You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The query generated and shown with format=debug looks correct, and it returns a single time bucket when run directly against Druid, but when the Fili response comes back, you get 2 time buckets for the "all" query, which should have been 1.
This is also borne out in the logs, which show that a query split is happening, even though "all" grain queries should never be split.
The text was updated successfully, but these errors were encountered:
The expectedIntervals is flattening the concept of a bucket with an interval. Under any TimeGrain, intervals must be non disjoint at the resolution of buckets. But under the all granularity disjoint intervals can share the same bucket.
Either there needs to be a bailout special case for 'all' here, or the splitter has be be layered so it can handle buckets which are not expressible in a single interval.
The query generated and shown with
format=debug
looks correct, and it returns a single time bucket when run directly against Druid, but when the Fili response comes back, you get 2 time buckets for the "all" query, which should have been 1.This is also borne out in the logs, which show that a query split is happening, even though "all" grain queries should never be split.
The text was updated successfully, but these errors were encountered: