Closed Alexander-Dubrawski closed 3 years ago
Didn't see that issue last time I tested the cockpit (around 1-2 weeks ago). I'll check it later today ...
I investigated the issue. I found out that the problem doesn't exist if we:
We only get this strange delay if we run the cockpit on a VM. The backend seems to be responsive at all time and if I ask for the throughput directly from the backend I get the current numbers without a delay, but in the frontend, they are shown with a delay. So I think the problem lies in the frontend, but I can't really tell where.
This Problem seems to not exist anymore.
Please describe the bug:
If we start a workload in the frontend the changes (throughput, latency,...) are visible just after around 20 to 25 seconds in the frontend. At the moment I am not sure if the problem lies in the frontend or backend (or even docker). I think we should further investigate the problem.
Please describe steps to reproduce:
What is the expected behavior? Please describe:
The change in latency or throughput should be visible at least after 3 to 4 second (three seconds is the offset for the timestamp in the influx)
Additional context:
I tested it with a VPN connection and the cockpit was running on a VM. Maybe this had an influence as well on the performance.