Not clear why the full library rescan was triggered – all I did was add an album from TIDAL. Is the rescan supposed to be so resource intensive that it causes music playback to stall and skip tracks?
Thanks for the detailed report, we were able to review a fresh set of Roon Server logs which gave us a clear picture of whats going on.
The rescan you’re seeing is your TIDAL library re-syncing. Roon re-imports your whole cached TIDAL content whenever your TIDAL favorites change, so adding or removing a single favorite album causes it to re-process everything rather than just the one album. In your logs we can see two of these fire seven minutes apart on 27 August, at 3:31 and 3:39 pm your time, right in the middle of a listening session.
On its own that would just be housekeeping. The problem is that Roon Server’s memory use has been climbing steadily. Your server has been running about ten days without a restart, and over that time memory has gone from roughly 3 GB to 6.7 GB. At that level there’s very little room left to work in, and when a re-import lands on top of it the server briefly stops responding, in the worst cases for 10 to 16 seconds at a stretch. That’s what your endpoints experience as a dropout: no audio is being sent during those pauses, the buffer empties, and Roon stops the stream. All seven dropouts in your logs line up exactly with one of these pauses.
Two things that would help:
Please restart Roon Server. That clears the accumulated memory back down to around 1 GB and should stop the dropouts straight away. It’s also the cleanest way to confirm what we’re seeing, if playback is clean afterwards and then gradually degrades again over the following week, that tells us what we need to know.
In the meantime, batching up your TIDAL favoriting outside of listening hours will remove the trigger. That’s a workaround rather than a fix, but it should buy you some quiet in the short term.
I’ve flagged the memory-growth pattern with our team, a routine background metadata job on 20 August accounts for a large step up that never came back down, and that’s the part we want to look at more closely.
If it happens again after the restart, send us a fresh set of logs along with the time it occurred and we’ll take another look.
At your convenience, please restart Roon Server two times in a row to activate some additional verbose logging we have enabled that will help capture what’s happening here for our engineers. You can use Roon normally after the restarts; you shouldn’t need to do anything other than restart twice to enable this logging.
Thank you for completing those restarts. The verbose logging is now active on your Nucleus Titan.
To capture the specific data our engineering team needs, we now need you to reproduce the exact event while this logging is running. When you have a moment, please follow these steps:
Start playing music to your setup as you normally would.
Add a new album or track from TIDAL to your favorites to trigger the library re-sync.
Keep listening until you experience the playback dropout or skip.
Once the dropout occurs, please reply to this thread with the specific local time it happened. We will immediately pull the diagnostic logs containing the verbose data and attach them to the developer ticket for analysis.