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
We have support for metrics but don't utilize them well yet. We should think of and add reasonable metrics that we can use to monitor the health and status of the supervisor.
The text was updated successfully, but these errors were encountered:
Regarding metrics, that could be a good team exercise of identifying metric locations. It promotes an understanding of the service behavior. At first glance,
every event can support a counter metric at least, with more metrics specific to the nature of the event.
API calls over the wire probably already have metrics for free, which just need to be unified with events in Grafana
The op-node has generic events-systems metrics utils which we can reuse
And I think it may be useful to just create a kind of like a status package like we have with op-node. And then just wait for all kinds of events, and use the event contents to update the metrics. That avoids having to install metrics into every individual module.
We have support for metrics but don't utilize them well yet. We should think of and add reasonable metrics that we can use to monitor the health and status of the supervisor.
The text was updated successfully, but these errors were encountered: