Include full image URLs in deploy_timestamp{} metric

[etsauer]

Feature description

currently our deploy_timestamp metric looks like this:

deploy_timestamp{app="kenwilli-basic-spring-boot", endpoint="http", exported_namespace="kenwilli-basic-spring-boot-prod", image_sha="sha256:46332b417660900361f7185830a4eb6d5ddc7e3002944ba26ed260e83f415197", instance="10.129.1.254:8080", job="deploytime-exporter", namespace="app-name-consistency", pod="deploytime-exporter-1-66w2p", service="deploytime-exporter"}

In order to have improved data integrity, we should collect the full image URL so that a user can trace this metric back to the image registry that it was pulled from.. I propose we include an additional label that includes the full image sha like so:

deploy_timestamp{app="kenwilli-basic-spring-boot", endpoint="http", exported_namespace="kenwilli-basic-spring-boot-prod", image_sha="sha256:46332b417660900361f7185830a4eb6d5ddc7e3002944ba26ed260e83f415197", image_url="quay.io/myapp/basic-spring-boot@sha256:46332b417660900361f7185830a4eb6d5ddc7e3002944ba26ed260e83f415197", instance="10.129.1.254:8080", job="deploytime-exporter", namespace="app-name-consistency", pod="deploytime-exporter-1-66w2p", service="deploytime-exporter"}

How this will help you

This will provide us with the data needed to provide a user experience that allows users to trace the data that we show them back to the raw metrics, helping them to independently validate the data we collect, learn how we calculate metrics, and build trust in the tool.

It will also help to expose any bugs we might have in our collection or calculation

Feature priority

⌛ It will greatly improve my experience using Pelorus

Code of Conduct

• I agree to follow Pelorus's Code of Conduct