Hmm, looks like they are 5 hours behind (or really, 19 hours ahead). But you are right, the language settings should have no impact on local time. I will email for some more debugging.

Thanks - mine is macOS 11.5.2 with Chrome 93, so slightly different, so maybe the Chrome 94 beta is changing some things?

Not exactly related but suggests something is changing:

https://www.reddit.com/r/chrome/comments/ocxicd/new_chrome_update_has_a_bug_with_youtube/

We do all our internals in GMT/UTC so if the sum of time zone and wall-clock time is correct, it should keep working.

Here is a tester script to compare with our server:
https://justgetflux.com/timesync.html

So yes, if "UTC time" in the OS is wrong, we will transition at the wrong time.

We only use tz changes as a cue to adjust location (like for traveling between timezones), but cancelling this dialog is just fine too.

We don't know of any updates in High Sierra that should affect this - did you add any hardware/external displays or anything like that.

Usually this means you need to set your location from the f.lux preferences menu.

I believe Linux has the same problem. I don't know if f.lux has a Linux version or not, but I use a similar program called redshift, and it has this mouse pointer issue. The little bit of research I did indicated that it is a very difficult problem to fix, so I just live with it. A minor annoyance.

Also started flickering here, not sure if related but quite recently upgraded to macOS 11.4.

Macbook pro 13" 2017
External monitor: Eizo EV2436W

Both monitors, the macbook one and the external monitor are flickering synchronously.

We ask the system about the installed profile for each display and try to use the profile correctly. So in our tests here, with a Thunderbolt display (standard gamut) and a Macbook (wide gamut) the two look identical, because we use different transforms for each display.

Is it possible that the external display is actually wide gamut, but macOS does not know about this? If this were true you may see other apps displaying colors that were extra saturated, and the fix would be to install a good ICC profile for the external display.

@ralvydc Just tested Gradual Transition and it appears to start the Transition 1 hour before the target time. So if my Bedtime is 9pm (9 hours before my WakeUp time of 6am), with Gradual Transitions set, f.lux starts decreasing blue light at about 8pm, taking 1 hour to get to my Bedtime setting. Perfect.

Thanks for the quick support.

[EDIT/UPDATE]
Confirming it is the company's security software killing f.lux, nothing on f.lux's end. Appreciate your looking into it! This thread can be closed, or removed.