Background Agent Environment is Unstable

[Tfcbot]

Describe the Bug

My background agent environment is constantly failing with “Failed to start VM” and when I try to rebuild it I get "Failed to to open environment window

Steps to Reproduce

1. Create a background agent request

Expected Behavior

Background agent starts working on request

Screenshots / Screen Recordings

Operating System

MacOS

Current Cursor Version (Menu → About Cursor → Copy)

Version: 1.2.4 (Universal)
VSCode Version: 1.99.3
Commit: a8e95743c5268be73767c46944a71f4465d05c90
Date: 2025-07-10T16:55:16.443Z
Electron: 34.5.1
Chromium: 132.0.6834.210
Node.js: 20.19.0
V8: 13.2.152.41-electron.0
OS: Darwin arm64 24.5.0

Does this stop you from using Cursor

Sometimes - I can sometimes use Cursor

[T1000]

Hi @Tfcbot where were temporary issues, could you check if this is still happening?

[Tfcbot]

Still happening unfortunately

[T1000]

Thanks, appreciate the update. This will need to be checked more in detail as its not a frequent case but not as easy to debug.

1. Do you already have an environment.json file in the repo and does it have a snapshot ID?

2. Could you check Developer Tools > Console for errors? (clear log before opening the Background Agent setup)

[Ravnurin1]

This was happening to me this morning as well.

Started a few Background Agents via Slack, and they all took a while to actually start, and trying to open them in Cursor didn’t even work - the chat and code wouldn’t actually load.

Just checked, and I actually have a Background Agent that’s been stuck on “Generating…” for 6 hours

[Tfcbot]

This was the last request ID: 5ae4c22c-c1fe-4255-91e8-a05008d66bf3

I dont have a snapshot ID

[T1000]

I’m checking with Cursor Team further and will update.

As that request fails, could you check Settings > Network > Run Diagnostics?