Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

One shot revert CR/DR #7557

Open
olivierlambert opened this issue Apr 15, 2024 · 1 comment
Open

One shot revert CR/DR #7557

olivierlambert opened this issue Apr 15, 2024 · 1 comment

Comments

@olivierlambert
Copy link
Member

Is your feature request related to a problem? Please describe.
Let's imagine your main site is unreachable for an unknown amount of time. You will start your DR/CR VMs on your recovery site. But then, after a bit, the main site is back online. Since your replicated VMs changed, you can't simply shut them down and boot the main site: you'll lose everything written since the replication site was started.

Right now, you can live migrate back the VM (or warm migrate) but it won't erase the source VMs. More than taking time to migrate the data back, the chore needed afterward is huge. For example, if you had ACLs, self-service, backup jobs or anything else on the original VMs (now useless), you have to reconfigure everything.

Describe the solution you'd like
The idea is to overwrite all the original VMs with the content from the replicated VMs. This way, you can go back on your feet very quickly. To do that, a button "One shot Revert" that will:

  1. check that replicated and source VMs are shutdown (or they will be shutdown as the process starts)
  2. check that we can find the source VMs for all the replicated VM
  3. check that we can match all the source VM disks
  4. provide a big warning that all the source VMs content will be overwritten by the current state of replicated VMs
  5. inject the data from all the replicated disks to the source disks
  6. optionally, automatically start the source VMs when it's done

This way, your existing CR/DR job can just continue as usual, as well as permissions, or whatever custom stuff you had on your original VMs.

Note: we only want to transfer the data, nothing else from the replicated VMs (no metadata sync: name label etc.). Because we want to keep the original metadata VMs as is.

Describe alternatives you've considered
Manual migration to the original site, which could be really cumbersome.

Additional context
This is a one shot action, with a lot of warning before doing it. But in the end, it's a lot of time saved in case of getting back on the main site. If we lost this original/main site for good, we don't need that, we'll use live or warm migration.

@olivierlambert
Copy link
Member Author

@marcungeschikts to discuss for potential next releases

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

1 participant