Hi! I'm running OpenHands using the "Docker Directly" approach from the docs (openhands:1.7 with agent-server 1.19.1-python). docker run -it --rm --pull=always \ -e AGENT_SERVER_IMAGE_REPOSITORY=ghcr.io/openhands/agent-server \ -e AGENT_SERVER_IMAGE_TAG=1.19.1-python \ -e LOG_ALL_EVENTS=true \ -v /var/run/docker.sock:/var/run/docker.sock \ -v ~/.openhands:/.openhands \ -p 3000:3000 \ --add-host host.docker.internal:host-gateway \ --name openhands-app \ docker.openhands.dev/openhands/openhands:1.7 The app container starts fine, and the sandbox container is also running. I can successfully reach the agent server from inside the openhands-app container: curl http://host.docker.internal: → returns 200 with JSON 145820 - openhands.app_server.sandbox.docker_sandbox_serviceINFO docker_sandbox_service.py:268 - Sandbox server not running: http://host.docker.internal:43527 : Server disconnected without sending a response. 145820 - openhands.app_server.app_conversation.live_status_app_conversation_serviceERROR live_status_app_conversation_service.py:443 - Error starting conversation [05/24/26 145820] INFO [generated in 0.00256s] ('ERROR', '500: Sandbox entered error state: oh-agent-server-QT9yKziirJmiwqfxb6pfs', '{"sandbox_id": null, base.py:1846 "conversation_id": null, "initial_message": null, "system_message_suffix": null, "processors": null, "llm_model": null, "selecte ... (98 characters truncated) ... ": null, "trigger": null, "pr_number": [], "parent_conversation_id": null, "agent_type": "default", "public": null, "plugins": null, "secrets": null}', '2026-05-24 145803.995139', '2026-05-24 145820.621031', '9c3d19409147489eac65b4113d79078e')

Hey @creamy-account-90359, if it’s the same to you, I’d suggest that you take a look at our new agent canvas: https://github.com/OpenHands/agent-canvas It should make it a lot easier to run, including in docker if you wish. We’re discussing it more on the #C06QT0AGY4W channel!