Solved!
Go to SolutionVADP backup problems after upgrade to 7.6.4
We had Networker 7.6.2 and VMware vSphere 4 => everything worked fine.
Upgraded VMware to vSphere 5 => everything worked fine.
Created new VMFS 5 datastores => not supported in my Networker 7.6.2 version.
Opened support case => please upgrade Networker to 7.6.3 => full of problems, more failed backups than working ones.
Upgraded to Networker 7.6.4.3 Build 1070 and still have some problems:
First problem:
VADP backups are working fine when I start them manual, but at night, when a lot of backups are running, then my VADP backups stall and I get the following media critical event:
"Waiting for ... writable volume(s) to backup pool 'Default' disk(s) or tape(s) on ..." although all my groups are connected with Media Pools with Datadomain devices.
Second problem:
All the FULL-VM restores I try also fail with the following error:
88342:nsrvadp_recover: Cannot fetch 'NSR hypervisor' resource for VC '.....'.
Anyone had the same problem or has a idea what the problem could be?
Anil_Kumar_Hari
13 Posts
899
0
Posted November 6th, 2012 21:00
Hi Kim,
Which cumulative hotfix did you apply in version 7.6.3? The 7.6.3.7 version is a good one if you wished to upgrade.
Anyway, for the first problem, it looks to be an issue with the configuration. Are you sure you have the group name ticked in the media pool properties AND, nothing mentioned in the client properties -> Pool attribute?
For the second problem, please paste the complete output that you get. Also, the contents in the application information section of the VMs and the proxy host would be helpful.
Thank you for the reply.
No idea what hotfix of 7.6.3 I installed, the support engineer delivered me the software.
I am sure that the group is linked as Data Source for the correct Media Pool. At the client, the Pool attribute is empty.
For the restore problem, here is the output:
Recovering files of client '< >' from server '< >'.
Connecting to host < >.
Recovering to original VC system.
Recovering to original ESX host system.
libCURL SSL support is ENABLED
NetWorker install path: C:\Program Files\Legato\nsr\.
88342:nsrvadp_recover: Cannot fetch 'NSR hypervisor' resource for VC '< >'.
Restore of virtual machine failed!...
Completion time: 7/11/2012 8:48:42
Total time taken: 0 hr(s) 00 min(s) 12 sec(s)
Server Application information:
VADP_MAX_BACKOFF_TIME=10
VADP_TRANSPORT_MODE=san|nbd
VADP_HOST=< >
VADP_BACKUPROOT=C:\VMMOUNT
VADP_MAX_RETRIES=0
Client Application information:
VADP_HYPERVISOR=< >
VADP_VM_NAME=< >
Hi Kim,
I had the same problem (second) with:
RedHat 6 -> NW Server 7.6.4.2 Build 1060
W2K8 R2 -> Storage Node 7.6.4.2 Build 1060
VADP_MAX_BACKOFF_TIME=10
VADP_TRANSPORT_MODE=hotadd
VADP_HOST=< >
VADP_BACKUPROOT=D:\
VADP_MAX_RETRIES=1
I did replace the IP of the vCenter by the FQDN in the window "VADP restore" during the save set recover, and then no more "Cannot fetch 'NSR hypervisor'.
It starts the recover and goes a more further step..
83342:nsrvadp_recover: Error writing to VMDK: 0 bytes written instead of 84 bytes.
3:nsrvadp_recover: cannot write to C:\Program Files\EMC NetWorker\nsr\logs\daemon.raw: No such file or directory
16294:nsrvadp_recover: Encountered an error recovering C:\Program Files\EMC NetWorker\nsr\logs\daemon.raw
53363:nsrvadp_recover: Recover of rsid 1981528804 failed: Error receiving files from NSR server `NW server'
73724:nsrvadp_recover: One or more recover threads failed to exit successfully
Failed to recover files: Parent recover thread exited with errors
Restore of virtual machine failed!...
Hi,
7.6.4.3 has the hot fix for:
NW139374 Unable to do VADP incremental backups for some VMs. Error "Could not obtain the drive letter for one of the drives"
NW142426 snapshot files are not removed after VADP image level san mode restore with more than 1 disks
NW142266 - VCB Proxy server and SDN: Bad error - The 'storage nodes' contains name of remote dedicated storage node 'nsrserverhost'
I would upgrade to version 7.6.4.3 to avoid all the above too.
Mustafa
I tried with the FQN instead of the IP-address and now I get the following error (almost identical as Bidonjack):
Warning top extent index reached maximum extent count
83342:nsrvadp_recover: Error writing to VMDK: 8192 bytes written instead of 175104 bytes.Warning top extent index reached maximum extent count
83342:nsrvadp_recover: Error writing to VMDK: 0 bytes written instead of 261632 bytes.3:nsrvadp_recover: cannot write to C:\Windows\System32\config\SOFTWARE{aa705413-272c-11e2-86c8-0010185af6de}.TMContainer00000000000000000002.regtrans-ms: No such file or directory
16294:nsrvadp_recover: Encountered an error recovering C:\Windows\System32\config\SOFTWARE{aa705413-272c-11e2-86c8-0010185af6de}.TMContainer00000000000000000002.regtrans-ms
53363:nsrvadp_recover: Recover of rsid 24357 failed: Error receiving files from NSR server `< >'
73724:nsrvadp_recover: One or
more recover threads failed to exit successfully
Failed to recover files: Parent recover thread exited with errors
Restore of virtual machine failed!...
For the backup problem: I also selected the correct Pool in the client configuration (instead of leaving it blank). I'll give it a try tonight.
kim.devos
11 Posts
899
0
Posted November 8th, 2012 02:00
Update:
Backup problem solved by adding the the correct Pool in the client configuration (instead of leaving it blank). Apparently, if the client must wait to long for a session on the Pool defined on the group, then the client uses the Pool defined on its own configuration.
The restore problem is a known problem. There is a patch available, but then we need to downgrade to 7.6.4.1 Build 1049 (or wait for the patch to be become available for our version).
Edit: removed NDMP file-level restore problem of deduplicated file-system, apparently not possible