Wondering if I can get some guidance here. It's not really a break / fix issue so I wasn't sure if I should open a SR (I can if you ask me to)
We use VMware's SRM / Live Recovery for DR. If I trigger a failover of a number of VMs to my DR site, configure re-protection from DR -> Prod and then fail the VMs back to prod, all the backups break.
They are still seen in the jobs but show 0 bytes when I refresh the list. I am assuming this is happening because once the VMs come back from DR as new VMs, they get a new MoREF ID. This means I need to go into each job, remove the 0 byte VMs, add them back and basically have it run full backups on those VMs.
Then I end up with 2 copies of the VMs in each job on my Data Domains, it affects retention, replication and I need to manually delete the "original" VMs from the job once I have sufficient restore points on the "new" VMs.
Is there any way around this (or anything planned to address this)? It's really a mess when we're dealing with hundreds and hundreds of VMs and I need to do full failover twice a year for my DR policies.
It has gotten to the point where this is very annoying and surely I cannot be the only one in this situation? A few hours of DR testing turns into days of work for me to "put everything back as it was"
What can we do ?
-
pmichelli
- Expert
- Posts: 116
- Liked: 31 times
- Joined: Mar 16, 2023 5:47 pm
- Contact:
-
Sturniolo
- Veeam Software
- Posts: 81
- Liked: 52 times
- Joined: Feb 19, 2019 3:08 pm
- Full Name: Andy Sturniolo
- Contact:
Re: MoREF ID + SRM/Live Recovery Failover
Hey @pmichelli
When a VM gets a new MoRefID it will break the backups... however in VBR 12.3 we have a MoRefID migration tool. Check out more info here... https://community.veeam.com/blogs-and-p ... -12-2-8714 The link also has links to the helpcenter regarding the feature. If you have any questions, please let me know.
When a VM gets a new MoRefID it will break the backups... however in VBR 12.3 we have a MoRefID migration tool. Check out more info here... https://community.veeam.com/blogs-and-p ... -12-2-8714 The link also has links to the helpcenter regarding the feature. If you have any questions, please let me know.
-
pmichelli
- Expert
- Posts: 116
- Liked: 31 times
- Joined: Mar 16, 2023 5:47 pm
- Contact:
Re: MoREF ID + SRM/Live Recovery Failover
Thank you for the super quick reply. I will take a look at what you posted. Happy new year !
-
pmichelli
- Expert
- Posts: 116
- Liked: 31 times
- Joined: Mar 16, 2023 5:47 pm
- Contact:
Re: MoREF ID + SRM/Live Recovery Failover
Hi @Sturniolo
I read the docs, I'm not sure how to apply this if I am using SRM for DR, since it would not be all the VMs that get failed over and back, plus the original vCenter will be the same as it was.
Example : vcenter1 holds VMs called test1 , test2 and test3
They are replicated to vcenter2 using SRM
If I trigger the failover in SRM, these VMs are now living on vcenter2
I setup re-protection and fail the VMs back to vcenter1
Veeam will now see these as 0 byte VMs in the respective jobs. So normally I have to go in, re-add the VMs and delete the 0 byte ones but this triggers a new full backup of these VMs on the 1st job run, and then leaves 2 copies of the data in the repository which I then need to clean up manually once retention catches up and I have sufficient restore points on the "new" VMs. Do I have any options here? That MoRef tools seems to be for your entire stack assuming you've built a new vCenter.
I read the docs, I'm not sure how to apply this if I am using SRM for DR, since it would not be all the VMs that get failed over and back, plus the original vCenter will be the same as it was.
Example : vcenter1 holds VMs called test1 , test2 and test3
They are replicated to vcenter2 using SRM
If I trigger the failover in SRM, these VMs are now living on vcenter2
I setup re-protection and fail the VMs back to vcenter1
Veeam will now see these as 0 byte VMs in the respective jobs. So normally I have to go in, re-add the VMs and delete the 0 byte ones but this triggers a new full backup of these VMs on the 1st job run, and then leaves 2 copies of the data in the repository which I then need to clean up manually once retention catches up and I have sufficient restore points on the "new" VMs. Do I have any options here? That MoRef tools seems to be for your entire stack assuming you've built a new vCenter.