You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Nov 28, 2022. It is now read-only.
Che version: N/A IDE extension version: N/A IDE version: N/A Kubernetes cluster: N/a
Description of the enhancement:
This is step 1 in a multi step process to improve the performance dashboard to automatically pick up and display metrics broadcast on a /metrics endpoint. This work will cover appmetrics-dash and javametrics-dash gaining code to poll a /metrics endpoint on localhost every 10 seconds and to display the data retrieved in as simple a form as possible, in order to prove the concept.
Proposed solution:
Bare bones spec:
Dashboard will poll http(s)://localhost/metrics every 10 seconds
Dashboard shows data retrieved - block of text or table
What's the main difference between appmetrics-dash and javametrics-dash today? I think one of the goals of this would be to have a single dash that can poll any container's /metrics, possibly with some detection to graph specifics for different languages or frameworks based on what's in the feed.
The main difference is in the type of metrics exposed that are language-specific - Node, for example, has the Event Loop Latency that is useful for asymmetric programs to monitor to see when their next portion of code will be executed, something that Java doesn't have.
Most of the dashboard's functional backends come from the same common codestream https://github.com/runtimetools/graphmetrics so it's absolutely feasible to aim for more commonality. As stated in the description though, this is step 1 of a larger item that will span several issues, so I'm more interested in keeping this particular issue quite tightly defined. Ordinarily I would have raised an Epic to tie the multiple forthcoming issues together, but I couldn't see a way of doing that :-)
Codewind version: vNext
OS: All
Che version: N/A
IDE extension version: N/A
IDE version: N/A
Kubernetes cluster: N/a
Description of the enhancement:
This is step 1 in a multi step process to improve the performance dashboard to automatically pick up and display metrics broadcast on a /metrics endpoint. This work will cover
appmetrics-dash
andjavametrics-dash
gaining code to poll a /metrics endpoint on localhost every 10 seconds and to display the data retrieved in as simple a form as possible, in order to prove the concept.Proposed solution:
Bare bones spec:
@mattcolegate to cover Javametrics, @rwalle61 covers Appmetrics.
Use Prometheus-enabled application as our test app.
cc @tobespc
The text was updated successfully, but these errors were encountered: