-
Notifications
You must be signed in to change notification settings - Fork 4.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Docker terminal state #124
Comments
CC @xingyaoww who pointed this out |
Another data point: the agent keeps trying to use virtualenv, and then getting confused when it doesn't work 🙃 |
I am little confused, I try the step locally in latest main branch. But when sending |
@yufansong Thanks for bringing up the issue! We had some lengthy discussion about this yesterday -- basically we were thinking whether it make sense for us to support sending actions from the user's end at this early stage. We temporarily decided to remove the ability from the user for them to execute commands directly, so Here's another step to reproduce this issue:
See this screenshot for more details: |
Closing this in favor of #226 |
Describe the bug
Currently, our DockerInteractive terminal loses some state. In particular, if the agent runs a
cd
command, the next command doesn't run inside that directory.Steps to Reproduce
uvicorn opendevin.server.listen:app --reload --port 3000
websocat ws://127.0.0.1:3000/ws
(in a second terminal)Expected behavior
ls
command showsfile.txt
Actual behavior
ls
command shows directoryfoo
Additional context
I'm not sure what other state we might be losing with docker's
exec
command. The only other one I can think of is exported environment variables.Suggested solution
Two ways we could go here:
ssh
cwd
at the end of eachexec
, and use that as workdir for the nextexec
The text was updated successfully, but these errors were encountered: