[BUG] Remote agent runs stuck on "Setting up a cloud container" — cannot cancel
Preflight Checklist
- [ ] I have searched existing issues and this hasn't been reported yet
- [x] This is a single bug report (please file separate reports for different bugs)
- [x] I am using the latest version of Claude Code
What's Wrong?
Body:
Scheduled and manual remote agent runs are stuck indefinitely at the "Setting up a cloud container" step
and never progress to "Clone repository." The issue affects both the automated schedule (3:00 AM EDT)
and manual triggers.
What's happening:
- All runs show status "Running" but are frozen at container setup
- The cancel button is unavailable — runs cannot be cancelled from the UI
- Affected runs: today at 3:00 AM (scheduled), 3:24 PM (manual), 3:30 PM (manual)
- Agent name: "Second Brain — Nightly Journal", runs daily at 3:00 AM EDT
What Should Happen?
What should happen (normal flow):
- Setting up a cloud container — Anthropic spins up a fresh Linux VM in the cloud
- Clone repository — your connected GitHub repo gets cloned into that VM
- Run setup script — installs dependencies (Node, any tools your agent needs)
- Start Claude Code — Claude Code CLI launches and runs your scheduled prompt against your vault
The whole thing takes ~2-3 minutes and then the agent does its work, writes files, and the container
shuts down.
Error Messages/Logs
Steps to Reproduce
Steps to reproduce:
- Trigger a manual run on any scheduled remote agent
- Run enters "Setting up a cloud container" and never advances
- UI provides no way to cancel the stuck run
Expected behavior: Container provisions, repository clones, Claude Code starts, prompt executes.
Actual behavior: Stuck at container setup indefinitely. Multiple concurrent stuck runs accumulate with
no way to clear them.
Environment: Claude Code (Mac desktop app), remote agent via claude.ai
Claude Model
Sonnet (default)
Is this a regression?
Yes, this worked in a previous version
Last Working Version
_No response_
Claude Code Version
2.1.123
Platform
Anthropic API
Operating System
macOS
Terminal/Shell
Terminal.app (macOS)
Additional Information
_No response_
11 Comments
Found 2 possible duplicate issues:
This issue will be automatically closed as a duplicate in 3 days.
🤖 Generated with Claude Code
This is control-surface failure during remote execution setup.
If remote agent runs get stuck on “Setting up a cloud container” and cannot be canceled, the product is failing in two layers at once: startup never completes, and the operator loses the ability to abort the bad state.
Why it matters:
I would want clarity on:
Users can tolerate setup failures better than they can tolerate losing control over the failed run. If startup and cancellation both break, the control model is incomplete.
Don't believe I can cancel or force cancel a run through the UI, only option is to actually delete the entire routine itself. The moment I run something, whether it's a one time or recurring execution, it immediately goes into cloud container setup and runs indefinitely.
Also experiencing this issue. Filed a related report at #57280 with routine ID
trig_013nLmBUUXQxozoSTjaChL7h. Run has been stuck on container setup for 30+ minutes with no error and no cancel option.Experiencing this issue on trig_013DCKNcSTMzw1VobPAGPMBV as well right now.
Same for me in the code module on Claude Desktop on Mac... wanted to do a simple thing with an existing repo and just spinning on Setting Up a Cloud Container
Claude 1.6608.2 (ebf1a1) 2026-05-08T23:17:27.000Z
I'm currently experiencing this also on web (Chrome browser) based claude (Opus 4.6) session, since yesterday. I have other sessions that are not affected.
Same bug here on area:routines. Routine trig_01GiiTzmR4swPDGZZdg3mxPr (sources-enabled with git_repository clone). Three manual runs today (UI Run now and API POST /run) all hung at "Setting up a cloud container" for 1+ hour. Scheduled run at 02:05 BST today did get a container. Cannot cancel hung runs from UI. Claude Code Version 1.6608.2 (ebf1a1), Windows.
Scheduled run also hangs, not just manual. Trigger trig_01Aycx52J6UXT1HbgfS2odqc, session at 02:06 BST today, still Running after 16+ hours
Same problem with trig_01HB8GKYsUorP3GFePq6XKv1
Closing for now — inactive for too long. Please open a new issue if this is still relevant.