Scan media library stuck at 89.1% & Metadata not updating

[FlowCode 1]

hi,

 

so I rearranged my files and folders of my movies, but when I ran emby server, I can't complete the scan library or update the metadata (only a couple have metadata either because an existing metadata in their folder).

I ran the scan and refresh metadata for continuous 48 hours without any result. Not a single metadata was added, and scan library never passed 89.1%

I tried vacuum option and restarted emby, but it opened normally without any progress or indication if the vacuum is in progress or done.

 

Kindly your usual support, what can I do to solve this.

is re-installing would help?

 

scan library stuck at 89.1% 

attached is the last log

embyserver.txt

[ebr 16185]

Hi.  What is your "F:" drive?  Looks like it is unavailable.

[FlowCode 1]

yeah its another HDD which I already removed from my library yesterday.

I figured this might be the issue, but even after I removed it from my libraries, still same issue nothing changed.

stuck at 89.1% for scan, and metadata is not updating

[ebr 16185]

Is the logfile changing when you think it is "stuck"?

[FlowCode 1]

yes.

 

Ive seen multiple entries:

LibraryMonitor: Watching directory followed by the folders included in my library.

 

but nothing changes, still stuck as we speak and no metadata got updated

[Luke 42081]

Hi, have you added any photo libraries?

[FlowCode 1]

hi Luke,

No I haven't,

However, when i skimmed through the log, I noticed that some plugins consume a lot of time (maybe slowing down the scan), so long story short, I removed couple of plugins such as iconic and subtitles plugin along with some others, and finally scan started to work (I think), passed 89.1%, its 93% as we speak and its progressing incredibly slow, hopefully the issue got resolved, not sure yet.

I'll update you once Im done.

 

Appreciate your support

[Luke 42081]

Ok I would just be patient and allow the scan to finish.

[Ugnaughts 53]

im testing different methods of scanning of large media library's.  emby seems to do it as fast as possible for fetching from the web, about a title per second. but when you get a library of thousands of strm files then it becomes frustrating to say the least. this is kind of of tough topic since one size dont fit all, usually, for this topic.  but i have done allot of testing and comparing so far. 

here is a few of the meaningful tests ive done

it took 315 minutes to do 13757 strm files. with no metadata stored in the folders, and default selection of emby for metadata except to save nfo and artwork

it took 279 minutes to do 13757 strm files. with metadata stored in the folders, and default selection of emby for metadata except to save nfo and artwork

it took [waiting currently for results] minutes to do 13757 strm files. with metadata stored in the folders, and no selection of emby for metadata except read nfo

i also tried a popular 3rd party app called tinymm to update folder with metadata. the free version only has movie db but 10 eur gets all metadata including imdb. im most likly going to get the pro version so i can get imdb which is how my strm files are saved. if turning off all metadata in emby greatly improves the speed of the scan. and maybe use it if its faster

here's my thoughts

if emby separated the scanning from the main emby process as a process of its own. this would keep the scanning from freezing emby and unresponsive during large media scans that take many hours to days. once again emby is doing as fast as possible for one process.

also make the new scan tool to be able to run multiple instances . then then main emby process creates a work que. that the individual scan process work from till done. i would base the speed of the scan tool off the lowest resource server that uses emby. so say if they are using qnap nas then select only 1 scan tool process. but if the are on a pc maybe select 3-5. 

back when i used plex, i noticed that was how they did it. but don't get me wrong plex had allot of issues and even with there scan tool algorithm and it would eventually use all the cpu for its scans, with no selection of how much . but if emby keeps there algorithm and implements a separate scan tool as multiple process that takes from a que. then that's as good as it will get ever. other than small tweaks to algorithm. 

also as a side note once you get all files initially scanned you should be fine. because the changes shouldn't be enough to really stress emby.