Graphs not matching up with script expectations

Grafana k6 OSS Support
,
1

We are trying to carry out soak-tests.
Implemented such scenario:

        executor: ramping-arrival-rate,
        startRate: 1,
        timeUnit: '1s',
        preAllocatedVUs: 10000,
        stages: [
          { target: 30000, duration: '30m' },
          { target: 30000, duration: '1h' },
          { target: 1, duration: '5m' }
        ]

In default function there are two requests with some checks.

Our setup is following:
k6 + InfluxDBv1 + Grafana
In Grafana we are using following dashboard:

We are expecting to see in dashboard “Requests per Second” numbers close to target (iterations) values (~30000 or ~60000 (as there are two requests)), but, instead we observing strange behaviour and numbers:

community.grafana.com-k6-influxdbv1-grafana-results-not-matching-up-with-script-expectations-grafana-graphs-with-strange-numbers
community.grafana.com-k6-influxdbv1-grafana-results-not-matching-up-with-script-expectations-grafana-graphs-with-strange-numbers2492×1112 174 KB

InfluxDB query used in first graph/panel “Requests per Second”:

SELECT sum("value") FROM "http_reqs" WHERE $timeFilter GROUP BY time(1s) fill(null)

InfluxDB query used in second graph/panel “Requests per Second”:

SELECT 
    count("value") * 1000 / $__interval_ms
FROM /^$Measurement$/ 
WHERE $timeFilter 
GROUP BY time($__interval) fill(null)

Same picture and results with following scenarios:

        executor: ramping-arrival-rate,
        startRate: 1,
        timeUnit: '1s',
        preAllocatedVUs: 1000,
        stages: [
          { target: 3000, duration: '30m' },
          { target: 3000, duration: '1h' },
          { target: 1, duration: '5m' }
        ]

          executor: ramping-arrival-rate,
          startRate: 1,
          timeUnit: '1s',
          preAllocatedVUs: 10000,
          stages: [
            { target: 30000, duration: '1h' }
          ]

(at least numbers (orders of magnitude) do not change much)

Why we are getting such results? Is it actually expected behaviour and numbers and we just missing something? Or something really wrong with our setup/script/expectations?

2

Welcome @zephyrzen to our community. Please share k6 console output to have more details to understand.

4 
	 checks..................................: 99.99%   ✓ 13388396    ✗ 62           
     custom_http_req_blocked_gauge...........: 0.003502 min=0         max=87.674653  
     custom_http_req_connecting_gauge........: 0        min=0         max=57.014291  
     custom_http_req_duration_gauge..........: 3.515758 min=1.357827  max=2529.667062
     custom_http_req_receiving_gauge.........: 0.048327 min=0.011842  max=50.004228  
     custom_http_req_sending_gauge...........: 0.017857 min=0         max=42.812183  
     custom_http_req_tls_handshaking_gauge...: 0        min=0         max=87.315612  
     custom_http_req_waiting_gauge...........: 3.449574 min=0.946926  max=2529.565179
     data_received...........................: 3.3 GB   592 kB/s
     data_sent...............................: 2.9 GB   521 kB/s
     dropped_iterations......................: 10157564 1800.978237/s
     group_duration..........................: avg=142.11ms min=2.2ms    med=13.76ms  max=2.53s   p(90)=486.71ms p(95)=564.52ms
     http_req_blocked........................: avg=27.54µs  min=0s       med=5.29µs   max=87.67ms p(90)=7.55µs   p(95)=9.93µs  
     http_req_connecting.....................: avg=1.34µs   min=0s       med=0s       max=57.01ms p(90)=0s       p(95)=0s      
     http_req_duration.......................: avg=138.9ms  min=1.35ms   med=11.03ms  max=2.52s   p(90)=481.88ms p(95)=559.72ms
       { expected_response:true }............: avg=138.91ms min=2.39ms   med=11.03ms  max=2.52s   p(90)=481.88ms p(95)=559.72ms
     http_req_failed.........................: 0.00%    ✓ 62          ✗ 3347068      
     http_req_receiving......................: avg=64.63µs  min=11.84µs  med=51.26µs  max=50ms    p(90)=76.44µs  p(95)=89.59µs 
     http_req_sending........................: avg=50.65µs  min=0s       med=23.06µs  max=42.81ms p(90)=40.52µs  p(95)=52.12µs 
     http_req_tls_handshaking................: avg=20.02µs  min=0s       med=0s       max=87.31ms p(90)=0s       p(95)=0s      
     http_req_waiting........................: avg=138.79ms min=946.92µs med=10.9ms   max=2.52s   p(90)=481.74ms p(95)=559.57ms
     http_reqs...............................: 3347130  593.460035/s
     iteration_duration......................: avg=284.32ms min=4.46µs   med=271.72ms max=2.54s   p(90)=577.88ms p(95)=625.03ms
     iterations..............................: 1673565  296.730017/s
     vus.....................................: 1        min=0         max=100        
     vus_max.................................: 100      min=100       max=100        
running (1h34m00.0s), 000/100 VUs, 1673565 complete and 0 interrupted iterations
5

@zephyrzen

You have a lot of dropped_iterations

And I don’t understand why the vus_max is just 100 in the summary, if you set the preAllocatedVUs to 1000 and 10000