spotify
Add cluster analytics/stats endpoint to the API
Right now its hard to discover information about how a Heroic cluster is being used, requiring something like a separate QueryLog analytics setup. Some of this information is actually pretty discoverable within Heroic itself, and just needs a decent way of querying it.
We already have an API endpoint that can return the total number of active timeseries. Some other possibilities that have varying levels of difficulty:
- number of active requests by endpoint
- requests that have been running for over a set amount of time (similar to something like pg_stat_statements)
- echo back current config
- information on the various pipeline connections (PubSub, Cassandra, BigTable)
As an example for comparison, OpenTSDB has an endpoint that returns all active queries and the 256 most recently completed ones.
@sming Not directly, although having this kind of data would help us make more informed decisions when prioritizing our efforts.
Yes, it looked like a "supporting cast" kind of ticket. Will message you on Slack.