The result shows all http req failed but Prometheus received data

[Kirchen99]

Prometheus Version: v2.37.0 xk6-client-prometheus-remote Version: build from resource according to the README.md k6 Version: k6 v0.40.0 ((devel), go1.19.1, darwin/amd64)

In this example the status code is compared with 200, but Prometheus give 204 back.

https://github.com/grafana/xk6-client-prometheus-remote/blob/6c1e7b99d2123ef2a87b39dcadb033dd1aabf7f8/examples/basic.js#L24

What did you do?

try the example from this repo with a local prometheus: https://github.com/grafana/xk6-client-prometheus-remote/blob/main/examples/basic.js

What did you expect?

No error

What did you see instead?

Error messages such as:

http_req_failed: 100% and

✗ is status 200
      ↳  0% — ✓ 0 / ✗ 10

but the Prometheus received the data.

default ✓ [======================================] 1 VUs  10s

     ✗ is status 200
      ↳  0% — ✓ 0 / ✗ 10

     checks.....................: 0.00%   ✓ 0        ✗ 10 
     data_received..............: 640 B   62 B/s
     data_sent..................: 2.8 kB  274 B/s
     http_req_blocked...........: avg=155.09µs min=4µs     med=6.5µs   max=1.49ms  p(90)=156.09µs p(95)=827.04µs
     http_req_connecting........: avg=15.4µs   min=0s      med=0s      max=154µs   p(90)=15.39µs  p(95)=84.69µs 
     http_req_duration..........: avg=29.69ms  min=16.42ms med=24.23ms max=81.08ms p(90)=37.05ms  p(95)=59.06ms 
     http_req_failed............: 100.00% ✓ 10       ✗ 0  
     http_req_receiving.........: avg=61.69µs  min=38µs    med=58µs    max=127µs   p(90)=77.49µs  p(95)=102.24µs
     http_req_sending...........: avg=85.9µs   min=27µs    med=46.5µs  max=444µs   p(90)=127.19µs p(95)=285.59µs
     http_req_tls_handshaking...: avg=0s       min=0s      med=0s      max=0s      p(90)=0s       p(95)=0s      
     http_req_waiting...........: avg=29.54ms  min=16.35ms med=24.13ms max=80.51ms p(90)=36.9ms   p(95)=58.7ms  
     http_reqs..................: 10      0.969329/s
     iteration_duration.........: avg=1.03s    min=1.01s   med=1.02s   max=1.08s   p(90)=1.03s    p(95)=1.06s   
     iterations.................: 10      0.969329/s
     vus........................: 1       min=1      max=1
     vus_max....................: 1       min=1      max=1

{"remote_ip":"127.0.0.1","remote_port":9090,"url":"http://localhost:9090/api/v1/write","status":204,"status_text":"204 No Content","proto":"HTTP/1.1","headers":{"Date":"Mon, 12 Sep 2022 10:28:25 GMT"},"cookies":{},"body":"","timings":{"duration":22.487,"blocked":0.006,"looking_up":0,"connecting":0,"tls_handshaking":0,"sending":0.092,"waiting":22.327,"receiving":0.068},"tls_version":"","tls_cipher_suite":"","ocsp":{"produced_at":0,"this_update":0,"next_update":0,"revoked_at":0,"revocation_reason":"","status":""},"error":"","error_code":0,"request":{"method":"POST","url":"http://localhost:9090/api/v1/write","headers":{"Content-Encoding":["snappy"],"Content-Type":["application/x-protobuf"],"User-Agent":["k6-remote-write/0.0.2"],"X-Prometheus-Remote-Write-Version":["0.0.2"]},"body":"","cookies":{}}}

[mstoykov]

Hi @Kirchen99, thanks for reporting that. This seems like a simple example mistake or behaviour change.

The test chekcs for 200 but prometheus returns 204 as you notice. I have no idea if that is normal though and if it has changed ... recently.

cc @replay any idea on this?

[replay]

I believe to remember that it is normal for Prometheus to return 204 on successful ingestion. Instead of looking for 200 the check should look for a status code in the range [200, 300)