InfluxDB 500 error (server overloaded or impaired) when Prometheus tries to send data.

[j.d.cr...@gmail.com]

Hello.

I'm not sure if this is more an influx issue or prometheus, but I figured I had to start somewhere.

I set up my prom launch parameters as follows:

storage.remote.influxdb-url http://my.domain.me.com:8086
storage.remote.influxdb.database prometheus

And it seems to be "working," as in I can access it from the prom instance with curl. However, in the influx logs, I get this (or something like it):

[httpd] 10.0.2.2 - - [18/Jan/2017:16:57:48 +0000] "POST /write?consistency=&db=prometheus&precision=&rp=default HTTP/1.1" 500 79 "-" "InfluxDBClient" 3d7ea072-dd9f-11e6-8a1a-000000000000 340

500 error docs:
https://docs.influxdata.com/influxdb/v1.1/guides/writing_data/#http-response-summary

The 500 error is what's concerning, I don't have auth set up on the influx instance and I have a "prometheus" db to match the startup parameters. It'll spit out multiple of these messages per second, no more than 5-7. I've checked with Chronograf, curl, and Grafana that the influx instance is accessible with no auth, and those seem to work just fine.

Is there something I'm missing on the prometheus side? I'm really not handling that many data points; my ingestion rate right now, according to the Grafana dashboard, is about 275, hence the confusion about the 500 error.

Again, sorry if this is wholly an influx issue, I just wasn't quite sure where to start.

Jack

[j.d.cr...@gmail.com]

Sigh, figured it out.

I missed that there is *no* retention policy named "default" when Influx makes a DB. Prometheus assumes that, and I assumed it was right, so I didn't even think of it until I found another section on the API responses:

> 500 Internal Server Error The system is overloaded or significantly impaired. Can occur if a user attempts to write to a retention policy that does not exist. The returned JSON offers further information.

Posting so I might help someone else. Sorry!