http_req_failed metric displays pass/fail counts reversed despite 0% failure rate

[keshavpdl]

Brief summary

Description:
When running a test with no failed HTTP requests (http_req_failed rate = 0.00%), the k6 CLI raw output and JSON results show the http_req_failed pass and fail counts reversed. Specifically, the failure rate is correctly reported as 0.00%, but the pass count is shown as 0 and the fail count as the total number of requests, which is misleading.

Example snippet from raw output:

http_req_failed................: 0.00%   ✓ 0        ✗ 21
http_reqs......................: 21

Impact:
This can cause confusion interpreting test results, as the failure rate and counts appear inconsistent.

Additional notes:
The failure rate metric itself is correct, so this appears to be a display/reporting issue for the http_req_failed metric pass/fail counts.

SS:

k6 version

k6 v1.6.1 (commit/2ac2bb560e, go1.25.6,)

OS

linux/amd64

Docker version and image (if applicable)

No response

Steps to reproduce the problem

1. Create a simple k6 script that makes a number of HTTP requests, all expected to succeed. For example:

import http from 'k6/http';
import { check } from 'k6'; 
export default function () {
 const res = http.get('https://test-api.k6.io');
check(res, {
'status is 200': (r) => r.status === 200,
});
}

2. Run the test with 20+ iterations (e.g., k6 run --iterations 21 script.js).
3. Observe the CLI or JSON output for the http_req_failed metric:
   http_req_failed................: 0.00% ✓ 0 ✗ 21
   Note that the failure rate is 0.00% but pass count is 0 and fail count equals total requests.

Expected behaviour

• Pass count (✓) should be equal to the total requests (21 in this case).
• Fail count (✗) should be zero since failure rate is 0.00%.

Actual behaviour

• Pass count is 0.
• Fail count equals the total number of requests.

[jonaslfleck]

In addition to this inversion of the http_req_failed metric itself, we have observed the same logic error affecting custom metrics. Custom metrics that use rate or count statistics are also being reported with inverted pass/fail counts in the summary, even though the underlying metric values and calculated rates are correct.

This indicates that the issue is not isolated to only the built-in http_req_failed metric, but impacts how pass/fail counts are computed and displayed for both built-in and custom metrics in the test summary output.

k6 version
k6 v1.6.1 (commit/2ac2bb560e, go1.25.6,)

OS
linux/amd64

Docker version and image (if applicable)
No response

Steps to reproduce the problem:

1. Create a simple k6 script that makes a number of HTTP requests, all expected to succeed. For example:

import http from 'k6/http';
import { check } from 'k6';
import { Rate } from 'k6/metrics';

const failed = new Rate('custom_metric_failed');

export default function () {
 const res = http.get('https://test-api.k6.io');

    const checkResult = check(res, {
    'status is 200': (r) => r.status === 200,
    });

    failed.add(!checkResult);
}

2. Run the test with 20+ iterations (e.g., k6 run --iterations 21 script.js).
3. Observe the CLI or JSON output for the http_req_failed metric:

custom_metric_failed...........: 0.00%   ✓ 0       ✗ 21

Note that the failure rate is 0.00% but pass count is 0 and fail count equals total requests.