Investigate unexpected queue flushing with cache and queued tracking on Redis #20208
Labels
Potential Bug
Something that might be a bug, but needs validation and confirmation it can be reproduced.
Milestone
We have a reports of unexpected queue flushing when using a single Redis instance for both the Matomo cache and for the Queued tracking feature in a multi-server environment. The cache and queue were in difference databases on the same Redis instance.
Moving the Matomo cache to a separate Redis instance resolved this behavior for the customer.
Ref L3-386
This issue is a reminder that we should test this scenario to confirm that the Matomo cache and Queued tracking can be run on the same Redis instance and if not then this should be documented or fixed.
Expected Behavior
The Matomo cache and queued tracking should be able to co-exist on the same Redis instance but in separate databases.
Current Behavior
It's reported that this doesn't work, we need to confirm this.
Steps to Reproduce (for Bugs)
Setup Matomo cache on Redis
Setup Queued Tracking on Redis (same instance, separate database)
Push some queued tracking requests into Redis
Attempt a Matomo cache flush and see if it clears the tracking queue
Try other things that may trigger a flush?
The text was updated successfully, but these errors were encountered: