Skip to main content
Version: 8.5 / 3.13.0

Camunda engine data reimport

Camunda 7 only

There are cases where you might want to remove all Camunda 7 engine data from Optimize which has been imported from connected Camunda engines but don't want to lose Optimize entities such as collections, reports, or dashboards you created.


Triggering a reimport causes the current data imported from the engine to be deleted and a new import cycle to be started. That also means that data which has already been removed from the engine (e.g. using the history cleanup feature) is irreversibly lost.

When triggering a reimport, all existing event-based processes get unpublished and reset to the mapped state. This is due to the fact that event-based processes may include Camunda engine data, yet the reimport does not take into account which sources event-based processes are actually based on and as such clears the data for all of them.

You then have to manually publish event-based processes after you have restarted Optimize.

To reimport engine data, perform the following steps:

  1. Stop Optimize, but keep Elasticsearch running (hint: to only start Elasticsearch without Optimize, you can use or elasticsearch-startup.bat scripts).

  2. From the Optimize installation root run ./reimport/ on Linux or reimport/reimport.bat on Windows and wait for it to finish

    • In Docker environments, you can override the command the container executes on start to call the reimport script, e.g. in docker-compose this could look like the following:
    version: '2.4'

    command: ["./reimport/"]
  3. Start Optimize again. Optimize will now import all the engine data from scratch.

  4. If you made use of event-based processes you will have to manually publish them again.