Search code examples
backupdatabase-backupsrethinkdb

RethinkDB backup data

I read article about backing up data, but some issues is not clear for me:

  1. What happens with data, that will be changed after backup process was started?
  2. Does backup operation work only on current machine? Or will it collect data from all shards in cluster? If only on current, should I start backup process on all servers?
  3. Is it slow operation so I should forbid all operation to db while backup in progress?
Solution

    1. If a row changes while the backup is going on, the new value may or may not be in the backup. This is generally OK because RethinkDB only offers single-row atomicity anyway, but if you have a workload where that isn't OK then your other options are to use a filesystem that lets you snapshot the data on disk, or to add a new server to your cluster and set it as a replica of the table you want to back up.

    2. It collects data from all shards.

    3. It can take a very long time.