Update 19:32 CET: Most import files from the queue were processed successfully after adding additional processing capacity. We're continuing to monitor the situation.
We're seeing an uptick in amount of imports produced by our customers. This causes imports to stay in "in progress" state for longer than usual amount of time.
Import functionality works properly and files are being processed correctly, however, this happens slower than usual.
Affected period of time: no downtime
Total: no downtime
Why did it happen?
We're seeing performance issues in the amount of time it takes for the import consumer to process individual rows of import files. This causes the global import queue to stall while bigger import files are being processed.
What are we doing to avoid such issues?
1. We've added an additional server to help processing smaller imports while bigger imports are processed on our main server
2. We will investigate ways to improve individual per-row performance of import consumers to make sure overall performance of import subsystem is acceptable.