[BUG] Claude Code VS Code extension causes kernel panics (watchdogd timeout/Jetsam OOM) on MacBook Pro M3 Pro with 18 GB RAM
Preflight Checklist
- [x] 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?
Running multiple Claude Code instances in VS Code causes macOS kernel panics due to memory exhaustion on MacBook Pro M3 Pro with 18 GB RAM. The system reboots without warning — no error dialog, no chance to save work.
Two distinct panic types observed:
Jetsam OOM kill of watchdogd — memory pressure causes macOS to kill the critical watchdogd process, triggering an immediate kernel panic (exit reason namespace 2, subcode 0xa).
Watchdog timeout — watchdogd fails to check in for 90+ seconds due to system-wide resource starvation.
The problem only occurs when a project folder is opened in VS Code with Claude Code active. VS Code with Claude Code but without an open project works stably. The system is completely stable when Claude Code is not running. The problem started after updating Claude Code to the latest version. Updating macOS to 26.4 did not resolve the issue.
What Should Happen?
Claude Code should manage its memory consumption to avoid exhausting system RAM. Multiple instances should coexist without causing system instability. At minimum, Claude Code processes should gracefully reduce memory usage or exit before reaching levels that trigger macOS Jetsam kills of critical system processes like watchdogd.
Error Messages/Logs
Panic Log 1 — Jetsam OOM Kill:
panic(cpu 1 caller 0xfffffe0042a6c3ac): watchdogd[43124] exited
-- exit reason namespace 2 subcode 0xa description: none
Compressor Info: 11% of compressed pages limit (OK) and 23% of segments limit (OK)
with 4 swapfiles and OK swap space
Panicked task 0xfffffe1983557968: 200 pages, 1 threads: pid 43124: watchdogd
4 swapfiles indicate the system was already heavily swapping before the crash.
Panic Log 2 — Watchdog Timeout:
panic(cpu 0 caller 0xfffffe003923fcfc): watchdog timeout:
no checkins from watchdogd in 90 seconds
(496 total checkins since monitoring last enabled)
Compressor Info: 5% of compressed pages limit (OK) and 13% of segments limit (OK)
with 1 swapfiles and OK swap space
Kernel Extensions in backtrace:
com.apple.driver.AppleInterruptControllerV3(1.0d1)
com.apple.driver.AppleARMWatchdogTimer(1.0)
com.apple.driver.AppleT6030CLPC(1.0)
Panicked task: pid 0: kernel_task
This panic occurred ~80 minutes after boot with no sleep cycles. AppleT6030CLPC (M3 Pro performance controller) is in the backtrace.
Activity Monitor observation during normal operation:
kernel_task consuming 43% CPU, WindowServer 20.7%, logioptionsplus_updater 20%. All temperature sensors normal (CPU 49–56°C, GPU 49°C, SSD 39°C) — confirmed via Macs Fan Control 1.5.20, overheating ruled out.
Steps to Reproduce
Open VS Code on MacBook Pro M3 Pro (18 GB RAM) with 2 external monitors connected via HDMI
Open a project folder in VS Code (important — without an open folder the extension works stably)
Launch 3 Claude Code instances in separate VS Code windows, each with its own project
Work normally — coding, reviewing, chatting with Claude
After some time (varies from 15 minutes to a few hours), macOS kernel panics and reboots without warning
Additional observations:
VS Code with Claude Code installed but without an open project folder — stable
All apps running except Claude Code — stable
Claude Code running with open projects — kernel panics reproduce consistently over several days
Claude Model
Opus
Is this a regression?
Yes, this worked in a previous version
Last Working Version
_No response_
Claude Code Version
2.1.84
Platform
Anthropic API
Operating System
macOS
Terminal/Shell
VS Code integrated terminal
Additional Information
_No response_
This issue has 2 comments on GitHub. Read the full discussion on GitHub ↗