I am using the Telegraf + InfluxDB + Grafana stack to display SQL-Server metrics.
My goal is to down-sample the old data, for example - after a week, group the 10s inputs to 2m inputs.
I cannot simply down-sample the data without breaking the graphs.
Following the guides:
Example 3 in continous query guide
and down-sampling guide
The original inputs + down-sampled inputs + broken grafana query:
As you see, the columns are just not the same.
The grafana graphs cannot work with the new down-sampled data.
Without down-sampling, the graphs just won’t load due to the high amounts of points kept.
How can this not be a must have feature? And how can I solve this?