cancel
Showing results for 
Search instead for 
Did you mean: 

How to configure NBU service group in GCO with VVR

NBU_13
Level 6
Accredited Certified

Discussion split off from https://www-secure.symantec.com/connect/forums/gco-configuration-between-2-nodes-cluster-and-1node-c...:

I have configured - site 2 with 1 node cluster - I have install the SFWHA6.1 software and created dynamic disk group, volume and configured cluster with vip address, service group for wac and I tried to install Netbackup software in 1 node cluster, it is asking vip, I gave new vip address and it asked for path for shared disk. I mention manually as F: drive and clicked cluster configuration, it was not successful. but service group and resources are created with offline, 

error - cluster configuration  is unsuccessful, some of the resources are not come online

Please guide to fix it.

 

3 ACCEPTED SOLUTIONS

Accepted Solutions

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

You basically need to install and configure 2 x1 node clusters using the same virtual hostnames on both sites.

Your first manual to read through is NetBackup in Highly Available Environments Administrator's Guide 

Specifically this topic:

Installing and configuring a globally clustered NetBackup master server with catalog replication

Pay attention to VVR requirements for NBU databases.

You will see reference to  NetBackup Clustered Master Server Administrator's Guide  for 1-node cluster installation instructions.

Please note that only new NBU master servers can be clustered.

If you want to reconfigure existing master server, you will need consulting services from NBU Catalog Manipulation partner. See http://www.symantec.com/docs/TECH31385

View solution in original post

Wally_Heim
Level 6
Employee

Hi  NBU_13,

The issue that you are running into is that the NBU wizard is designed to only work with Clustered Diskgroups.  It sounds like you are not creating clustered diskgroups because you do not have shared disks since each site in a single node cluster. 

The vxclus command that Mike mentioned might help but it might not.  You do need separate physical disks from the OS drive.  Can you tell us a little more on the exact disk hardware and controllers being used?

The reason I'm asking is that many internal RAID controllers do not support SCSI 2 or SCSI 3 reservations at all. If you are using one of these controllers, then the vxclus command will not make any difference.

As long as you have separate physical disks from the OS, this is how I would approach resolving this issue.

 

1. Create a clustered disk group (do not deport it or you will not be able to reimport it again as a clustered diskgroup.

2. Create the volume structure as needed for NBU and VVR SRL.

3. Install NBU in cluster mode.

4. Once the NBU installation in completed, Configure VVR replication in VEA or CLI at this time.

5. Run the VVR Service Group configuration wizard to setup the VVR service group at this site.

6.  Offline the NBU and VVR service groups.

7.  Save and close the cluster configuration at this site and then stop the cluster with the following command.

   hastop -all

8.  In VEA or CLI, import the diskgroup as a Secondary diskgroup.

9.  Start the cluster again, by running

   hastart -all

10. Replace the VMDg resource in the VVR service group with the VMNSDg resource.  This resource works with Secondard diskgroups for single node cluster installations and VMware installations that do not need SCSI-3 reservations.

11. test that the service group comes online on this site.

 

Repeat these steps at the DR site using the same NBU Virtual server name.  The VIP can be different as needed for that site.  During VVR configuration setup the DR site as the VVR secondary.  And before online testing of the service group in the DR site, make sure that replication is up to date and the NBU groups are linked as Global groups.

Please let us know if you have any questions or problems.  Keep in mind that Symantec Technical Support is available if you run into any problems and need more direct asisstance with this issue.

Thank you,

Wally

 

View solution in original post

Wally_Heim
Level 6
Employee

Hi NBU_13,

Sorry, I the NBU installers is actually supported by the NBU team so I don't have much experiance with troubleshooting the NBU installer.  You may need to post in the NBU forum or open a case with NBU Techncial Support to get help with this one.

However, I would recommend checking the following before you move on to engage NBU help.

 

1. Make sure that SFW-HA is in a running state with the following command.

     hasys -state

If the cluster has not been created, then create it on this node and rerun the NBU installer once the cluster is in a running state.

 

2. Try using a drive letter that is F:\ or higher.  Some cluster aware applications do not like drive letters below F:\.  I'm not sure if NBU checks the drive leter or not.

 

I would think that the issue maybe that the cluster is not running but that is just a guess on my part.

Thank you,

Wally

View solution in original post

8 REPLIES 8

mikebounds
Level 6
Partner Accredited

How did you install NBU on primary site in 2-node cluster.

In general applications are installed as follows

Install application on node 1 using shared disk.

Move data installed on shared disk out of the way and import shared disk on node2 using same drive letter that was used on node 1 and install application on node 2 using shared disk.

If using VVR, then if VVR is not configured yet, then install application on DR node using same drive letter that was used on node 1.  Then sync VVR which will overwrite shared data on DR node with shared data that was created on primary cluster.

Sometimes there are wizards to help you with this process.

Mike

NBU_13
Level 6
Accredited Certified

Hi Mike,

In site 1, I have 2 node cluster, while installing netbackup on it with cluster install option, it asked the vip address and it was auto selected the shared disk for Netbackup database and I have netbackup installed in site 1, Iam able to import and deport shared disk between 2 nodes in site 1. (also I have configured disk group as cluster disk group)

But in site 2, I have only 1 node cluster, so I have install SFHA software and configure disk group (dynamic disk group), because it is not allowed to create cluster disk group(due to no shared disk), and created volume with same F: drive.

Now, Iam trying to install netbackup in site 1 (1 node cluster) with cluster installation option, it was asked vip address, I gave new vip address (not a physical address) and it was not auto select F: drive in the shared disk path (Netbackup database), I have selected manually the F: drive and started cluster configuration, it was failed with created the Netbackup service group and resources, but it is showing as unsuccessful Iam not able to move to next steps.

 

mikebounds
Level 6
Partner Accredited

If you have no shared disk for 1-node cluster - i.e you only have internal disks on the same controller as the boot disk, then you need to use "vxclus UseSystemBus ON” and this will allow you to create a cluster diskgroup.  Note you need a separate disk to the boot disk so you can create a diskgroup - i.e the F: can't be a partition on the same disk as the C: drive - the F: dirve must be on a separate disk in its own diskgroup and you will need to create an SRL in this diskgroup for VVR (which can use the same or a separate disk) and leave space so that VVR can create DCMs (small bit maps).

Mike

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

You basically need to install and configure 2 x1 node clusters using the same virtual hostnames on both sites.

Your first manual to read through is NetBackup in Highly Available Environments Administrator's Guide 

Specifically this topic:

Installing and configuring a globally clustered NetBackup master server with catalog replication

Pay attention to VVR requirements for NBU databases.

You will see reference to  NetBackup Clustered Master Server Administrator's Guide  for 1-node cluster installation instructions.

Please note that only new NBU master servers can be clustered.

If you want to reconfigure existing master server, you will need consulting services from NBU Catalog Manipulation partner. See http://www.symantec.com/docs/TECH31385

Wally_Heim
Level 6
Employee

Hi  NBU_13,

The issue that you are running into is that the NBU wizard is designed to only work with Clustered Diskgroups.  It sounds like you are not creating clustered diskgroups because you do not have shared disks since each site in a single node cluster. 

The vxclus command that Mike mentioned might help but it might not.  You do need separate physical disks from the OS drive.  Can you tell us a little more on the exact disk hardware and controllers being used?

The reason I'm asking is that many internal RAID controllers do not support SCSI 2 or SCSI 3 reservations at all. If you are using one of these controllers, then the vxclus command will not make any difference.

As long as you have separate physical disks from the OS, this is how I would approach resolving this issue.

 

1. Create a clustered disk group (do not deport it or you will not be able to reimport it again as a clustered diskgroup.

2. Create the volume structure as needed for NBU and VVR SRL.

3. Install NBU in cluster mode.

4. Once the NBU installation in completed, Configure VVR replication in VEA or CLI at this time.

5. Run the VVR Service Group configuration wizard to setup the VVR service group at this site.

6.  Offline the NBU and VVR service groups.

7.  Save and close the cluster configuration at this site and then stop the cluster with the following command.

   hastop -all

8.  In VEA or CLI, import the diskgroup as a Secondary diskgroup.

9.  Start the cluster again, by running

   hastart -all

10. Replace the VMDg resource in the VVR service group with the VMNSDg resource.  This resource works with Secondard diskgroups for single node cluster installations and VMware installations that do not need SCSI-3 reservations.

11. test that the service group comes online on this site.

 

Repeat these steps at the DR site using the same NBU Virtual server name.  The VIP can be different as needed for that site.  During VVR configuration setup the DR site as the VVR secondary.  And before online testing of the service group in the DR site, make sure that replication is up to date and the NBU groups are linked as Global groups.

Please let us know if you have any questions or problems.  Keep in mind that Symantec Technical Support is available if you run into any problems and need more direct asisstance with this issue.

Thank you,

Wally

 

NBU_13
Level 6
Accredited Certified

Hi wally,

Thank you for information,

I have installed SFHA on 2 servers as node 1 and node 2, I have drive E:\ on both server, Ran vxclus UseSystemBus ON, while I try to configure disk group, I have option to select Create clustered diskgroup and create clsuter group, I have created disk groupo with create cluster group, but while installing netbackup, install netbackup cluster option is gray out, Iam not able to install netbackup in cluster mode.

please advise me.

Wally_Heim
Level 6
Employee

Hi NBU_13,

Sorry, I the NBU installers is actually supported by the NBU team so I don't have much experiance with troubleshooting the NBU installer.  You may need to post in the NBU forum or open a case with NBU Techncial Support to get help with this one.

However, I would recommend checking the following before you move on to engage NBU help.

 

1. Make sure that SFW-HA is in a running state with the following command.

     hasys -state

If the cluster has not been created, then create it on this node and rerun the NBU installer once the cluster is in a running state.

 

2. Try using a drive letter that is F:\ or higher.  Some cluster aware applications do not like drive letters below F:\.  I'm not sure if NBU checks the drive leter or not.

 

I would think that the issue maybe that the cluster is not running but that is just a guess on my part.

Thank you,

Wally

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

Wally is right - the basic Cluster config must be done and VCS up and running before you start with NBU installation.

All of this is documented in the NBU Clustered Master server guide that I have referenced above.

Please take your time and go through the requirements and checklist.