[kalmar]

Honest question: how do people use influxdb for monitoring and alerting? Our metrics feed into influx, and I cannot get answers to simple questions like “what is the failure rate” because arithmetic across measurements isn't possible [0]. I could shoehorn things into a schema to make it work, but in the limit I end up with one mega measurement.

[0]: https://github.com/influxdata/influxdb/issues/3552

[aequitas]

Maybe try to use Graphite as additional query API for influx. We switched to Influx from Graphite but some queries where unable or cumbersome to translate into Influx query language (especially inside Grafana). In our case we use Graphite-api with a Influx plugin instead of Graphite's own frontend. https://github.com/InfluxGraph/influxgraph

[_jomo]

It's possible to write raw queries in Grafana by changing the query edit mode.

[aequitas]

True, but even then there are some queries you can't do which graphite can. And you loose the nice editor grafana offers.

[agnivade]

> Honest question: how do people use influxdb for monitoring and alerting?

In our case, influx+grafana+alert notifications work well.

Yes, the query language needs a lot of work. It doesn't support anything beyond simple queries.

[zaarn]

I've had similar problems.

I feed analytics from my webpage into InfluxDB and it is impossible to compute a histogram of pageload times of the last 10k hits.

[agnivade]

Right. You would have to approximate the last 10k hits to the time period.

Also, check out the grafana histogram plugin. Works great for these scenarios.

[zaarn]

I've tried the grafana histogram plugin but it doesn't work; simply get a 404 error on load and makes the entire row unusuable.