Hello all

I'm trying to figure out the gnocchi workflow to get metric ready and configured properly, so I installed Devstack with gnocchi set.

My backend configuration is working by default and I did the following using gnocchiclient:

(Pre-requirement) Create and instance and get its Id and also get the project_id where instance runs --- run gnocchi-upgrade just after DevStack is ready

1. Created a resouce-type "instance"
2. Created a resource with its needed associations: "metric (new one named uptime), archive-policy (the default low), attributes - resourceId (instanceId) and project (projectId) and the type (the new instance)"
3. Reviewed the index DB and all apparently is fine.
4. Start the workers (gnocchi-api and gnocchi-metricd)
5. Reviewed logs (no errors inside)
6. Bad news I don't have timeseries for my metric

QUESTIONS

A. How does gnocchi to match the UPTIME from my instance? (this is my first missing thing - I do not how to set that)

B. How to configure the UNITS example "sec." for my metric? (from gnocchi client there is not an option or argument)

C. For other cases of use "Instance's CPU temperature, Network BW usage, Storage I/O, etc" how is it supposed to approach that configuration from the metric point of view and OpenStack IaaS?

Please help me to understand these empties of knowledge to run in my path.