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
As proposed by @nbren12 in #1 (comment), we will be running these tests as a kubernetes cron job. The clusters will need some way to publish their results, likely just a static http endpoint.
How should this work? I see two basic possibilities:
Each testing cluster publishes its results to a public bucket in its own cloud storage. With this, write credentials can be packaged into the kubernetes environment.
All clusters publish to a central location. This would require some sort of common credentials.
The former certainly seems easier.
We also need to decide what we want to log. It will be important to capture detailed logs from all parts of the stack, as well as performance reports (timing, etc).
The text was updated successfully, but these errors were encountered:
I agree (1) is easier and allows a more federated model. It would be nice to setup some kind of table of badges, we could bake this into our test script, but maybe there is an easier way.
As proposed by @nbren12 in #1 (comment), we will be running these tests as a kubernetes cron job. The clusters will need some way to publish their results, likely just a static http endpoint.
How should this work? I see two basic possibilities:
The former certainly seems easier.
We also need to decide what we want to log. It will be important to capture detailed logs from all parts of the stack, as well as performance reports (timing, etc).
The text was updated successfully, but these errors were encountered: