Hey @dgeorgakopoulos,
Thank you for the detailed timestamps and screenshots, and for the clear summary. We activated diagnostics and went through your Roon Server logs around the times you reported, and we have a much clearer picture now. We also want to correct the direction of our earlier replies, because the logs point somewhere different.
First, you were right about one thing: since the interruptions happen on both USB and optical TOSLINK, and only when Roon is in the chain, this is not about the NODE’s USB port or its digital output. We can set that question aside.
What the logs actually show:
There are two different ways you have been reaching the NODE, and they fail for two different reasons.
When you play to the NODE as a native Roon Ready endpoint (the “direct” zone, shown in your logs as “Foyer”), Roon and the NODE have to agree on a shared clock so audio stays sample-accurate. In your logs, that clock handshake is failing badly. Every sync between Roon Server and the NODE shows the NODE’s reported clock drifting by roughly a million parts per million, where a healthy endpoint drifts by only a handful. A few seconds into your local-file test at 15:30 on June 30, this built up until the NODE reported a dropout and Roon stopped and tore the stream down. That tear-down loop is what you hear as stops and noise on the direct connection. Importantly, we see the exact same broken clock pattern on every direct-playback attempt in the logs, including ones from June 27, so this is consistent rather than a one-off.
When you play via AirPlay 2 (the “Foyer earphones” zone), the clock problem goes away because AirPlay handles its own timing, which is why AirPlay plays. But we do see the NODE repeatedly renegotiating its AirPlay session during the day, which lines up with the occasional clicks you get on that path.
So the common factor is not Roon’s playback engine, which has not changed, but the timing handshake between Roon and this particular NODE, at the clock level. The most likely thing that changed around the time you noticed this is a BluOS firmware update on the NODE, since that is the part producing the clock values Roon is rejecting.
What we would like you to try, in order
- Fully power-cycle the NODE. Unplug it for about 30 seconds rather than using standby, then power it back up. A corrupted clock state on the endpoint often only clears with a cold restart, so please try this first.
- Check the BluOS version on the NODE and when it last updated. In the BluOS app, look under the player's settings for the firmware version and update history. If you can tell us when it last updated relative to when the direct connection stopped working, that timeline is genuinely useful to us.
- Run one test with the NODE on a wired Ethernet connection instead of Wi-Fi. We are seeing some long round-trip times to the NODE, so this cleanly separates a network issue from a clock issue. If the direct path still breaks up on Ethernet, that confirms it is on the NODE's side and we would loop in Bluesound at that point.
- Re-add the NODE in Roon, as you suggested. Go to Settings, then Audio, disable the NODE, then enable it again so Roon rebuilds the device and reruns the handshake. This is quick and worth doing, though it will not help if the firmware clock is the underlying cause.
After the cold reboot, please try the direct (non-AirPlay) zone again, and if it still breaks up, give us the exact local time and the track. A timestamp from the direct connection specifically, rather than AirPlay, is the most useful thing you can send us next.
Looking forward to your results. 