Influxdb reported a large number of errors when run k6


hello guys, when i perform k6 run --out influxdb=http://xx.xx.xx.xx:8086/xxx.js, influxdb reported a large number of errors, how can I solve this problem?
Below are k6 version and influxdb version:
image
image

k6 sends quite a lot of data and it seems like your are overwhelming your InfluxDB instance, it can’t keep up with what k6 is sending it. The only things you can configure on the k6 side are these, and while K6_INFLUXDB_PUSH_INTERVAL and K6_INFLUXDB_CONCURRENT_WRITES might help a bit, I suggest giving more resources to your InfluxDB server. I also see a few Request Entity Too Large InfluxDB errors in your logs, which means you also need to adjust its configuration for that, see this other forum topic about it: 413 Request too large error

Alternatively, you can try another output like Prometheus remote-write, or saving the raw k6 results to a .json or .csv file and saving them to InfluxDB manually, at a slower pace, after the test has finished executing.

hi @ned , I have adjusted influxdb‘s configuration option, but still threw a lot errors “Request Entity Too Large”, I observed that the load on the influxdb server was normal.


See the other thread I linked to, 413 Request too large error - #3 by codebien, you probably have to configure INFLUXDB_HTTP_MAX_BODY_SIZE option for InfluxDB. But this is more of an InfluxDB question, not a k6 question.

1 Like