On Nov 24, 2025, at 6:53 AM, Eric Covener <[email protected]> wrote:
> I've had two relatively recent and relatively painful investigations > come down to this closure of keepalive connections on activity when an > individual child process has no idle workers: How do browsers handle this situation? I'm wondering if this is the cause of very rare connection failures I've seen but have never been able to duplicate. > Does it warrant a WARNING level message? It is nice to at least get > the connection details into the log. I think this would be useful so people could see how often it happens. If it's never happening in practice, the warning would be sufficient, but if people see the warning often, a better fix with blocking or queuing could then be considered. -- Robert L Mathews
