InfluxDB2 Integration with Telegraf Agent not working
Hi all,
The InfluxDB2 Integration is currently not working with InfluxData´s Telegraf Agent with an active InfluxDB2 Input Module because the InfluxDB2 Integration lacks the /api/v2/ping API Interface (based on the error Logs it is asking for it).
For sending InfluxDB2 metrics, the /api/v2/write API Interface is totaly enough to send metrics! https://github.com/influxdata/telegraf/tree/master/plugins/inputs/influxdb_v2_listener
Cheers
Jonas Hahn
Comments: 2
-
yesterday
OliverI did a workaround by using nginx as a reverse proxy (sending http 204 on ping endpoint) and forward the write endpoint to influxdb. Not a solid solution (you have to scale nginx very well because of the amount of requests) but at least a fix for some scenarios. You might also look into VictoriaMetrics as a influxdb replacement. It supports influxdb v1 and v2 protocol for writing metrics.
-
yesterday
Marcel Arentz AdminI'm sorry to hear that the input listener plug-in is not providing the /v2/ping endpoint. Checkmk is using this endpoint to verify that the instance is available and we can reliable send metrics to this instance. This is a very important part of the integration which we cannot change without strong reasons.
If you want to use the plug-in of telegraf I would rather ask for adding this endpoint there. Please also note the official statement of the API documentation for the ping endpoint:
"Get the status of the instance
Retrieves the status and InfluxDB version of the instance.
Use this endpoint to monitor uptime for the InfluxDB instance. The response returns a HTTP 204 status code to inform you the instance is available."