[Question] Connected to Databricks Workspace but unable to sync workspace folder destination
Mpicca opened this issue · 5 comments
Describe the bug
When attempting to sync to any folder in my workspace explorer i receive the following error:
logger: 'Extension',
operationId: undefined,
operationName: undefined,
loggingFunction: undefined,
timestamp: 1682639326027,
error: {
stack: 'Error: Local file URI scheme must be file. Found dbws (\)\n' +
' at new LocalUri (c:\Users\matti\.vscode\extensions\databricks.databricks-0.3.7-win32-x64\out\extension.js:38071:19)\n' +
' at ConnectionManager.attachSyncDestination (c:\Users\matti\.vscode\extensions\databricks.databricks-0.3.7-win32-x64\out\extension.js:39838:35)\n' +
' at WorkspaceFsCommands.attachSyncDestination (c:\Users\matti\.vscode\extensions\databricks.databricks-0.3.7-win32-x64\out\extension.js:42693:5)\n' +
' at m.h (c:\Users\matti\AppData\Local\Programs\Microsoft VS Code\resources\app\out\vs\workbench\api\node\extensionHostProcess.js:98:125789)',
message: 'Local file URI scheme must be file. Found dbws (\)'
},
level: 'error',
message: 'Attach Sync Destination error'
To Reproduce
Steps to reproduce the behavior:
- In workspace explorer, hover over folder, right click and click "Attach folder as sync destination"
- Does not sync and Output shows error above.
Screenshots
If applicable, add screenshots to help explain your problem.
System information:
Databricks Extension Version : 0.3.11
Hi @Mpicca. It doesn't seem that your update went through properly, because you still have databricks.databricks-0.3.7
in the logs.
We have removed right click and attach option from workspace explorer in the latest version due to issues with that option. You can attach a workspace folder by clicking the gear icon next to sync destination configuration item.
Please give it a try and let us know if you are still having troubles.
okay I've update the extension, now receive the following output:
logger: 'Extension',
operationId: undefined,
operationName: undefined,
loggingFunction: undefined,
timestamp: 1682705292440,
error: {
stack: 'Error: Local file URI scheme must be file. Found dbws (\)\n' +
' at new LocalUri (c:\Users\matti\.vscode\extensions\databricks.databricks-0.3.11-win32-x64\out\extension.js:89856:19)\n' +
' at ConnectionManager.attachSyncDestination (c:\Users\matti\.vscode\extensions\databricks.databricks-0.3.11-win32-x64\out\extension.js:91755:35)',
message: 'Local file URI scheme must be file. Found dbws (\)'
},
level: 'error',
message: 'Local file URI scheme must be file. Found dbws (\)'
}
{
logger: 'Extension',
operationId: '5b58f20a-8f7c-4408-bd41-cd14d514b958',
operationName: 'DatabricksEnvFileManager.writeFile',
loggingFunction: 'DatabricksEnvFileManager.writeFile',
timestamp: 1682705169235,
error: {
stack: 'Error: EISDIR: illegal operation on a directory, read',
message: 'EISDIR: illegal operation on a directory, read',
errno: -4068,
code: 'EISDIR',
syscall: 'read'
},
level: 'error',
message: "Can't load .env file"
Hi @Mpicca. Are you trying to use this extension with Databricks Power Tools extension? We don't use dbws URIs but the Power Tools extension does. They also initialise a databricks workspace in the "Workspace Explorer" view of vscode. This might be causing us to pickup the wrong workspace.
If this is indeed the case, can you try either
- Closing the databricks workspace in the "Workspace Explorer", OR
- Uninstalling the power tools extension trying again with only installing this extension.
I think I have the same error. I had the « Databricks power tools » extension and tried with or without it it finally came to the same problem.
My behavior is the following :
- Cloning a new repository
- Setting up the sync (works at start)
- Sync stop working with state "ERROR"
- Unable to sync the directory again (even by changing the sync destination path)
Unfortunately, I wasn't able to see any error logs that could help us to debug. Which logs do you need (and how to access them) ?
Hi @LvffY. This is how you get logs https://docs.databricks.com/dev-tools/vscode-ext.html#send-usage-logs-to-databricks. If your logs contain sensitive data, you can ping me on linkedin and I can share my email.