Variable for elapsed duration since start of analysis run

[jessesuen]

A prometheus query might want to query by the total number of seconds since the start of the rollout, as opposed to a fixed, hard-wired duration like [5m]. This would enable us to perform queries that get totals since the start of the rollout, not just the last X minutes.

This might be important in queries such as success rates in low-traffic services, which may be heavily skewed in recent history by lack of activity.

[github-actions]

This issue is stale because it has been open 60 days with no activity.

[nebojsa-prodana]

Hello, we are migrating from our internal custom-built metric checks engine to analysis templates. We are missing a feature like this (or #405) that would allow us to know how long the pods of the new canary version have been up so that we can get the proper statistical confidence in the metric checks.

Is this still on the roadmap? Would you be willing to accept contributions?