A few weeks ago I run into the problem where I have performed recovery of some virtual machines. After performing recovery deleted the VMs from the recovery site. But it wasn’t deleted from VR database
– Attempt to reconfigure replication from the primary site failed with the following error ”
“Unable to configure replication for the virtual machine because the group cannot be created. Another virtual machine has same instance UUID already exists on protection site.
I did some workaround and didn’t find the solution. So after logged a complaint in Vmware support , here is the solution :
Setep-1 Took a backup of VR databases on both the sites. Following are the steps:
cd /opt/vmware/vpostgres/9.3/bin
./psql -U vrmsdb
Setep-2 – Took a snapshot of the appliances.
Step-3– Run this SQL Query on the primary site appliance. Zero rows retrieved.
SELECT a.group_movalue AS “PRIMARY GID”, b.name AS “virtual_machine Name”, ‘https://’ || c.address || ‘:8043/mob/?moid=’ || a.group_movalue || ‘&vmodl=1’ AS “Link to Paste” FROM primaryvirtualmachineentity a, virtualmachineentity b, localserverentity c WHERE a.movalue = b.movalue ORDER BY name;
Step-4– From SSH session to DR site replication appliance, cd to navigate to this directory:
cd /opt/vmware/vpostgres/9.3/bin
Step-5 Run this command to access the embedded database:
./psql -U vrmsdb
Step-5 Run this query on the remote VRMS to get the GID of the orphaned job and generate a URL to review it using the MOB:
SELECT a.group_movalue AS “SECONDARY GID”, b.name AS “VM Name”, ‘https://’ || c.address || ‘:8043/mob/?moid=’ || a.group_movalue || ‘&vmodl=1’ AS “Link to Paste” FROM secondaryvirtualmachineentity a, virtualmachineentity b, localserverentity c WHERE a.movalue = b.movalue ORDER BY name;
Step-6– Copy the MOB URL you acquired from the database and paste in to your browser.
- To find the exact credentials required, open this URL:
- https://VC_IP/mob/?moid=SessionManager&doPath=currentSession
Step-7– Select the destroy method and click Invoke Method to remove the VRMS database entry related to the problematic replication job.
The orphaned job no longer exists in the VR UI and you can configure replication for the affected virtual machine as normal.