Dear @fdiblen,
Thank you for that architecture design.
I have a few questions and comments. Please note that I am really not an expert, so some of them may seem naive and it's also based on my understanding and vision.
-
from my understanding of your design, the user is requesting for their repositories to be added/tracked by the dashboard. is that correct?
-
one could also have a dashboard monitoring a predefined catalogue of software repositories (I think that's what the proposal intended). For example, in ESCAPE, our catalogue in the OSSR and we would have a dashboard to aggregate the indicators from T3.2 into global metrics. Then users do not really add their software to the dashboard but to the community catalogue. (If they want to measure their specific software quality, they should use T3.2 pipelines in their own CI.)
-
that would remove the need for A&A and rate limiter. One cluster would be responsible for installing and running the dashboard on their catalogue and on their own servers.
-
a first dashboard prototype could even be a page generated on a list of software repositories (so nothing dynamic and without time evolution)
Dear @fdiblen,
Thank you for that architecture design.
I have a few questions and comments. Please note that I am really not an expert, so some of them may seem naive and it's also based on my understanding and vision.
from my understanding of your design, the user is requesting for their repositories to be added/tracked by the dashboard. is that correct?
one could also have a dashboard monitoring a predefined catalogue of software repositories (I think that's what the proposal intended). For example, in ESCAPE, our catalogue in the OSSR and we would have a dashboard to aggregate the indicators from T3.2 into global metrics. Then users do not really add their software to the dashboard but to the community catalogue. (If they want to measure their specific software quality, they should use T3.2 pipelines in their own CI.)
that would remove the need for A&A and rate limiter. One cluster would be responsible for installing and running the dashboard on their catalogue and on their own servers.
a first dashboard prototype could even be a page generated on a list of software repositories (so nothing dynamic and without time evolution)