Buster Neece

@mnapoli looking into that issue, it looks like they did actually merge a fix for it into their 4.x branch, which is intended to support the new PHP 8.x features...

@mnapoli I'm worried that may be the only way to actually take advantage of PHP 8.0-level features inside the closures that are compiled, since they've closed PRs that would add...

@odan The `RequestResponseNamedArgs` controller invocation strategy is provided by, and part of, the Slim repo. While it isn't the default, it's certainly within the scope of Slim.

@odan Like I mentioned in the original issue, the type strictness of a call is determined by the _calling_ file, not the _called_ file, so removing strictness on my controller...

@iDJHost This may be an issue with Docker or with how your partitions are mounted in the VPS itself. Usually, Docker stores its own internal data in a directory similar...

@matthewunmuted Does this station have HLS streaming enabled? This may be something to do with that, as that involves storing the smaller chunks of HLS data locally, converting them for...

@Omnichannel-nz @matthewunmuted I'm not actually sure that there's anything we can do on our end to resolve this. I think there's a distinct possibility that in Chrome, the HTML5 ``...

@toots I can certainly try it, but this is applying to streams that _don't_ have live broadcasts, so that `to_live` branch shouldn't ever be executing at all. It's definitely using...

Following up on this, we've updated to the latest stable version and it appears the issue persists exactly as described. Mitigations include disabling smart crossfade or disabling crossfade altogether.

@toots Not sure that they're under the crossfade duration itself, but they're short enough that the "smart crossfade" doesn't have enough track data to analyze. That seems to be the...