What is the next step

[faressoft]

What is the next step to visualize these data with Graphana ?

[disjunction]

The next step is to set up prometheus and point it to your service: https://prometheus.io/docs/prometheus/latest/getting_started/

You'll be able to see the metrics in prometheus itself

After you make sure the metrics are properly read by prometheus you can connect Grafana to prometheus. That's it https://prometheus.io/docs/visualization/grafana/

[mikepuglisi]

Is there a prebuilt grafana dashboard that lends itself to the data express-prom-bundle outputs?

I'm new to all prometheus and grafana, but I have the express-prom-bundle /metrics exposed and available to prometheus. The next step for me is to present it in grafana. I already have kubernetes cluster data exposed using prebuilt dashboards, but I don't know how to create a custom dashboard yet, and I'm not sure I even want to if there are prebuilt dashboards that convert data such as the following to a histogram in grafana.

# HELP http_request_duration_seconds duration histogram of http responses labeled with: status_code, method, path, year
# TYPE http_request_duration_seconds histogram
http_request_duration_seconds_bucket{le="0.1",status_code="200",method="GET",path="/",year="2018"} 0
http_request_duration_seconds_bucket{le="0.4",status_code="200",method="GET",path="/",year="2018"} 0

http_request_duration_seconds_bucket{le="0.7",status_code="200",method="GET",path="/",year="2018"} 0
http_request_duration_seconds_bucket{le="+Inf",status_code="200",method="GET",path="/",year="2018"} 2

[disjunction]

@mikepuglisi I don't have any predefined dashboards, but I can share a couple of queries I use:

1. request count per 5m. Shows a chart per http method and path. Here I use sum because I have multiple instances of the same app, and increase shows the request in given time range instead of just continuously growing chart

• Query: sum(increase(http_request_duration_seconds_count[5m])) by (method, path)
• Legend format: {{ method }} {{ path }}

2. Errors chart. Shows a chart per HTTP error code, displaying all 5xx and only 400 error. (I ignore other 40x errors, as those are often generated by bots and alike).

• Query: sum(increase(http_request_duration_seconds_count{status_code=~"5.*|400"}[1h])) by (status_code)
• Legend format: {{ status_code }}

Hope it helps!

[poteirard]

It would be very beneficial for the users of this libraries to have a Grafana dashboard just like other libs like swagger-stats. I would love to help on this but unfortunately I'm new with this...

[marwenbk]

did anyone make a Dashboard?