Quote:
Originally Posted by erland
![View Post]()
It's probably a good idea to start a new thread if we need to discuss this further.
But for the duplicate musicbrainz ID's the reason is likely the changed principle Musicbrainz did 2010 when they started to use the same musicbrainz id on a track on a normal album and the same track on a compilation album. Unfortunately this breaks the possibility for TrackStat to uniquely identify music files with musicbrainz id, the only solution to get rid of the problem is to deactivate the musicbrainz support in TrackStat through the "Enable musicbrainz tags" option in TrackStat settings.
If you have duplicate musicbrainz id's the number of duplicates are going to increase every time TrackStat does a refresh, this happens at startup and after rescan unless you have disabled the refresh functionality. If you disbable musicbrainz tag functionality it means that if you rename/move files TrackStat won't be able to keep the statistics for them. To get rid of the existing duplicates, unfortunately the only solution is to:
1. Take a backup of TrackStat data through TrackStat settings page
2. As an extra safety, you can also shutdown SBS/LMS and take a copy of the persist.db file in the SBS/LMS Cache directory
3. In TrackStat settings page select to "Remove all data" and then specify the full path to the backup file taken in point 1 and use the "Restore from File" functionality. The restore operation will make sure to only import one instance with the same musicbrainz id, so this will get rid of the duplicates. Remember that the import runs in backgroud so you need to look in the server.log file to see when it's completely finished. The "Find alternative files" option generally makes the restore process a lot slower, so unless you need the restore to try to relocate files which you have moved previously which didn't have musicbrainz tags I would suggest disabling "Find alternative files" option. Please note that in a large library with "Find alternative files" the restore can easily take many hours. In my own library with about 4000 tracks it just takes few minutes and is significantly faster than that if I disable "Find alternative files" option.
If you get warning about "Unused " entries, this is typically statistics pointing to files that either have been removed or moved/renamed and didn't have musicbrainz tags. If you just want to remove these there is a "Delete unused statistics" button in TrackStat settings page. I have not tried this in a large library but I suspect it could take some time to execute and while doing so I think the server will be locked, so you probably don't want to do this when you are about to listen to some music. Generally the unused entries doesn't cause any problem unless you have a lot of them so it makes the database a lot larger than it needs to be.
Feel free to reply in this thread if you just have a quick question, but please use a separate thread if you feel that you have a lot of questions.
But for the duplicate musicbrainz ID's the reason is likely the changed principle Musicbrainz did 2010 when they started to use the same musicbrainz id on a track on a normal album and the same track on a compilation album. Unfortunately this breaks the possibility for TrackStat to uniquely identify music files with musicbrainz id, the only solution to get rid of the problem is to deactivate the musicbrainz support in TrackStat through the "Enable musicbrainz tags" option in TrackStat settings.
If you have duplicate musicbrainz id's the number of duplicates are going to increase every time TrackStat does a refresh, this happens at startup and after rescan unless you have disabled the refresh functionality. If you disbable musicbrainz tag functionality it means that if you rename/move files TrackStat won't be able to keep the statistics for them. To get rid of the existing duplicates, unfortunately the only solution is to:
1. Take a backup of TrackStat data through TrackStat settings page
2. As an extra safety, you can also shutdown SBS/LMS and take a copy of the persist.db file in the SBS/LMS Cache directory
3. In TrackStat settings page select to "Remove all data" and then specify the full path to the backup file taken in point 1 and use the "Restore from File" functionality. The restore operation will make sure to only import one instance with the same musicbrainz id, so this will get rid of the duplicates. Remember that the import runs in backgroud so you need to look in the server.log file to see when it's completely finished. The "Find alternative files" option generally makes the restore process a lot slower, so unless you need the restore to try to relocate files which you have moved previously which didn't have musicbrainz tags I would suggest disabling "Find alternative files" option. Please note that in a large library with "Find alternative files" the restore can easily take many hours. In my own library with about 4000 tracks it just takes few minutes and is significantly faster than that if I disable "Find alternative files" option.
If you get warning about "Unused " entries, this is typically statistics pointing to files that either have been removed or moved/renamed and didn't have musicbrainz tags. If you just want to remove these there is a "Delete unused statistics" button in TrackStat settings page. I have not tried this in a large library but I suspect it could take some time to execute and while doing so I think the server will be locked, so you probably don't want to do this when you are about to listen to some music. Generally the unused entries doesn't cause any problem unless you have a lot of them so it makes the database a lot larger than it needs to be.
Feel free to reply in this thread if you just have a quick question, but please use a separate thread if you feel that you have a lot of questions.
(I only included the lines with possible problems.)
Quote:
Originally Posted by Custom Scan Inconsistency/Problems
Problem type Possible problem Number of
Not refreshed entries in Custom Scan song data Yes 2022
Songs with long urls (Longer than 255 and shorter than 511 characters) Yes 11
Duplicate albums with same title without musicbrainz tags Yes 42
Not refreshed entries in Custom Scan song data Yes 2022
Songs with long urls (Longer than 255 and shorter than 511 characters) Yes 11
Duplicate albums with same title without musicbrainz tags Yes 42
The Musicbrainz IDs are still in my files. Is that what this is reporting? If so, this is likely benign and there's nothing to do (except to remove the Musicbrainz IDs).
Quote:
Originally Posted by Squeezebox Server Inconsistency/Problems
Problem type Possible problem Number of
Duplicate musicbrainz song tags Yes 919
Duplicate musicbrainz album tags Yes 4
Duplicate musicbrainz artist tags Yes 2
Duplicate musicbrainz song tags Yes 919
Duplicate musicbrainz album tags Yes 4
Duplicate musicbrainz artist tags Yes 2
What was curious about the results here is that the "Not refreshed entries" are not clickable links. Is that by design?
Quote:
Originally Posted by TrackStat Inconsistency/Problems
Problem type Possible problem Number of
Not refreshed entries in TrackStat Yes 1004
Not refreshed entries in TrackStat history Yes 1374
Songs with long urls (Longer than 255 and shorter than 511 characters) Yes 11
Not refreshed entries in TrackStat Yes 1004
Not refreshed entries in TrackStat history Yes 1374
Songs with long urls (Longer than 255 and shorter than 511 characters) Yes 11