Best Practices for Backup and Restore

Hi,
this may be a very basic question, but I did not find any practical information on this:

  • We build an application on Camunda 8 and want to launch it soon
  • Therefore we considered the backup and restore solution.
  • While we understand the (sophisticated :-)) hot backup approach, we are unsure about how to work in a situation where data gets lost or corrupted:

-Assume we do a nightly (? is that the right frequency) hot backup.

  • During the day (say afternoon) the data gets corrupted and we need to fall back to the backup.

What should we best do with the activities and messages that happened during the day? They modified the process instances, but how do we get these applied to the restored Camunda?

Do the process instances need to be fixed manually? Or can we use (or implement) some kind of “replay” of all the messages and worker activities in the mean time? (Probably not)
Should we keep an independent history of in/outgoing requests for that purpose on application level=
Do idempotent workers help in that context?

Or could we minimize the problem by constantly taking backups (every 1,5, 10, 60 or whatever minutes?)
We can only speculate what the best way would be. Also, what it means to manually adjust all process instances.
Are there practical experiences with this?

Thanks for any ideas,
Andreas

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.