Memory leak

[iandunn]

Continuing from #6

I got a spot script... warning from Firebox, and high CPU/memory usage again this morning. That's while running running master @ 4c8a092 (after #11 was merged) and hitting the local REST API every 3 seconds (instead of 60), but with service worker caching enabled, and after 10+ hours.

The browser still seems relatively responsive, though, except in the devtools Network tab, which users won't normally be accessing anyway. If anyone else is seeing problems and has time to test, that might be nice, but it might also be fine to go ahead and see what happens during WCEU. If there aren't any reports of problems, then it's probably not worth digging into further.

One thing worth noting, though, is that phones have significantly less CPU/memory than desktops, but will probably also cycle things out of memory more aggressively, especially when the display is off or the browser isn't the active app. Tablets may be the most vulnerable to this kind of thing, since users are more likely to keep them open throughout the day, but they're still relatively underpowered when compared to desktops.

[iandunn]

Has anyone heard reports of this in the wild? If not, maybe the service worker caching and/or typical usage patterns makes it a non-issue.

Cutting down on the size of the API responses (#6) might also help.

If it's not causing tangible problems, then it's probably not worth spending any more time on, and can be closed. If there are reports of it, then we should move this issue to https://github.com/WordPress/wordcamp.org/

@avillegasn , @vdwijngaert , @vedanshujain, @westonruter, @iceablemedia , @nem-c , @tfrommen

[avillegasn]

I didn't heard any reports about memory leaks.