claude agents: daemon and bg-spare processes spin at ~100% CPU when idle
Status Fixed / completed
Reported on v2.1.153
Maintainer reply ✓ Yes — bogini
Workaround ✓ Mentioned in description ↑
Activity 7 comments · opened May 28, 2026 · closed May 28, 2026
💡 Likely answer: A maintainer (bogini, collaborator)
responded on this thread — see the highlighted reply below.
Description
After running claude agents, the daemon and background worker processes spin at ~100% CPU per process even when there are no active tasks. The processes never exit on their own.
Steps to Reproduce
- Run
claude agentsin a terminal - Let it sit idle (no tasks running)
- Open Activity Monitor (macOS) or
top
Observed Behavior
Multiple claude.exe processes each consuming ~99–100% CPU:
claude agents PID 1203 ~100% CPU
├── daemon run PID 1212 ~100% CPU
│ ├── bg-pty-host PID 1226 ~100% CPU
│ └── bg-pty-host PID 1229 ~100% CPU
└── bg-spare PID 1267 ~100% CPUTotal CPU usage: ~500% (5 cores pegged) while completely idle.
Expected Behavior
Idle processes should block on socket I/O and consume ~0% CPU.
Root Cause (suspected)
The Node.js child processes appear to be busy-waiting on Unix sockets instead of using blocking I/O or an event loop sleep. This looks like a missing await or an unintentional tight polling loop in the PTY host / spare process code.
Workaround
pkill -f 'claude agents'
pkill -f 'claude\.exe.*(daemon|bg-)'Environment
- Claude Code version: 2.1.153
- OS: macOS 15 (Darwin 24.6.0)
- Node.js: v22.22.3
7 Comments
Found 2 possible duplicate issues:
This issue will be automatically closed as a duplicate in 3 days.
🤖 Generated with Claude Code
Addressed by a merged fix. Please reopen with a fresh repro if you still see this on a current version.
▎ This issue was closed as fixed, but I'm still experiencing it on v2.1.154 (the latest version as of 2026-05-28).
▎
▎ Environment:
▎ - Claude Code version: 2.1.154
▎ - OS: macOS Darwin 24.6.0
▎
▎ Steps to reproduce:
▎ 1. Run claude agents in a terminal
▎ 2. Let it sit idle
▎ 3. Observe claude.exe bg-spare/bg-pty-host processes spinning at ~99% CPU each (3 processes × ~99% CPU)
▎
▎ Observed:
▎ Three claude.exe processes (PIDs vary) consuming ~99.4%, 99.4%, 99.3% CPU while idle. Only solution is to kill them manually.
▎
▎ Could this be reopened or tracked as a regression?
Still occurring on v2.1.161 (2026-06-03).
Environment:
Observed: A
--bg-spareprocess (PID 80814) ran from the previous evening with ~281 min CPU time, pinning one core at 100% CPU. The process survived daemon restart and was not cleaned up byclaude-cleanup. Requiredkill -9to terminate.This appears to be a regression that has persisted from 153 → 154 → 161.
Could a maintainer please reopen this issue? The bug is still present in v2.1.161.
Full environment details:
The
--bg-spareorphan process from a previous session ran overnight (~281 min CPU time) pinning one core at 100% CPU.claude-cleanupdid not remove it — onlykill -9worked.Additional observation that may help narrow down the root cause:
The 100% CPU spin only occurs when the spare processes are idle (no active sessions). When a session is actively running — e.g. an agent task is in progress — the CPU usage drops to normal levels. As soon as the session goes idle again, the spin resumes.
This is consistent with a busy-wait loop in the spare process idle path: real I/O events from an active session drive the event loop normally, but when waiting for a new connection the process falls into a tight poll loop instead of blocking on the socket.
This issue has been automatically locked since it was closed and has not had any activity for 7 days. If you're experiencing a similar issue, please file a new issue and reference this one if it's relevant.