Separate metrics summary for each request in default function

Grafana k6 OSS Support
2

The summary aggregates all of the results it shows. The only current way to get around that, before we implement Add explicit tracking and ignoring of metrics and sub-metrics · Issue #1321 · grafana/k6 · GitHub, is to define a threshold on a sub-metric. I gave an example for that in Ignore http calls made in Setup or Teardown in results? - #2 by nedyalko, but if I understand your use case correctly, it would be something like this:

import http from 'k6/http'

export let options = {
    vus: 5,
    duration: "10s",
    thresholds: {
        // Some dummy thresholds that are always going to pass. You don't even
        // need to have something here, I tried it and just this
        // `'http_req_duration{scenario:default}': []` is enough to trick
        // k6, but that's undefined behavior I can't promise we won't break
        // in the future...
        'http_req_duration{my_tag:create}': ['max>=0'],
        'http_req_duration{my_tag:read}': ['max>=0'],
        // You don't have to define custom tags, you can use any of the already defined
        // system tags: https://k6.io/docs/using-k6/tags-and-groups#system-tags
        'http_req_duration{method:POST}': ['max>=0'],
        'http_req_duration{url:https://httpbin.test.k6.io/anything}': ['max>=0'],
        'http_req_duration{status:200}': ['max>=0'],
    },
}

export default function () {
    http.post('https://httpbin.test.k6.io/delay/2', null, { tags: { my_tag: 'create' } });
    http.get('https://httpbin.test.k6.io/delay/1', { tags: { my_tag: 'read' } });

    http.post('https://httpbin.test.k6.io/anything', 'some body');
}

Alternatively, it might be better if you work with the raw metrics data directly by exporting it to JSON/CSV or sending it to InfluxDB or some other external service. You can see more details about that in Results output

1 Like