abhishekrb19 opened a new issue #10283:
URL: https://github.com/apache/druid/issues/10283
Querying a new data source with no data -- unpublished/no segments, returns
a CalciteContextException.
### Affected Version
0.18.1
### Description
Bring up a supervisor with data source "ds". Issue a SQL query against the
new data source once the supervisor is up and running:
```
SELECT * FROM ds
```
returns the following exception:
```
Unknown exception / org.apache.calcite.runtime.CalciteContextException: From
line 1, column 15 to line 1, column 17: Object 'ds' not found /
org.apache.calcite.tools.ValidationException
```
Looking at the logs, the broker appears to be doing a SQL validation:
```
Caused by: org.apache.calcite.sql.validate.SqlValidatorException: Object
'ds' not found
at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native
Method) ~[?:1.8.0_162]
at
sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:62)
~[?:1.8.0_162]
```
As a workaround, we'd have to resort to batch-inline ingest, where we ingest
a single canned row corresponding to this data source, so there's at least one
segment that is available/published and hence is query-able. I think it'd be
nice to return "no data", which would also be consistent with the canonical
native JSON query and other SQL-like engines.
/cc: @gianm @clintropolis @jihoonson. Thanks!
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
