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
Is your feature request related to a problem? Please describe.
Once datahub is deployed, regardless of method (docker-compose, K8s, bare-bones, etc...) there is no way to obtain metrics & alerts related to the health of the deployed components.
I.e: There is no endpoint I can poll to retrieve the health of GMS. Should it fail, I have no way to automate alerts to notify me.
Describe the solution you'd like
Given that DataHub is a mostly java project it should be relatively easy to expose JMX metrics such that they can then be consumed by a Prometheus instance or similar.
Ideally there should exist operational metrics for each component of Datahub with associated documentation on how to use it.
The text was updated successfully, but these errors were encountered:
Agree for need of exposing JMX/Prometheus(which is widely used nowadays) metrics. For MCE and MAE, it is pretty-much out-of-box using Spring actuator APIs. For GMS also, can also be done using JMX exporter.
In case you agree it to be relevant for datahub, happy to contribute to this feature.
I think it is very relevant but I'm not a maintainer or contributor to the overall project. I simply created this issue as a request from a slack thread in the official channel.
Is your feature request related to a problem? Please describe.
Once datahub is deployed, regardless of method (docker-compose, K8s, bare-bones, etc...) there is no way to obtain metrics & alerts related to the health of the deployed components.
I.e: There is no endpoint I can poll to retrieve the health of GMS. Should it fail, I have no way to automate alerts to notify me.
Describe the solution you'd like
Given that DataHub is a mostly java project it should be relatively easy to expose JMX metrics such that they can then be consumed by a Prometheus instance or similar.
Ideally there should exist operational metrics for each component of Datahub with associated documentation on how to use it.
The text was updated successfully, but these errors were encountered: