Question about the official k6 dashboard for prometheus-backed metrics

Hello, I have been testing this official dashboard for prometheus-backed metrics with my test cases. I have some questions about it.

  1. There is one metric under “performance overview” to show p95 response time. However it’s an aggregated view across all requests. Is there a way to break this metric off to show p95 response time per API?
  2. A follow-up question. Looks like the “Requests by URL” table under HTTP row was designed to achieve my first question. However it doesn’t present data well when the API contains dynamic element.
    Example: API:${request_id}. if every each request has a unique request_id, it’s gonna list each request on the table without aggregation, which is kinda useless, like snapshot below… How do I modify to group this URL into one?

Hi @alex

Thanks for testing the official prometheus-backed dashboard :grinning:

We believe you could achieve the second point using url grouping. It’s a common use-case. If you are still not seeing what you expect, please share a (sanitized) script and we’ll be able best advice :woman_technologist:t4:.

I hope this helps.