Datasources are fully available when no historical is present - Githubissues

apache / druid

Apache Druid: a high performance real-time analytics database.

https://druid.apache.org/

Apache License 2.0

13.46k stars 3.7k forks source link

Datasources are fully available when no historical is present #16794

Open mounikanakkala opened 2 months ago

mounikanakkala commented 2 months ago

Affected Version

Version 28.0.1

Description

Cluster setup

One coordinator, one overlord
One broker, one router
Zero historicals
Zero middlemanagers

Steps to reproduce the problem

Keep one historical with a tier, say Tier_A
Keep default retention rule to load at least one replica to Tier_A
Any datasource on the cluster should be loaded to Tier_A historical. Datasources are queryable
Now, remove the historical.
Datasources will be shown as available, but when queried, there will be an error.
```
Error: INVALID_INPUT
```

Object 'datasource_name' not found

mounikanakkala commented 2 months ago

Hello,

May I know if you have any update on this?