Improve disaster recovery docs

[joaofnfernandes]

@joaofnfernandes commented

As long as cluster has quorum:

• Use remove command to remove node that is down from cluster
• Manually uninstall DTR if necessary from that node, and make sure everything is OK
• Use join command

If there has been loss of quorum you'll need to use the restore command. It creates a new DTR from an existing backup.

Also improve backup docs to tell how users can test if a DTR backup has been successful

• docker/dtr backup
• tar -tf to list contents of backup and spot for corrupt backup
• Create new UCP cluster (can be a single node)
• Restore UCP from a backup to have user data
• Restore DTR

You might have problems if you restore DTR in different domain from where you created the backup.

---

@pdevine commented on Mon Oct 17 2016

We should really be doing this before 2.2.0, but it shouldn't hold up 2.1 GA

[joaofnfernandes]

This has been mostly addressed by now.

[docker-robott]

Closed issues are locked after 30 days of inactivity.
This helps our team focus on active issues.

If you have found a problem that seems similar to this, please open a new issue.

/lifecycle locked