I use the spring-boot Camunda engine and we already have deployed multiple processes working perfectly.
The problem is that we didn’t defined a TTL on the processes in the modeler so our history tables has grown huge !
So I tried to configure a process engine time to live, to do soI tried to define some properties to cleanup the history :
But our history tables are still huge …
Is there a problem with this configuration ?
Also, is there a way to check if the cleanup was trigerred ?
Thanks for your help
The clean-up is recorded in the metrics:
You can query the metrics with this endpoint: Get Metrics in Interval | docs.camunda.org
Hope this helps, Ingo
Okay, thanks for the input !
I see that
history-cleanup-removed-process-instances deleted 500 processes yesterday (which is the max batch size according to the documentation).
At this rate, it will take months to clean the history
Does the cleanup batch only run once a day ? I understood it is ran whenever it is in the
Also, I have another question : some of my processes in the history have an ‘ACTIVE’ status, so they don’t have an endTime. Is there a way to flush them depending on their startDate ?
Is there any recommended way to define / archive historical data to Data Warehouse solution before it gets deleted from live DB?
I’m sorry @Camunda_newbir but I think this is a different problem. Could you please open a dedicated topic for this ?