Currently, the import scheduler iterates through its existing observers to notify status updates. However, it doesn't synchronize on the synchronized observer list when doing this iteration. It might be that new subscriptions/unsubscriptions that occur at the same time as this iteration could cause a race condition that blocks importing, possibly causing the issue reported in https://jira.camunda.com/browse/SUPPORT-8453.

Given:

I am running Optimize docker and I have open status connections

When:

The status of connections change

Then:

The engine import is unaffected and continues as expected

ATs:

• We iterate through the observers in accordance with the synchronized list API