SERVER01   SERVER01.mydomain.co.in

[IT_GCOSI]

As I said from the beginning :

SERVER02 and SERVER04 are DAG Servers.

SERVER01 is the DAG Witness.

In the DNS, DAG IP address is pointing on the IP of SERVER02

I've changed the DAG Witness for SERVER02 and I still get this error :  Info bpbrm(pid=3280) from client DAG1: TRV - BPRESOLVER has executed on server (SERVER01)

I can't get it.

[rawilde]

As long as the logs\bpresolver folder exists, it should log (even with no verbosity set). From the bpbrm logs form the server, you can see what network name NetBackup is starting bpresolver on.  I may suggest opening a support case.  Seems like there are some network name issues (as NetBackup sees them)