Investigating Chrome Sleep/Resume Failures on My M3 MacBook Air

This has happened countless of times, the version I recorded was 142.0.7444.162 (Official Build) (arm64). WhatsApp fails to open at all when this happens and requires a full Chrome quit and relaunch. Annoying if you have over 100 tabs open at any one time. Facebook also fails to load messages or go to groups when this bug is present. Flagging with this with Google but I hope it already has been. It's no surprise when Apple basically told everyone overnight to use ARM (#mseries) or die. lol Google Chrome

Clearing WhatsApp or Website cache/cookies/site data does not solve it. a full Chrome restart is required each time it happens. I suspected a memory-management issue, but did not establish a root cause. Never had it happen on an Intel Mac.

Experience this since last year from memory when I got my M3 MacBook Air, 16GB RAM, 512GB SSD. Apple

Separately, I observed macOS freezes while using Chrome with low storage and heavy swap usage. I did not isolate the cause. The only way out of that is to do a full macOS force power off and turn on again. Has happened many times also.

What this investigation establishes

On my setup, repeated sleep/resume cycles were followed by failures in WhatsApp and parts of Facebook. Clearing site data did not solve the problem; quitting and restarting Chrome did. I had not seen the same behaviour on my Intel Mac. Those are observations, not proof of an ARM64 or V8 defect.

What I would test next

The original import included lengthy ChatGPT and Gemini diagnoses. Those were hypotheses, not test results, and I have removed the unsupported conclusions. A more useful follow-up would record:

• The exact Chrome and macOS versions, available storage and tab count when the failure occurs.
• Whether it reproduces with a fresh profile and extensions disabled.
• Whether the same sleep/resume sequence affects another browser.
• Console and network errors before restarting, with private account information removed.
• Whether changing one condition at a time alters the result.

These are proposed checks, not tests I am claiming to have completed. A restart was my workaround; I do not have a confirmed upstream fix to report.