Fields are quite strange in InfluxDB.
On the one hand, they are required.
On they other, they are not useful in complex SQL queries, as they lack support for some important operations such as GROUP BY.
Can we please add support for GROUP BY on fields, and make fields optional?
For custom telemetry data, users must populate at least one field, such as a dummy field. That’s wasteful.
But for stock telemetry data such as environment metrics (cpu, disk, mem, etc.), GROUP BY is still not an option. In some cases DISTINCT may provide a workaround, but I can’t find syntax which InfluxDB’s SQL implementations support.
