Weird measurements in InfluxDB

Hi all,
I don’t know since when this happens, but I have strange measurements in our Icinga2 InfluxDB and cannot find out, where this is coming from.
OS: Ubuntu 18.04.4 LTS
Influx: 1.8.4-1
Icinga2: r2.12.3-1 / 2 Masters
Config:

object InfluxdbWriter "influxdb" {
  host = "1.2.3.4"
  port = 8086
  database = "icinga2"
  password = "xxx"
  flush_threshold = 4096
  flush_interval = 60s
  enable_send_thresholds = true
  enable_send_metadata = true
  enable_ha = true
  host_template = {
    measurement = "$host.check_command$"
    tags = {
      hostname = "$host.name$"
    }
  }
  service_template = {
    measurement = "$service.check_command$"
    tags = {
      hostname = "$host.name$"
      service = "$service.name$"
    }
  }
}

Beside my correct measurements, I have sth. like this:
image

Problem here, all my “Cleanup” Scripts don’t work anymore and I have absolutely no idea, where these strange measurements are coming from.

Cheers,
Marcus

I have not seen this with InfluxDB, but it looks like an encoding error. Perhaps you can identify the host/service object in some way and verify very the encoding gets messed up (Plugin, Icinga, InfluxDB). If no other attribute helps to identify, perhaps the readable part “isco_standard” from the first one displayed helps in a search.

It does not look like an Icinga problem, also Influx DBs are f… up… Don’t know why. I guess, I will try Influxdb2 with v1 compatibility mode

Hello, good idea, as far as i know icinga only supports v1 of influxdb protocol, you could also use telegraf as a gateway for icinga from v2 to v1.