A k6 extension implements Prometheus HTTP exporter as k6 output extension.
Using xk6-prometheus output extension you can collect metrics from long running k6 process with Prometheus. All custom k6 metrics (Counter,Gauge,Rate,Trend) and build-in metrics will be accessible as appropiate Prometheus metrics on a given HTTP port in Prometheus importable text format.
To build a k6
binary with this extension, first ensure you have the prerequisites:
- Go toolchain
- Git
Then:
- Download
xk6
:
$ go install go.k6.io/xk6/cmd/xk6@latest
- Build the binary:
$ xk6 build --with github.com/alexander-falca/xk6-prometheus@latest
You should use at least
v0.31.0
version because xk6-prometheus extension registers itself as output extension. This feature introduced in thev0.31.0
version of k6.
Without parameters the Prometheus HTTP exporter will accessible on port 5656
.
$ ./k6 run -d 1m --out prometheus script.js
/\ |‾‾| /‾‾/ /‾‾/
/\ / \ | |/ / / /
/ \/ \ | ( / ‾‾\
/ \ | |\ \ | (‾) |
/ __________ \ |__| \__\ \_____/ .io
execution: local
script: script.js
output: prometheus (:5656)
scenarios: (100.00%) 1 scenario, 1 max VUs, 1m30s max duration (incl. graceful stop):
* default: 1 looping VUs for 1m0s (gracefulStop: 30s)
running (1m01.0s), 0/1 VUs, 54 complete and 0 interrupted iterations
default ✓ [======================================] 1 VUs 1m0s
data_received..................: 611 kB 10 kB/s
data_sent......................: 4.1 kB 67 B/s
http_req_blocked...............: avg=3.37ms min=2.86µs med=3.82µs max=181.96ms p(90)=11.15µs p(95)=13.52µs
http_req_connecting............: avg=2.19ms min=0s med=0s max=118.34ms p(90)=0s p(95)=0s
http_req_duration..............: avg=125.14ms min=118.99ms med=120.68ms max=237.66ms p(90)=121.45ms p(95)=124.07ms
{ expected_response:true }...: avg=125.14ms min=118.99ms med=120.68ms max=237.66ms p(90)=121.45ms p(95)=124.07ms
http_req_failed................: 0.00% ✓ 0 ✗ 54
http_req_receiving.............: avg=5.1ms min=85.32µs med=792.2µs max=118.29ms p(90)=860.41µs p(95)=903.71µs
http_req_sending...............: avg=20.68µs min=12.53µs med=16.69µs max=75.97µs p(90)=29.39µs p(95)=37.87µs
http_req_tls_handshaking.......: avg=0s min=0s med=0s max=0s p(90)=0s p(95)=0s
http_req_waiting...............: avg=120.01ms min=118.17ms med=119.78ms max=127.48ms p(90)=120.6ms p(95)=120.71ms
http_reqs......................: 54 0.885451/s
iteration_duration.............: avg=1.12s min=1.11s med=1.12s max=1.3s p(90)=1.12s p(95)=1.16s
iterations.....................: 54 0.885451/s
vus............................: 1 min=1 max=1
vus_max........................: 1 min=1 max=1
The output extension accept parameters in a standard query string format:
k6 run --out 'prometheus=param1=value1¶m2=value2¶m3=value3'
Note apostrophe (
'
) characters around the--out
parameter! You should use it for escape&
characters from shell (or use backslash before&
characters).
The following paremeters are recognized:
parameter | description |
---|---|
namespace | Prometheus namespace for exported metrics (default: "", empty) |
subsystem | Prometheus subsystem for exported metrics (default: "", empty) |
host | Hostname or IP address for HTTP endpoint (default: "", empty, listen on all interfaces) |
port | TCP port for HTTP endoint (default: 5656) |
It is recommended to use k6
as either namespace
or subsystem
to prefix exported metrics names with k6_
string.
Here is the relevant part of the metrics HTTP response:
# HELP k6_data_received The amount of received data
# TYPE k6_data_received counter
k6_data_received{group="",scenario="default"} 430237
# HELP k6_data_sent The amount of data sent
# TYPE k6_data_sent counter
k6_data_sent{group="",scenario="default"} 2888
# HELP k6_http_req_blocked Time spent blocked before initiating the request
# TYPE k6_http_req_blocked summary
k6_http_req_blocked{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.5"} 0.00377
k6_http_req_blocked{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.9"} 0.008411
k6_http_req_blocked{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.95"} 0.016064
k6_http_req_blocked{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="1"} 872.706547
k6_http_req_blocked_sum{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 872.8805460000001
k6_http_req_blocked_count{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 38
# HELP k6_http_req_blocked_current Time spent blocked before initiating the request (current)
# TYPE k6_http_req_blocked_current gauge
k6_http_req_blocked_current{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 0.003246
# HELP k6_http_req_connecting Time spent establishing TCP connection
# TYPE k6_http_req_connecting summary
k6_http_req_connecting{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.5"} 0
k6_http_req_connecting{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.9"} 0
k6_http_req_connecting{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.95"} 0
k6_http_req_connecting{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="1"} 424.67631
k6_http_req_connecting_sum{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 424.67631
k6_http_req_connecting_count{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 38
# HELP k6_http_req_connecting_current Time spent establishing TCP connection (current)
# TYPE k6_http_req_connecting_current gauge
k6_http_req_connecting_current{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 0
# HELP k6_http_req_duration Total time for the request
# TYPE k6_http_req_duration summary
k6_http_req_duration{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.5"} 433.957267
k6_http_req_duration{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.9"} 850.686511
k6_http_req_duration{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.95"} 975.504212
k6_http_req_duration{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="1"} 3292.618482
k6_http_req_duration_sum{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 21587.420325
k6_http_req_duration_count{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 38
# HELP k6_http_req_duration_current Total time for the request (current)
# TYPE k6_http_req_duration_current gauge
k6_http_req_duration_current{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 418.772552
# HELP k6_http_req_failed The rate of failed requests
# TYPE k6_http_req_failed histogram
k6_http_req_failed_bucket{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",le="0"} 38
k6_http_req_failed_bucket{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",le="+Inf"} 38
k6_http_req_failed_sum{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 0
k6_http_req_failed_count{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 38
# HELP k6_http_req_receiving Time spent receiving response data
# TYPE k6_http_req_receiving summary
k6_http_req_receiving{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.5"} 1.729595
k6_http_req_receiving{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.9"} 5.581774
k6_http_req_receiving{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.95"} 512.29381
k6_http_req_receiving{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="1"} 1533.230124
k6_http_req_receiving_sum{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 2532.616402999998
k6_http_req_receiving_count{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 38
# HELP k6_http_req_receiving_current Time spent receiving response data (current)
# TYPE k6_http_req_receiving_current gauge
k6_http_req_receiving_current{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 2.20869
# HELP k6_http_req_sending Time spent sending data
# TYPE k6_http_req_sending summary
k6_http_req_sending{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.5"} 0.015493
k6_http_req_sending{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.9"} 0.026709
k6_http_req_sending{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.95"} 0.030378
k6_http_req_sending{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="1"} 0.095429
k6_http_req_sending_sum{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 0.7408489999999998
k6_http_req_sending_count{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 38
# HELP k6_http_req_sending_current Time spent sending data (current)
# TYPE k6_http_req_sending_current gauge
k6_http_req_sending_current{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 0.013392
# HELP k6_http_req_tls_handshaking Time spent handshaking TLS session
# TYPE k6_http_req_tls_handshaking summary
k6_http_req_tls_handshaking{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.5"} 0
k6_http_req_tls_handshaking{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.9"} 0
k6_http_req_tls_handshaking{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.95"} 0
k6_http_req_tls_handshaking{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="1"} 0
k6_http_req_tls_handshaking_sum{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 0
k6_http_req_tls_handshaking_count{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 38
# HELP k6_http_req_tls_handshaking_current Time spent handshaking TLS session (current)
# TYPE k6_http_req_tls_handshaking_current gauge
k6_http_req_tls_handshaking_current{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 0
# HELP k6_http_req_waiting Time spent waiting for response
# TYPE k6_http_req_waiting summary
k6_http_req_waiting{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.5"} 431.363566
k6_http_req_waiting{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.9"} 535.981583
k6_http_req_waiting{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="0.95"} 975.42239
k6_http_req_waiting{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io",quantile="1"} 1759.361649
k6_http_req_waiting_sum{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 19054.063072999994
k6_http_req_waiting_count{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 38
# HELP k6_http_req_waiting_current Time spent waiting for response (current)
# TYPE k6_http_req_waiting_current gauge
k6_http_req_waiting_current{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 416.55047
# HELP k6_http_reqs How many HTTP requests has k6 generated, in total
# TYPE k6_http_reqs counter
k6_http_reqs{expected_response="true",group="",method="GET",name="http://test.k6.io",proto="HTTP/1.1",scenario="default",status="200",url="http://test.k6.io"} 38
# HELP k6_iteration_duration The time it took to complete one full iteration
# TYPE k6_iteration_duration summary
k6_iteration_duration{group="",scenario="default",quantile="0.5"} 1434.73512
k6_iteration_duration{group="",scenario="default",quantile="0.9"} 1945.377781
k6_iteration_duration{group="",scenario="default",quantile="0.95"} 2301.369932
k6_iteration_duration{group="",scenario="default",quantile="1"} 4293.211801
k6_iteration_duration_sum{group="",scenario="default"} 60485.744295
k6_iteration_duration_count{group="",scenario="default"} 38
# HELP k6_iteration_duration_current The time it took to complete one full iteration (current)
# TYPE k6_iteration_duration_current gauge
k6_iteration_duration_current{group="",scenario="default"} 1419.901911
# HELP k6_iterations The aggregate number of times the VUs in the test have executed
# TYPE k6_iterations counter
k6_iterations{group="",scenario="default"} 38
# HELP k6_vus Current number of active virtual users
# TYPE k6_vus gauge
k6_vus 1
# HELP k6_vus_max Max possible number of virtual users
# TYPE k6_vus_max gauge
k6_vus_max 1