Hi,
I’m using Telegraf 1.4.0-rc2 and I have a few errors in the logs that look like this:
2017-08-24T14:31:35Z E! InfluxDB Output Error: Post http://10.22.29.113:8086/write?consistency=any&db=staging: net/http: request canceled (Client.Timeout exceeded while awaiting headers)
2017-08-24T14:31:35Z E! Error writing to output [influxdb]: Could not write to any InfluxDB server in cluster
2017-08-24T14:31:45Z E! InfluxDB Output Error: Post http://10.22.29.113:8086/write?consistency=any&db=staging: net/http: request canceled (Client.Timeout exceeded while awaiting headers)
2017-08-24T14:31:45Z E! Error writing to output [influxdb]: Could not write to any InfluxDB server in cluster
2017-08-24T14:31:50Z E! InfluxDB Output Error: Post http://10.22.29.113:8086/write?consistency=any&db=staging: dial tcp 10.22.29.113:8086: getsockopt: connection refused
2017-08-24T14:31:50Z E! Error writing to output [influxdb]: Could not write to any InfluxDB server in cluster
2017-08-24T14:31:50Z E! InfluxDB Output Error: Post http://10.22.29.113:8086/write?consistency=any&db=staging: read tcp 10.22.29.218:36100->10.22.29.113:8086: read: connection reset by peer
2017-08-24T14:31:50Z E! Error writing to output [influxdb]: Could not write to any InfluxDB server in cluster
It’s possible I got these errors at the same time I was running manual queries using the Data Explorer in Chronograf.
The question is, do these errors mean data was lost? Or would Telegraf have queued them up and eventually written them out once it was able to restore it’s connection? Metric data is one thing, but I want to ensure that my log parsers are not losing data when these kinds of errors happen.
Thanks!