Details
-
Improvement
-
Resolution: Done
-
Major
-
None
-
None
Description
Feedback from David Haikney:
getting it built and up and running was straightforward. A few places ^ where docs could be tightened or clarified but nothing major
Need to make it as easy as possible for users to say "here's my clusters+creds, take me to the dashboard". Doing that for prometheus and health checker for each cluster is a bit messy.
Dashboard still needs work. We need to figure out how to preset the feedback of "something's red...here's what's wrong...here's what to do about it" - where the what to do could be as simple as a URL.
Document ssh-agent setup for git build, possibly remove once released.
There is a file you can update to add the Prometheus endpoint and health check has it's REST API on the usual port
https://github.com/couchbaselabs/observability/tree/main/examples%2Fnative%2Fdynamic%2Fprometheus%2Fcouchbase-servers
for health check you currently use the REST API
There is an example in the other dynamic directory: https://github.com/couchbaselabs/observability/tree/main/examples%2Fnative%2Fdynamic%2Fhealthcheck
You can just do it from your OS though, the usual port is mapped and I default the checker credentials to admin and password unless you override them: https://www.github.com/couchbaselabs/observability/tree/main/microlith%2Fentrypoints%2Fcouchbase_healthcheck.sh
I just use curl in the example to do it from within the container but the port is exposed so it'll work outside too: https://www.github.com/couchbaselabs/observability/tree/main/examples%2Fnative%2Fdynamic%2Fhealthcheck%2Ftargets.sh
So just run your usual command to configure cluster manager on localhost 7196
Look to see if we can provide a simple web form with Nginx that then generates everything we need to configure things.