Exchange 2010 backup of two DBs(two nodes: each node have one Live DB)

[Zahid_Haseeb]

Environment

Netbackup Server and Client versions = 7.1

Exchange Server = 2010 (two nodes)

Exchange Server Node1 has a database of let suppose 50Mailboxes (DB1)

Exchange Server Node2 has a database of let suppose 50 other Mailboxes (DB2)

Exchange Server node1 replicating its database to Exchange Server node2 (replicating not on node2 DB)

Exchange Server node2 replicating its database to Exchange Server node1 (replicating not on node1 DB)

If any Exchange node will fail the another node will mount the replicating DB(which was in healthy form) and the users who were hosted on the lost / failed Exchange Server can browse their Mailboxes from other Exchange Server node. Means both 50+50 users will be on a single Exchange Server node.

My questions are:

1.) Is that possible that I can take the backup of both DB's(node1 and node 2) in case of they are on different nodes and also in case of both DB's are on single node(in case of failure) 

What should i define in the Netbackup Exchange Policy Backup Selection List ?

What should i define in the Netbackup Exchange Policy Clients ?

Do we have to use the snapshot option on the attributes in Exchange Server policy ?

[rizwan84tx]

Since both the nodes host different Mailbox DBs, then we have backup both exchange server using MS-Exchange agent.

[Zahid_Haseeb]

Activity Monitor

4/28/2011 2:31:11 PM - Info nbjm(pid=2752) starting backup job (jobid=44) for client VirtualDAG, policy Exchange2010, schedule Exchcange2010 

4/28/2011 2:31:11 PM - Info nbjm(pid=2752) requesting MEDIA_SERVER_ONLY resources from RB for backup job (jobid=44, request id:{F9F57620-3975-4586-9AA2-84E1DD7EBDBD}) 

4/28/2011 2:31:11 PM - requesting resource Exchange2010

4/28/2011 2:31:11 PM - requesting resource NBU.NBU_CLIENT.MAXJOBS.VirtualDAG

4/28/2011 2:31:11 PM - requesting resource NBU.NBU_POLICY.MAXJOBS.Exchange2010

4/28/2011 2:31:11 PM - requesting resource EXCHANGE_RESOLVER.NBU.Exchange2010.VirtualDAG

4/28/2011 2:31:11 PM - granted resource NBU.NBU_CLIENT.MAXJOBS.VirtualDAG

4/28/2011 2:31:11 PM - granted resource NBU.NBU_POLICY.MAXJOBS.Exchange2010

4/28/2011 2:31:11 PM - granted resource EXCHANGE_RESOLVER.NBU.Exchange2010.VirtualDAG

4/28/2011 2:31:11 PM - estimated 0 Kbytes needed

4/28/2011 2:31:11 PM - begin Parent Job

4/28/2011 2:31:11 PM - begin Unknown Type, Start Notify Script

4/28/2011 2:31:11 PM - Info RUNCMD(pid=1960) started           

4/28/2011 2:31:11 PM - Info RUNCMD(pid=1960) exiting with status: 0        

Status 0

4/28/2011 2:31:11 PM - end Unknown Type, Start Notify Script; elapsed time: 00:00:00

4/28/2011 2:31:11 PM - begin Unknown Type, Step By Condition

Status 0

4/28/2011 2:31:11 PM - end Unknown Type, Step By Condition; elapsed time: 00:00:00

4/28/2011 2:31:11 PM - begin Unknown Type, Read File List

Status 0

4/28/2011 2:31:11 PM - end Unknown Type, Read File List; elapsed time: 00:00:00

4/28/2011 2:31:11 PM - begin Unknown Type, Resolver Discovery

4/28/2011 2:31:12 PM - started process bpbrm (4624)

4/28/2011 2:34:51 PM - Error bpbrm(pid=4624) Cannot connect to VirtualDAG        

4/28/2011 2:34:51 PM - Info bpbrm(pid=4624) VirtualDAG is the host to restore data     

4/28/2011 2:34:56 PM - Info (pid=0) done. status: 58: can't connect to client     

Status 26

4/28/2011 2:34:56 PM - end Unknown Type, Resolver Discovery; elapsed time: 00:03:45

4/28/2011 2:34:56 PM - begin Unknown Type, Stop On Error

Status 0

4/28/2011 2:34:56 PM - end Unknown Type, Stop On Error; elapsed time: 00:00:00

4/28/2011 2:34:56 PM - begin Unknown Type, End Notify Script

4/28/2011 2:34:56 PM - Info RUNCMD(pid=2252) started           

4/28/2011 2:34:56 PM - Info RUNCMD(pid=2252) exiting with status: 0        

Status 0

4/28/2011 2:34:56 PM - end Unknown Type, End Notify Script; elapsed time: 00:00:00

Status 26

4/28/2011 2:34:56 PM - end Parent Job; elapsed time: 00:03:45

client/server handshaking failed(26)

[Zahid_Haseeb]

No need of two policies only one policy will work. This is Microsoft recommanded HA solution which i have described above. We can take the backup via virtualDAG name which is the feature of this Exchange HA scenerio. The virtualDAG will take care both mailboxStore/Database by his own self

[Zahid_Haseeb]

The above error was resolved . I just stop the firewall on both Exchange Nodes(Does not stop from the Administrative Tools=>Services. This will disconnect you machine from Network).

See the below SNAP for your reference. Select the RED highlighted box and click on OK:

[Zahid_Haseeb · Accepted solution]

The above asked questions was being resolved by Symantec Support. I would also like to share these.

1.) Is that possible that I can take the backup of both DB's(node1 and node 2) in case of they are on different nodes and also in case of both DB's are on single node(in case of failure) 

2.) What should i define in the Netbackup Exchange Policy Backup Selection List ?

3.) What should i define in the Netbackup Exchange Policy Clients ?

4.) Do we have to use the snapshot option on the attributes in Exchange Server policy ? 

Answer#1) Either the both DB's(databases) are on their own machines or either on single node the virtualDAG look after both DB's locations in any case. The virtualDAG is mentioned on the Client TAB. See the SNAP for reference below:

Answer#2) What should i define in the Netbackup Exchange Policy Backup Selection List ?

Answer#3) What should i define in the Netbackup Exchange Policy Clients ?

We will mention the virtualDAG name in the Clients TAB

Answer#4) Do we have to use the snapshot option on the attributes in Exchange Server policy ?

a.) As far as this question is concern. We must select the Perform snapshot backups.

b.) In the Options select the VSS.

c.) As far as the Database Backup sources. There are three options.

First will take the backup of mounted DB's of both Nodes

Second will take the passive copies means the replicated copies (from the preferred server list means the passive copies will be backedup from Servers/Nodes which we mentioned in the preferred server list).

Third one means that if the passive copies are not available in case of node failure/corrupted copies etc the backup will be performed of Active 

[rizwan84tx]

Zahid,

Thanks for sharing the TIP....Pls mark your post as solution for others reference...