Initialize Spark Connect session in Jupyter init script#4454

Draft

ilia-db wants to merge 25 commits intomainfrom

ssh-tunnel-spark-init

Commits on Feb 12, 2026

WIP: Add serverless GPU compute support to SSH tunnel
ilia-db
committed
Add liteswap header value for traffic routing (dev/test only).
ilia-db
committed
Add liteswap option to the ProxyCommand.
ilia-db
committed
Fix lint error
ilia-db
committed
Add GPU accelerator support for serverless compute
ilia-db
committed
Don't log metadata response to stdout
ilia-db
committed
Move metadata URL logging to debug level
ilia-db
committed
Simplify SSH tunnel job polling
ilia-db
committed
Make sure that connection name is not empty in serverless mode
ilia-db
committed
Extract serverless environment key to a constant
ilia-db
committed
Add IDE flag and support for IDE integration for vscode and cursor
ilia-db
committed
Mark hidden flags for serverless compute, remove manual Http call
ilia-db
committed
Separate IDE options into constants
ilia-db
committed
Move unnecessary logging to debug level
ilia-db
committed
Remove dedicated cluster and serverless compute specific documentation from connect command help message
ilia-db
committed
New config logic
ilia-db
committed
Always recreate SSH config entry
ilia-db
committed
Fix sshconfig logic and tests on Windows
ilia-db
committed
Do not create secret scope if it already exists
ilia-db
committed
Better SSHD config escaping
ilia-db
committed
Avoid duplication of host config generation code
ilia-db
committed
Escape backslashes in SetEnv values
ilia-db
committed
Initialize Spark Connect session in Jupyter init script
ilia-db
committed
Format jupyter-init.py
ilia-db
committed
Use DatabricksSession for Spark Connect session initialization
ilia-db
committed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Initialize Spark Connect session in Jupyter init script#4454

Initialize Spark Connect session in Jupyter init script#4454
ilia-db wants to merge 25 commits intomainfrom
ssh-tunnel-spark-init

Commits on Feb 12, 2026

WIP: Add serverless GPU compute support to SSH tunnel

Add liteswap header value for traffic routing (dev/test only).

Add liteswap option to the ProxyCommand.

Fix lint error

Add GPU accelerator support for serverless compute

Don't log metadata response to stdout

Move metadata URL logging to debug level

Simplify SSH tunnel job polling

Make sure that connection name is not empty in serverless mode

Extract serverless environment key to a constant

Add IDE flag and support for IDE integration for vscode and cursor

Mark hidden flags for serverless compute, remove manual Http call

Separate IDE options into constants

Move unnecessary logging to debug level

Remove dedicated cluster and serverless compute specific documentation from connect command help message

New config logic

Always recreate SSH config entry

Fix sshconfig logic and tests on Windows

Do not create secret scope if it already exists

Better SSHD config escaping

Avoid duplication of host config generation code

Escape backslashes in SetEnv values

Initialize Spark Connect session in Jupyter init script

Format jupyter-init.py

Use DatabricksSession for Spark Connect session initialization