Hey All, we’ve been working on this continuously (+/- a bit of sleep) since reproducing it yesterday, and have narrowed this down to an interaction between Microsoft’s .NET 6.0 framework and some of TIDAL’s servers…except these are not really TIDAL’s servers–it’s actually Amazon CloudFront, which is the content distribution network that TIDAL uses.
CloudFront is comprised of thousands of servers all over the world. They do not all get software updates at the same time–Amazon, like all large providers, deploys code on a rolling basis. Perhaps starting with some parts of the world then going broader, or rolling back if there are issues.
We have been reproducing this issue by using a VPN to virtually transport ourselves to Montreal, since apparently there are some “bad” servers there.
Then, in the past hour or so, our ability to reproduce has become flaky then evaporated. Almost as if Amazon rolled out some new code that fixed this issue this morning. At least, that’s how it looks to us.
Because of rolling updates, it’s possible that there are still “bad” servers out there. This is not our infrastructure so we don’t have the details, so we’re not calling this fixed yet.
Nonetheless, I think it would be useful to get a read on how you guys are doing.
If you have a minute, please try this:
- Reboot the machine that runs Roon
- Play some music and see if this is still happening.
Afterwards, I’d like to know:
- Where you are located
- If you are still seeing the problem, or if it has gone away
This will give us a sense of what is happening and how this is changing. If there are still parts of the world where this is consistently bad, we may be able to reproduce it again using another location. Unfortunately, the only real concentration we saw in the data was in Canada, and those avenues seem to be closed now.