microsoft / vscode-copilot-release

Feedback on GitHub Copilot Chat UX in Visual Studio Code.
https://marketplace.visualstudio.com/items?itemName=GitHub.copilot-chat
Creative Commons Attribution 4.0 International
341 stars 37 forks source link

Sorry, no response was returned after returning response #2157

Open juanfadev opened 3 weeks ago

juanfadev commented 3 weeks ago

Type: Bug

Calling claude sonnet, it returned a response; but in the middle of the process of editing a file, it suddenly showed "Sorry. no response was returned" and all the progress on that file was lost.

Extension version: 0.22.0 VS Code version: Code 1.95.0 (912bb683695358a54ae0c670461738984cbb5b95, 2024-10-28T20:16:24.561Z) OS version: Windows_NT x64 10.0.22631 Modes: Remote OS version: Linux x64 5.15.153.1-microsoft-standard-WSL2

System Info |Item|Value| |---|---| |CPUs|AMD Ryzen 5 5600X 6-Core Processor (12 x 3700)| |GPU Status|2d_canvas: enabled
canvas_oop_rasterization: enabled_on
direct_rendering_display_compositor: disabled_off_ok
gpu_compositing: enabled
multiple_raster_threads: enabled_on
opengl: enabled_on
rasterization: enabled
raw_draw: disabled_off_ok
skia_graphite: disabled_off
video_decode: enabled
video_encode: enabled
vulkan: disabled_off
webgl: enabled
webgl2: enabled
webgpu: enabled
webnn: disabled_off| |Load (avg)|undefined| |Memory (System)|31.92GB (10.03GB free)| |Process Argv|--crash-reporter-id 168a6030-29a0-424a-8b20-1c8109cbef2c| |Screen Reader|no| |VM|0%| |Item|Value| |---|---| |Remote|WSL: Ubuntu| |OS|Linux x64 5.15.153.1-microsoft-standard-WSL2| |CPUs|AMD Ryzen 5 5600X 6-Core Processor (2 x 0)| |Memory (System)|3.83GB (0.35GB free)| |VM|0%|
A/B Experiments ``` vsliv368cf:30146710 vspor879:30202332 vspor708:30202333 vspor363:30204092 vswsl492cf:30256860 vscod805cf:30301675 binariesv615:30325510 vsaa593cf:30376535 py29gd2263:31024239 c4g48928:30535728 azure-dev_surveyone:30548225 a9j8j154:30646983 962ge761:30959799 pythongtdpath:30769146 pythonnoceb:30805159 asynctok:30898717 pythonmypyd1:30879173 h48ei257:31000450 pythontbext0:30879054 cppperfnew:31000557 dsvsc020:30976470 pythonait:31006305 dsvsc021:30996838 945dj816:31013170 dvdeprecation:31068756 dwnewjupytercf:31046870 impr_priority:31102340 nativerepl2:31139839 refactort:31108082 pythonrstrctxt:31112756 wkspc-onlycs-t:31132770 wkspc-ranged-t:31151552 cf971741:31144450 iacca2:31156134 notype1cf:31157160 5fd0e150:31155592 dwcopilot:31170013 ```
lramos15 commented 3 weeks ago

Do you have a reliable repro for this?

ynaiborlang commented 3 weeks ago

yep, claude not working for me as well, returning the same error message

craigphicks commented 3 weeks ago

I started a new chat and at least it began to return a reply.

neverset123 commented 2 weeks ago

same here. claude is much lower than gpt-4o...

lramos15 commented 2 weeks ago

Can you include your logs via CTRL + SHIFT + U -> GitHub copilot Chat

ynaiborlang commented 2 weeks ago

Image

as you can see, I am using Claude 3.5 but I'm not getting any response, the OpenAI models are working fine

ynaiborlang commented 2 weeks ago

Can you include your logs via CTRL + SHIFT + U -> GitHub copilot Chat

please find the image above, thank you

lramos15 commented 2 weeks ago

@ynaiborlang I cannot see the whole log because it is cut off. Can you paste it here.

andeeplus commented 2 weeks ago

I've got the same issue.

On the logs I can see the following:

[error] CodeExpectedError: cannot open xxx (Error: Unable to resolve nonexistent file xxx)

The error is triggered for several files that has been deleted in the past and they are not related with the current file context selected.

After all the errors I have this info message

2024-11-06 21:40:15.947 [info] request done: requestId: [3b272269-fdaa-43a0-937e-06578361e1c5] model deployment ID: []


Just for debugging purpouse I've tried to check the logs os the same message using GPT 4o

It logs the same errors and the same info at the end, this time preceded by, but it works.

2024-11-06 21:45:59.729 [info] message 0 returned. finish reason: [stop]
2024-11-06 21:45:59.730 [info] request done: requestId: [0be3a8ae-a7e2-4653-bb46-d98d71696e6c] model deployment ID: []

I'm on a npm workspace.


The fun thing is that if I open a new vscode instance, empty one, and I try to use Claude 3.5 it works without any issue.


If I open a different monorepo I get the same issue.


Tried with another project a simple one, no monorepo, new chat. It starts with this message:

Workspace(rerun without)...
Still building the workspace index, response may be less accurate.

It took 1/2 minutes to give the response but it worked


[SOLVED] Last shoot I went back to the first project, opened a new chat (before I was using the already active chat, I do not change it often, I use to keep always the same opened and use it - btw is it a good or bad practice?) and now Claude 3.5 is working again.

Not so intuitive but at the end looks like I can use it. I hope you can find this live debugging session helpful, I'm not an expert on this field but looks like there is a bug for sure on how chat sessions are handled with different models.

Also this should explain why it trigs the errors above, the chat is probably referencing to indices that has been previously removed. Thinking about this I'd say I should start new chats more often.

ynaiborlang commented 2 weeks ago

I've got the same issue.

On the logs I can see the following:

[error] CodeExpectedError: cannot open xxx (Error: Unable to resolve nonexistent file xxx)

The error is triggered for several files that has been deleted in the past and they are not related with the current file context selected.

After all the errors I have this info message

2024-11-06 21:40:15.947 [info] request done: requestId: [3b272269-fdaa-43a0-937e-06578361e1c5] model deployment ID: []

Just for debugging purpouse I've tried to check the logs os the same message using GPT 4o

It logs the same errors and the same info at the end, this time preceded by, but it works.

2024-11-06 21:45:59.729 [info] message 0 returned. finish reason: [stop]
2024-11-06 21:45:59.730 [info] request done: requestId: [0be3a8ae-a7e2-4653-bb46-d98d71696e6c] model deployment ID: []

I'm on a npm workspace.

The fun thing is that if I open a new vscode instance, empty one, and I try to use Claude 3.5 it works without any issue.

If I open a different monorepo I get the same issue.

Tried with another project a simple one, no monorepo, new chat. It starts with this message:

Workspace(rerun without)...
Still building the workspace index, response may be less accurate.

It took 1/2 minutes to give the response but it worked

[SOLVED] Last shoot I went back to the first project, opened a new chat (before I was using the already active chat, I do not change it often, I use to keep always the same opened and use it - btw is it a good or bad practice?) and now Claude 3.5 is working again.

Not so intuitive but at the end looks like I can use it. I hope you can find this live debugging session helpful, I'm not an expert on this field but looks like there is a bug for sure on how chat sessions are handled with different models.

Also this should explain why it trigs the errors above, the chat is probably referencing to indices that has been previously removed. Thinking about this I'd say I should start new chats more often.

this SOLVED for me as well, opening a new chat