Awesome
xk6-output-influxdb
k6 extension for InfluxDB v2, it adds the support for the latest v2
version and the compatibility API for v1.8+.
Why is this output not directly part of k6
core?
The k6
core already supports the InfluxDB v1 so the natural feeling would be to do the same for the v2
. Unfortunately, the v2
has introduced some breaking changes in the core parts of the API. This would make it difficult to support both versions without taking a bunch of compromises for maintaining the retro-compatibility or introducing breaking changes in the current user experience of the k6's InfluxDB output, with a high probability to create more confusion for k6's users. For this main reason, the k6
development team has decided to create a new and independent extension for InfluxDB v2.
Install
To build a k6
binary with this extension, first ensure you have the prerequisites:
- Go toolchain
- Git
- xk6
- Build with
xk6
:
xk6 build --with github.com/grafana/xk6-output-influxdb
This will result in a k6
binary in the current directory.
- Run with the just built
k6
binary:
K6_INFLUXDB_ORGANIZATION=<insert-here-org-name> \
K6_INFLUXDB_BUCKET=<insert-here-bucket-name> \
K6_INFLUXDB_TOKEN=<insert-here-valid-token> \
./k6 run -o xk6-influxdb=http://localhost:8086 <script.js>
Using Docker
This Dockerfile builds a docker image with the k6 binary.
Configuration
Options for fine-grained control for flushing and connections.
ENV | Default | Description |
---|---|---|
K6_INFLUXDB_ORGANIZATION | The Organization. | |
K6_INFLUXDB_BUCKET | The Bucket. | |
K6_INFLUXDB_TOKEN | The Token. | |
K6_INFLUXDB_ADDR | http://localhost:8086 | The address of the instance. |
K6_INFLUXDB_PUSH_INTERVAL | 1s | The flush's frequency of the k6 metrics. |
K6_INFLUXDB_CONCURRENT_WRITES | 4 | Number of concurrent requests for flushing data. It is useful when a request takes more than the expected time (more than flush interval). |
K6_INFLUXDB_TAGS_AS_FIELDS | vu:int,iter:int,url | A comma-separated string to set k6 metrics as non-indexable fields (instead of tags). An optional type can be specified using :type as in vu:int will make the field integer. The possible field types are int, bool, float and string, which is the default. Example: vu:int,iter:int,url:string,event_time:int. |
K6_INFLUXDB_INSECURE | false | When true , it will skip https certificate verification. |
K6_INFLUXDB_PRECISION | 1ns | The timestamp Precision. |
Docker Compose
This repo includes a docker-compose.yml file that starts InfluxDB, Grafana and k6. This is just a quick setup to show the usage; for real use case you might want to deploy outside of docker, use volumes and probably update versions.
Clone the repo to get started and follow these steps:
-
Put your k6 scripts in the
samples
directory or use thehttp_2.js
example. -
Start the docker compose environment.
docker compose up -d
# Output Creating xk6-output-influxdb_influxdb_1 ... done Creating xk6-output-influxdb_k6_1 ... done Creating xk6-output-influxdb_grafana_1 ... done
-
Use the k6 Docker image to run the k6 script and send metrics to the InfluxDB container started on the previous step. You must set the
testid
tag with a unique identifier to segment the metrics into discrete test runs for the Grafana dashboards.docker compose run --rm -T k6 run -<samples/http_2.js --tag testid=<SOME-ID>
For convenience, the
docker-run.sh
can be used to simply:./docker-run.sh samples/http_2.js
-
Visit http://localhost:3000/ to view results in Grafana.
This repository includes a basic dashboard. If you want to build a custom Dashboard, contact the k6 team in Slack.
Compatibility API
The v2 includes a InfluxDB v1.8+ compatibility API that adds endpoints for communicating with an InfluxDB v1.
Client API usage differences summary:
- Use the form username:password for an authentication token. Example: my-user:my-password. Use an empty string ("") if the server doesn't require authentication.
- The organization parameter is not used. Use an empty string ("") where necessary.
- Use the form database/retention-policy where a bucket is required. Skip retention policy if the default retention policy should be used. Examples: telegraf/autogen, telegraf.