Networker and SQL cluster error

Hello Ian,

Are you running the backup against any of the physical nodes or the virtual one?

Based on what I saw in Config Checker output you have Data ONTAP VSS provider, and this could be the reason of the failure, however I would need the nmm.raw file rendered from the active node, so I can gather more information.

VSS_PROVIDER_LIST

Provider name: Microsoft Software Shadow Copy provider 1.0 of Provider type: System, Version: 1.0.0.7 is available.

  Provider name: Data ONTAP VSS Hardware Provider of Provider type: Hardware , Version: 6. 3. 1. 5028 is available. Check for hardware provider support for NMM.

There is a workaround from Microsoft

ot instruct backup software to use VSS software provider (Microsoft) instead of HW provider. These spets refer to Windows 2003:

1. Click Start, click Run, type regedit , and then click OK.
2. Locate the following registry subkey, and then right-click it:
   HKey_Local_Machine\SYSTEM\CurrentControlSet\Control\BackupRestore

3. Point to New, and then click DWORD Value.
4. Type UseMicrosoftProvider as the name for the new DWORD value, and then press ENTER.
5. Double-click UseMicrosoftProvider. To enable the new feature, type 1 in the Value box. To disable the new feature, type 0 in the Value box.
6. Press ENTER, and then exit Registry Editor.
7. Restart the server.

I didn't try yet on Windows 2008, but I'm sure there should be some KB in Microsoft's web site.

You could also try to unregister Data ONTAP VSS provider as follows:

navssprv.exe -r service /u

Anyway I think this is a known limitation as per the following KB, maybe EMC could provide us further details on the RFE status (LGTsc31666)

https://solutions.emc.com/emcsolutionview.asp?id=esg106853

Thank you.

Carlos.

Hi Carlos

Thanks for your post.

Attached are the nmm.raw files from both nodes.

Thanks

Ian

Hi Ian,

Can you please ensure all services are up and running?

Also please run the following command and please paste in here the output:

nsrsnap_vss_save -c virtual_cluster_name -?

Also please take a look at this post, in regards to the configuration for multiple SQL instances:

https://community.emc.com/message/588309#588309

Thank you.

Carlos.

Hey Ian,

Any luck with the SQL backup?

Than you.

Carlos.

Hi Carlos,

Sorry about delay in answering you. We have managed to get past the original error which had to do with the Data ONTAP VSS provider. There was a registry entry and another entry in the client setup area in the Application Information attribute.

We are now experiencing another problem. When we try to create client resources for the physical cluster nodes we get an error that the server name already exists, but it doesn't. Any ideas how to get around this problem?

Thanks

Ian

Hello Ian,

Yes, first of all you can run this command:

mminfo -q client=short_name -r client,clientid

then

mminfo -q client=fqdn_client -r client,clientid

Where short_name is the short name for the physical nodes, and fqdn_client is the fqdn for the physical nodes.

If you get different ouotputs means that there is a duplicated entry in the Media DB, however, if you see the same output for both commands, or more likely "no matches found", then you will need to engage EMC support so they can use certain tool to get rid of the entry for that client.

Is there any client ID in the error message you are getting? something like "client xxx with clientID already exists" or similar, or the client ID is not mentioned in the error message?

Check that the name of the physical nodes is NOT in the alias of any other client, including the virtual node.

Also check if there is any folder within /nsr/index for the physical nodes.

Thank you.

Carlos.

Hi Carlos

I'm happy to say that the problems have been resolved. A co-worker of mine managed to sort out all the problems. Unfortuanately I was not present to see what he did to resolve it.

Thanks very much for all your assistance.

Ian