JDBC Invalid SessionHandle with dbSQL Warehouse
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
01-22-2025 01:05 PM
Connecting Pentaho Ctools dashboards to Databricks using JDBC to a serverless dbSQL Warehouse, it works fine on the initial load, but then if we leave it idle for awhile and come back we get this error:
[Databricks][JDBCDriver](500593) Communication link failure. Failed to connect to server. Reason: HTTP Response code: 400, Error message: INVALID_STATE: Invalid SessionHandle: SessionHandle
It seems like maybe we need to set some sort of timeout configuration similar to what is suggested here except I cannot find any documentation about what that would be for dbSQL. Has anyone encountered this?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
01-23-2025 10:08 PM
@pdiamond wrote:Connecting Pentaho Ctools dashboards to Databricks using JDBC to a serverless dbSQL Warehouse, it works fine on the initial load, but then if we leave it idle for awhile and come back we get this error:
[Databricks][JDBCDriver](500593) Communication link failure. Failed to connect to server. Reason: HTTP Response code: 400, Error message: INVALID_STATE: Invalid SessionHandle: SessionHandle
It seems like maybe we need to set some sort of timeout configuration similar to what is suggested here except I cannot find any documentation about what that would be for dbSQL. Has anyone encountered this?
Hello @pdiamond,
It sounds like you might need to configure a timeout setting for your JDBC connection to dbSQL Warehouse. While specific documentation for dbSQL isn't readily available, you can try adjusting the connection timeout settings in your JDBC URL or your application's connection pool configuration.
Best regards,
David Scott
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
01-30-2025 05:26 AM
I should have mentioned that we're using AuthMech=3 and in the JDBC docs (Databricks JDBC Driver Installation and Configuration Guide) I don't see any relevant timeout settings that would apply in that scenario. Am I missing something?

