Raw partition restore from one VM to another? (error 2817)

[Nate_D1]

Hi All,

NBU version 7.5.0.3

Windows version 2008 std

I am trying to test how long it takes for us to do a full restore of one of our more important servers. Its a critical server so what I would like to do is take its last full backup (Flashbackup-Windows policy type) and then restore it to a newly created VM.

So I created a new VM, joined it to the domain, etc. and then I went to the BAR, I went file>specify netbackup machines and policy type. I tell it the source client is my critical server, I hit 'edit client list' and added the name of the destination restore server. I chose the name of the destination server under 'destination client for restores'. I chose the correct policy type, I hit 'OK' and it brings up the list of backups.  I then click actions>select restore type> raw partition backups.  I select a full backup, check the top box (next to the server name), and then hit 'start restore.

When I start the restore I get the following error:

2/8/2013 9:22:08 AM - begin Restore
2/8/2013 9:22:35 AM - restoring image emerald.domain1.xxxxxx_1359799810
2/8/2013 9:22:35 AM - Info bprd(pid=9992) Restoring from copy 1 of image created 02/02/13 02:10:10    
2/8/2013 9:22:35 AM - requesting resource @aaafE
2/8/2013 9:22:35 AM - granted resource MediaID=@aaafE;DiskVolume=HQ_VMware_Full.Colossus.xxxxxxx_HQ;DiskPool=HQ_VMware_Full;Path=HQ_VMware_Full.Colossus.xxxxxxx_HQ;StorageServer=xxxxxxx_HQ;MediaServer=xxxxxxx
2/8/2013 9:23:00 AM - Info bpbrm(pid=3340) connect failed STATUS (18) CONNECT_FAILED        
2/8/2013 9:23:00 AM - Info bpbrm(pid=3340)     status: FAILED, (42) CONNECT_REFUSED; system: (10061) No connection could be made because the target machine actively refused it. ; FROM 0.0.0.0 TO emerald_restore.domain1.xxxxx 10.1.255.181 bpcd VIA pbx
2/8/2013 9:23:00 AM - Info bpbrm(pid=3340)     status: FAILED, (42) CONNECT_REFUSED; system: (10061) No connection could be made because the target machine actively refused it. ; FROM 0.0.0.0 TO emerald_restore.domain1.xxxxx 10.1.255.181 bpcd VIA vnetd
2/8/2013 9:23:00 AM - Info bpbrm(pid=3340)     status: FAILED, (42) CONNECT_REFUSED; system: (10061) No connection could be made because the target machine actively refused it. ; FROM 0.0.0.0 TO emerald_restore.domain1.xxxxx 10.1.255.181 bpcd
2/8/2013 9:23:00 AM - Error bpbrm(pid=3340) Cannot connect to emerald_restore.domain1.xxxxxx        
2/8/2013 9:23:00 AM - Info tar32(pid=0) done. status: 58: can't connect to client      
2/8/2013 9:23:00 AM - Error bpbrm(pid=3340) client restore EXIT STATUS 58: can't connect to client    
2/8/2013 9:23:00 AM - restored image emerald.domain1.xxxxx_1359799810 - (can't connect to client(58)); restore time 00:00:25
2/8/2013 9:23:00 AM - end Restore; elapsed time: 00:00:52
FlashBackup-Windows policy restore error(2817)

Any thoughts on this?

[Nate_D1]

I also attempted to perform a full backup of the destination server to make sure netbackup 'knows' about it. Im not sure what else to do! I did some searching on the error and didnt find much.

[CRZ]

You're getting a status 58 after all connection attempts to 10.1.255.181 fail.  (Is this the expected IP of the client?)

There are scads of status 58 resources available on Connect and in SymWISE and they're usually pretty easy to diagnose and troubleshoot.

(Also, one of your "xxxxx" edits didn't get changed to "xxxxx," if that's important.  ;) )

[Nate_D1]

I will take a look at the 58 errors, thanks for the heads up on the edit :). It is the expected IP.

[Mark_Solutions]

If you are re-directing this to another machine (your new VM) then it seems odd that it is connecting to the original machine in the log - i would have expected it to be connecting to a different client name (that of the new VM)

Please clarify exactly your original backup type here and the steps you have taken to do the restore

Thanks

[Nate_D1]

I am using an existing Full backup taken by the 'flash-backup windows' policy type, from a VM named 'Emerald' and trying to do a raw partition restore on a newly created VM named 'Emerald_Restore'.

In the log above I see the 'restoring image...emerald...."

restoring from copy1 and then gives an address to one of our ExaGrid dedup appliances.

The long and short of it is, I am trying to test how quickly we could do a restore of one of our more critical servers. Im not sure how much easier this would be if I was just restoring to itself, but I cant really do that in this case.

[Nate_D1]

Is there a technote or document that outlines this somewhere? Or am I doing something that is just not supported?

Anyone have any ideas for me? Kind of in a pickle here :(

[Yasuhisa_Ishika]

I'm wondering whether you have configured VADP backup or regular FlashBackup.

if you configured VADP backup with FlashBackup-Windows policy type, you should recovery entire VM with different VM name then move or copy vmdk as you like.
Or, if you took regular FlashBackup-Windows backup, you can restore it to alternate client of same OS with enought disk space.

Have you already manage status code 58?

[Nate_D1 · Accepted solution]

Hi All,

Problem solved, for some reason I read somewhere that I needed to use the 'RAW ' backup type when restoring the full VM. I found a video somewhere where they used the 'virtual machine' type and when you select that type it brings up a wizard that walks through the rest of the process. I had no issues after that.