system/os: UnRaid 6.2
Plex web: 2.8.1 / 2.9.3 (via Plex.tv interface)
Plex server: 1.1.4.2757
So, this is a bug that I’ve noticed before but did not report as I assumed it would be corrected in subsequent updated. It obviously hasn’t been so far.
The first part of it occurs when a new item is added to a media library (music). Most items will have the plex mix option available immediately upon library refresh, even when the plex mix itself for is not available for the item. This is not desired behaviour as the option should only be available when the song has a plex mix.
An additional issue is that the error message associated with the plex mix availability is inconsistently shown in the web interface.
If you attempt to trigger a plex mix directly from the album/song list (the additional options under “…” on individual songs) you will only get a partial player interface at the bottom of the screen, with a loading animation. The API request to playQueues correctly responds with “Sorry, there is no Plex Mix available for this track.” message but the exception is obviously not handled correctly.
Additional for devs, your Sentry error calls are not useful in this instance, due to constant rate limiting (429 response)
If the same song is already playing and you attempt to us the “Queue Plex Mix” from the player, the error message displays normally.
Now I’m quite aware that indexing additions to a large library (in my case 3.5K albums) is time-consuming and may take a while as a background process, so my complaint is not related to speed, but rather to the UX/API response exception handling issue presented in my notes above.
A completely separate issue is that I’ve looked at albums added 30 days ago and there are some that exhibit the same behavior (while others have normal Plex Mix function). This seems like a very long period so I’m wondering what is the recommended fix in those cases (re-indexing?).
Let me know if I need to attach logs.
UPDATE:
- Plex server log attached
I’ve attached the log to the original post. And this issue has been reproduced numerous times on my end, with server restarts, etc.