Requests with a query time range that's longer than the limit result in 5xx

[replay]

I have seen this log message in the querier:

POST /prometheus/api/v1/read (500) 397.162µs Response: \"the query time range exceeds the limit (query length: 2160h0m0s, limit: 768h0m0s) (err-mimir-max-query-length). To adjust the related per-tenant limit, configure -querier.max-partial-query-length, or contact your service administrator.

It indicates that a query was using a time range which was longer than the limit, as a result we have returned 500. This error should be a 4xx instead.

[pracucci]

/prometheus/api/v1/read is the remote read API. The issue here could be that the errors remapping we do doesn't apply to the remote read API.