Start a Conversation

Unsolved

This post is more than 5 years old

1830

January 28th, 2016 07:00

Recover Point For Vmware 4.3 Error:The same volume has been added at more than one RPA cluster.

Hi folks,

One of our project, i have faced an issue.

The project subject was Disaster Recovery side installation and replication between Production site to DR site. Of course the both side are using EMC hardware and softrware solution. RP4VM.

We used two different way to complete that project.  One of them, we took vm clones (which has a really huge disk size) to external storage. Then we migrated all VMs to DR site. And we started RP4VM sync job. And the sync job has been completed without any error.

Second option is, we migrated VM to DR site by using RP4VM solution directly. Because these VMs don't have much disk size. That's why we started the replication job directly.

Anyway,  when i manage RP4VM  disable job, test copy etc.  (which i prefered option 1)  , i got an error as i write on subject!   But when i try the others VM (which i prefered option two) i don't see any mistakes or error.

Therefore, how can i fix that problem as permanently? What is the root cause of that issue?

Looking forward all your response !

153 Posts

January 28th, 2016 07:00

Hi there,

Can you please elaborate on the error you’re seeing? What’s the status of the CG where these VMs are protected?

Thanks,

Zahid

EMC2

Twitter: @zDataGuy

January 28th, 2016 07:00

Hi Zahid

You can find these screenshots as below.

rp4vm_consistentgroup_screenshot.jpg

rp4vm_error.jpg

153 Posts

January 28th, 2016 08:00

So where those CGs pre-seeded? I would try removing the CG and re-protecting the VM and see if that eliminates the error. It could be a result of the migration you outlined below.

Thanks,

Zahid

Twitter: @zDataGuy

January 28th, 2016 09:00

Hello Zahid,

* The vm is un-protected and therefore replication job and CG are removed.

* Replication is re-protected again.

- The system found the VM automatically which runs on DR site and replicated before.

But, unfortunatelly the problem is not solved. I got an the same error again.

So what's the next step?

153 Posts

January 28th, 2016 09:00

Thanks. I am not clear on how the system found the VM automatically, are you referring to the Copy VM?

Contact me directly, we can jump on a webex.

Zahid

Twitter: @zDataGuy

January 31st, 2016 23:00

Hello Zahid,

If you available, we can jump on a webex ! Looking forward your reply,

153 Posts

February 1st, 2016 00:00

Sorry I am not available right now. Was the env reinstalled at one point? I am suspecting that the splitters see two Cluster UIDs. If you run the command get_internal_cluster_name and compare it to the content of the Current_env file under /scratch/logs on the ESXi node, send me those and if indeed there is a stale cluster UID we can run a command hijack_cluster to cleanup the old UID.

Sent from my iPhone

February 1st, 2016 02:00

Hello Zahid,

We've checked UID's and cluster name. But the problem is still contiuning. When you be available, we can set a date for webex?

February 17th, 2016 22:00

This problem resolved

This is the rule as the user did indeed copy the VM, so they must select "I copied it".

When the user chooses "I moved it", VMware does not change the VM's ID, so we ends up with 2 VMs having the same VM-ID, which is not supported by RP.

When choosing "I copied it", (which is exactly what the user did - copied the VM), VMWare changes the VM ID and everything works fine.

The fix from our end will be to block the creation of a CG with duplicate VM-IDs.

  1. 1.    Issue (impact on customer): Error: The same volume has been added at more than one RPA cluster
  2. 2.    Splitter type(s): ESX
  3. 3.    Impact on RP: Settings for the CG cannot be changed (can't test a copy, etc)
  4. 4.    Root cause: In some manner, a VM is copied to the remote site. Since the VM was copied, the VMDK files report the same GUID for the VM. This is not supported.
  5. 5.    Symptoms found in the logs:

675 Posts

February 18th, 2016 00:00

Yes, this is a known issue. The WA is to use "I copied it" which will change VM UUID. In future versions, RP4VM will block the creation of CG with duplicate VM UUIDs.

Hope that helps,

Idan Kentor

RecoverPoint Corporate Systems Engineering

Email: idan.kentor@emc.com

No Events found!

Top