generic .nfo file questions

[psybertech 33]

Hello again.

 

OK, all is working very well on my Emby server!, so this is just a few questions about .nfo files.

 

I do not use .nfo files currently.

 

I know how to turn on the saving of nfo files but I have questions on how they work exactly.

1. Once enabled, is there a fast way to create them initially? I know a full refresh per library works, but I'd like to know if there is a faster way to do the initial creation of an entire library that would be fast(er)
2. Once the nfo files exist, on each scan, are they checked against a timestamp or something so the scan time isn't increased checking each directory/file again?
3. Is the DB still used primarily? I have to assume yes and the nfo files are just there for initial/update scans, correct? (see #2)
4. Once this is setup, if I build a new server, the .nfo files should be checked/read in on the initial scan rather than trying internet scrapers like themoviedb, correct? Well, unless the nfo doesn't exist...
5. I am confused here too... If I manually edit metadata using Emby's metadata browser, will the .nfo files get updated? If I lock an item there, is that setting saved in the .nfo? or is that a db only setting? EDIT:  found this after a small test scan completed and a test... <lockdata>true</lockdata> - so yes, this is honored. GREAT!!!!
6. If I choose NOT to save images locally, will Emby download the images to the server structure during the initial scan for everything, or only as needed?
7. If I use .nfo files (set on the main server to save), but do not enable that on other servers, will the .nfo files be read on the additional servers? Or is that setting in the library section just to save them? Or is it bigger... meaning, in order to use .nfo files at all, that has to be enabled?

See, sorry, I am a bit confused! haha... it seemed simpler back in the early XBMC days, but that was over a decade ago.

 

My questions stem off these reasons...

• I love to test and play, so if I setup a new test server, it would be great to get the initial scan done in less than a week.. ideally a day would be nice.
• Same as above, except I want to run a backup server for when I place my main server offline for whatever reason - or just to have sort of a fallover server if the load is great on the main server and I'd like as close to a mirror of the main as possible but have the backup just scan daily or on-demand as opposed to every 45 mins to avoid conflicts and don't want to have to wait a day or more to play catchup to find my new stuff.

So, I guess I just would like to know the benefits on .nfo files and see if they will speed up scan times in some situations (full, initial scans once the .nfos exist) or slow scans down once they are built.

 

Man, I know I am making this sound more confusing than it is or should be.

 

Thanks in advance.

 

j

Edited July 19, 2019 by psybertech

[Happy2Play 9783]

Hello again.

 

OK, all is working very well on my Emby server!, so this is just a few questions about .nfo files.

 

I do not use .nfo files currently.

 

I know how to turn on the saving of nfo files but I have questions on how they work exactly.

1. Once enabled, is there a fast way to create them initially? I know a full refresh per library works, but I'd like to know if there is a faster way to do the initial creation of an entire library that would be fast(er)
2. Once the nfo files exist, on each scan, are they checked against a timestamp or something so the scan time isn't increased checking each directory/file again?
3. Is the DB still used primarily? I have to assume yes and the nfo files are just there for initial/update scans, correct? (see #2)
4. Once this is setup, if I build a new server, the .nfo files should be checked/read in on the initial scan rather than trying internet scrapers like themoviedb, correct? Well, unless the nfo doesn't exist...
5. I am confused here too... If I manually edit metadata using Emby's metadata browser, will the .nfo files get updated? If I lock an item there, is that setting saved in the .nfo? or is that a db only setting? EDIT:  found this after a small test scan completed and a test... <lockdata>true</lockdata> - so yes, this is honored. GREAT!!!!
6. If I choose NOT to save images locally, will Emby download the images to the server structure during the initial scan for everything, or only as needed?
7. If I use .nfo files (set on the main server to save), but do not enable that on other servers, will the .nfo files be read on the additional servers? Or is that setting in the library section just to save them? Or is it bigger... meaning, in order to use .nfo files at all, that has to be enabled?

See, sorry, I am a bit confused! haha... it seemed simpler back in the early XBMC days, but that was over a decade ago.

 

My questions stem off these reasons...

• I love to test and play, so if I setup a new test server, it would be great to get the initial scan done in less than a week.. ideally a day would be nice.
• Same as above, except I want to run a backup server for when I place my main server offline for whatever reason - or just to have sort of a fallover server if the load is great on the main server and I'd like as close to a mirror of the main as possible but have the backup just scan daily or on-demand as opposed to every 45 mins to avoid conflicts and don't want to have to wait a day or more to play catchup to find my new stuff.

So, I guess I just would like to know the benefits on .nfo files and see if they will speed up scan times in some situations (full, initial scans once the .nfos exist) or slow scans down once they are built.

 

Man, I know I am making this sound more confusing than it is or should be.

 

Thanks in advance.

 

j

 

1. If the libraries don't exist yet then they will be created as the media is imported.  But if your media has already been imported your have to do a refresh all to resave all your media to generate the nfo.

2. Depending on your setup Realtime monitoring will look for changes in your files then auto refresh them.  Otherwise it will be picked up on the next scheduled scan.

3. Yes

4. Additional servers should have all metadata options disable to you don't have both servers in a loop updating each other.

6. Depends on if you have "Download in advance" enabled per library.  But if running more then one server you will want to save with media.

7. Emby by default will read all existing metadata before going to provider site.  So as noted above it is best to save metadata and images with media.

[psybertech 33]

1. If the libraries don't exist yet then they will be created as the media is imported.  But if your media has already been imported your have to do a refresh all to resave all your media to generate the nfo.

2. Depending on your setup Realtime monitoring will look for changes in your files then auto refresh them.  Otherwise it will be picked up on the next scheduled scan.

3. Yes

4. Additional servers should have all metadata options disable to you don't have both servers in a loop updating each other.

6. Depends on if you have "Download in advance" enabled per library.  But if running more then one server you will want to save with media.

7. Emby by default will read all existing metadata before going to provider site.  So as noted above it is best to save metadata and images with media.

 

 

Thank you Happy!

 

I will think about saving the images along side media too. I was trying to save space on the NAS, but it really isn't that big of a hit on space in reality, especially if it helps. I have a ton of questions about how it stores actors, but that is another question for another day.

 

And as far as #2

2. Depending on your setup Realtime monitoring will look for changes in your files then auto refresh them.  Otherwise it will be picked up on the next scheduled scan.

 

I am using fstab CIFS mounts back to my NAS from my Ubuntu based server (and all tests/fallover will be too via VMs), and I think inotify is used on the Linux versions for realtime changes so ths doesn't work for me, so I'd like to know how this works...

 

Will it check every file, every time for the local timestamp on the NAS? Or is smart-ish dir scan used (i.e. some ls -tal or more intricate find command) to find recent media and compare dates to the last successful scan time?

 

I'd try to check it out myself but the main GIT for Emby is no longer around for me to poke at. 

 

I just have so many files and I want to keep scans down as much as possible. My scans have recently crept back up to ~8 minutes as it stands now and just an concerned with adding more to the structure and how it will impact a basic scan, let alone a full scan.

 

Thanks again!

 

j