Update: A workaround is now available that fixes the issue.
If you use Invidious to access YouTube videos, you may have noticed that most (all?) instances are broken right now. Attempts to play YouTube videos are met with error messages.
Messages such as “the video returned by YouTube isn’t the requested one” are thrown by Invidious instances at the time.
A bug report on the official GitHub repository confirms the issue. Good news is that the team is aware of the issue. Bad, that there is no fix yet to address it.
YouTube search works on all instances, but videos won’t play anymore. You can try any of the listed Invidious instances or others, and you will likely get the same result.
The issue was reported three days ago on the GitHub project site.
Another popular YouTube frontend, Piped, appears affected as well. Attempts to play videos are met with the error code “failed with error code 1002”.
Google seems to have implemented changes to YouTube that break video playback functionality of the frontends. It is too early to say whether this can be fixed at all.
Google seems to be tightening access to YouTube in an effort to increase revenue and lock out alternatives. Privacy conscious users like these alternatives, as they include no ads or tracking.
It is quite possible that Google is going to lock down YouTube further in the future. It is probably only a matter of time. It is likely that existing alternatives and new alternatives will see huge user jumps when that happens.
Now You: do you watch videos on YouTube? Did you run into these issues?
yt-dlp is also somewhat affected (WARNING for now):
> WARNING: [youtube] Skipping player responses from android clients (got player responses for video “aQvGIIdgFDM” instead of “BaW_jenozKc”)
https://github.com/yt-dlp/yt-dlp/issues/9554
Newpipe works well on my phone and tablet still, and that version is already old..so maybe this is just a PC thing.
I can confirm on my side that all Invidious instances I’ve tested are out of service.
Concerning Piped instances, all depends as it has always been the case of the instance and of the instance’s proxy (set by the user). Personally I encounter presently very few issues when I choose as by the past :
Instance, either of these :
[https://piped.adminforge.de/]
[https://piped.smnz.de/]
[https://piped.video/] (Cloudflare)
[https://piped.projectsegfau.lt/]
[https://piped.privacydev.net/] (the only to handle live streams)
Instance’s proxy (Instance / Preferences / Instance (=proxy server used by the instance) :
[adminforge.de]= [https://pipedapi.adminforge.de]
[smnz.de] = [https://pipedapi.smnz.de]
[kavin.rocks] = [https://pipedapi.kavin.rocks]
At this time, a few hiccups but not many. Occasionally mainly with latest videos.
Sorry for you, Google, but [https://yewtu.be/] is functional again at this time : 2024-04-07, 09:00 GMT
Seems to be fixed for now.
[https://github.com/iv-org/invidious/pull/4552]