I am running Grafana and InfluxDB on the same server, and have a number of metrics providing alerts without issue. However, one of the alerts fairly consistently emails me the following [Alerting] error immediately followed by an [OK] email.
tsdb.HandleRequest() error Get http://localhost:8086/query?db=solar&epoch=s&q=SELECT+sum%28%22power%22%29+-+sum%28%22ac_power%22%29+FROM+%22autogen%22.%22helioscope%22+WHERE+time+%3E+now%28%29+-+720h+GROUP+BY+time%281d%29+fill%28none%29: dial tcp [127.0.0.1:8086](http://127.0.0.1:8086/): connect: connection refused
While it’s just an annoyance since it immediately succeeds, I want to fix the problem before attempting to scale the system. For now it’s easy enough just to turn off the “If execution error or timeout” alerting.