Rob Landers

> Join us at static-php-cli and you can enjoy it daily ;) I would love to steal an hour of your time and learn from you, if you have it...

I’m not sure this is actually possible. I’ve taken two different approaches to varying degrees of success. Core problems: - CLI has PHP functions that are only available in the...

I tried that as well, but then you get symbol conflicts with libphp. (that was the second approach that I just now realized I forgot to write about in the...

@henderkes here's [my changes](https://github.com/php/php-src/compare/master...bottledcode:php-src:refactor/cli) -- I should add that I've also tried just building libphp by just including the new C files ... but that doesn't work either. I build...

Most likely this is a metrics reporting issue. Note that if you are used to working in Linux, frankenphp's timeouts are based on wall clock time, not execution time. If...

Interesting solution! I'm not sure this is an actual problem though. With thread autoscaling and the fact that caddy will hold requests until a thread is actually available, I'm not...

If all threads are consumed, should it keep serving traffic? I feel like this would just mask issues. For example, it may continue to respond to `/healthz` saying everything is...

I appreciate your take on this, no one wants the whole server to crash on a single failure. I think our difference is in _how_ we want the system to...

> The health check timing out will just cause containers to restart over and over (bad time, been there). Hmmm. I’m not sure what you mean. k8s have both liveness...

> The use case would be the one of the [hanging requests](https://github.com/php/frankenphp/blob/main/testdata/performance/hanging-requests.js), where an endpoint receiving a slow 15s request just 2% of the time will eventually hang up the...