@outer locust Still seeing some double scrobbles?
#Double scrobble
rugged cosmos
outer locust
That's a lot of time in between, are you sure it only played once?
rugged cosmos
I was wondering if it was scrobbled once at the start and at the end. Let me do some more testing right now and see what happens.
There were a lot BTW. I just showed one
rugged cosmos
That's a lot of time in between, are you sure it only played once?
Just watched this carefully
2025-04-20 00:05:23.778 DEBUG (MainThread) [music_assistant.player_queues] PlayerQueue Study Player loaded item ff54af94c14541cd8c59b8265270d21f in buffer 2025-04-20 00:05:24.279 DEBUG (MainThread) [music_assistant.player_queues] (pre)loading (next) item for queue Study Player... 2025-04-20 00:05:24.816 DEBUG (MainThread) [music_assistant.player_queues] Enqueued next track Fleetwood Mac - Rhiannon (Live) on queue Study Player 2025-04-20 00:06:04.227 INFO (ThreadPoolExecutor-0_5) [pylast] track.updateNowPlaying 2025-04-20 00:09:55.660 INFO (ThreadPoolExecutor-0_3) [pylast] track.scrobble 2025-04-20 00:09:55.855 DEBUG (MainThread) [music_assistant.player_queues] PlayerQueue Study Player loaded item 518971449e464d4ca2920e3d9c23c758 in buffer 2025-04-20 00:09:56.357 DEBUG (MainThread) [music_assistant.player_queues] (pre)loading (next) item for queue Study Player... 2025-04-20 00:09:56.914 DEBUG (MainThread) [music_assistant.player_queues] Enqueued next track Fleetwood Mac - Brown Eyes (Live) on queue Study Player 2025-04-20 00:10:05.213 INFO (ThreadPoolExecutor-0_5) [pylast] track.updateNowPlaying 2025-04-20 00:17:00.018 INFO (ThreadPoolExecutor-0_2) [pylast] track.scrobble 2025-04-20 00:17:00.605 DEBUG (MainThread) [music_assistant.player_queues] PlayerQueue Study Player loaded item ba5d2db46bdd4f71868d27ecf08d487b in buffer 2025-04-20 00:17:01.110 DEBUG (MainThread) [music_assistant.player_queues] (pre)loading (next) item for queue Study Player... 2025-04-20 00:17:01.669 DEBUG (MainThread) [music_assistant.player_queues] Enqueued next track Fleetwood Mac - Eyes of the World (Live) on queue Study Player 2025-04-20 00:17:06.875 INFO (ThreadPoolExecutor-0_14) [pylast] track.updateNowPlaying 2025-04-20 00:17:07.132 INFO (ThreadPoolExecutor-0_10) [pylast] track.scrobble 2025-04-20 00:17:40.660 INFO (ThreadPoolExecutor-0_13) [pylast] track.updateNowPlaying 2025-04-20 00:21:20.662 INFO (ThreadPoolExecutor-0_4) [pylast] track.scrobble
2025-04-20 00:21:21.352 DEBUG (MainThread) [music_assistant.player_queues] PlayerQueue Study Player loaded item cc41857164ec4d46afde7c4f4af80fdf in buffer 2025-04-20 00:21:21.854 DEBUG (MainThread) [music_assistant.player_queues] (pre)loading (next) item for queue Study Player... 2025-04-20 00:21:22.390 DEBUG (MainThread) [music_assistant.player_queues] Enqueued next track Fleetwood Mac - Gypsy (Live) on queue Study Player 2025-04-20 00:22:01.378 INFO (ThreadPoolExecutor-0_7) [pylast] track.updateNowPlaying 2025-04-20 00:25:27.436 INFO (ThreadPoolExecutor-0_12) [pylast] track.scrobble 2025-04-20 00:25:27.885 DEBUG (MainThread) [music_assistant.player_queues] PlayerQueue Study Player loaded item dab72d6e7c11402e9835c96c80ff2ab2 in buffer 2025-04-20 00:25:28.386 DEBUG (MainThread) [music_assistant.player_queues] (pre)loading (next) item for queue Study Player... 2025-04-20 00:25:28.942 DEBUG (MainThread) [music_assistant.player_queues] Enqueued next track Fleetwood Mac - Love in Store (Live) on queue Study Player 2025-04-20 00:25:31.375 INFO (ThreadPoolExecutor-0_1) [pylast] track.updateNowPlaying
I'm trying to get a screencap of it but I saw a scrobble right at the start of playback and then the now playing indication of the same track
Like this
outer locust
At the start of the playback? 
that's very strange, I'll have to experiment to see if I can reproduce it. Can you share the exact length of those two songs that got double scrobbled?
rugged cosmos
At the start of the playback? <:Hmm:961647351389388920> that's very strange, I'l...
OK I just noticed that sometimes on a track transition the progress bar moves fully up and back so I was wondering if you are watching that media position attribute then that might be a problem. I am using a squeezelite player and I have an open issue for that not working properly...?
outer locust
Yeah that would do it 
In short: the queue logic looks at that and sets a "fully_played" property on an event that the scrobble integrations listen to
I can't guarantee that that's the issue, but it's definitely possible
rugged cosmos
I have added a link to this conversation to the issue since it might be related to that problem I observed although the issue is about the HA entity attribute not being updated correctly on track transition. I will try a different player type and see if it happens with that.
outer locust
I don't know enough of the player differences to say if that could be the direct cause. It might be just a queue issue
rugged cosmos
In short: the queue logic looks at that and sets a "fully_played" property on an...
Interestingly though sometimes when I see that double entry where it says scrobbling now it doesn't always result in a double entry but I think that might be when the next track is longer. So the position stays at the length of the previous track and if that is longer than the new track it scrobbles and if it is shorter then it doesnt!
outer locust
I noticed the "pre-loading next track" in your logs, that's the thing I want to look at, as I don't know if I ever noticed that before
quiet belfry
pre-loading has nothing to do with scrobbling
That is just the queue that loads the streamdetails for the next track in the queue
rugged cosmos
I noticed the "pre-loading next track" in your logs, that's the thing I want to ...
I played the same album tonight on a cast device and no double scrobbles so I am leaning towards the squeezelite player triggering this as discussed
fossil crescent
FYI, I have also run into this problem when using squeezelite.
The weird thing that I see is that the MA UI updates to show the next track when the pre-loading starts (not when the track actually switches), but the player progress bar continues to increment as it plays the old track. If the previous track was longer than the next track it'll even indicate a time which is past the end of the track.
Then when playback of the next track actually starts, the player progress bar will jump back to the start of the track.
I think the extra scrobble at the start of a track is coming from this case - after the MA is showing the track has switched but before squeezelite has actually started playing the new track, when the position is displaying at, near, or even past the end of track.
I typically see the MA ui display the next track about 10 seconds before squeezelite actually starts playing the new track, and during this time MA is showing the track details of the next track with the position from the previous track.
rugged cosmos
We know what the problem is it just has to be fixed.
quiet belfry
Is there an existing bug report for that ?