Emby scanning logic

[Edrock200 53]

Have a question wrt Emby 4.5.4 scanning logic. I use cloudbox autoscan with emby to send targeted scan requests for newly added media. 

Based on what I've read and seen, if emby gets a request to scan in a path that already exists in the library, it will target scan that folder. If the scan request is a new folder that doesn't exist in the library, it will scan that entire librarys root folder. 

My question is, 1. is the above understanding correct, and 2. does emby scan in series or in parallel? E.g. if a full library scan or a scan of an entire root folder is running, and I open a show and select "scan media files" within that show, does emby wait for the other scans to complete or does it do them in parallel? Based on what I'm seeing it seems to be be in series but wanted to confirm.

 

Last question - does 4.6.4 change any of the scanning logic mentioned above?

Thanks in advance for any insights.

Ed

[Happy2Play 9785]

20 minutes ago, Edrock200 said:

1. is the above understanding correct

From my understanding yes.

16 minutes ago, Edrock200 said:

Based on what I'm seeing it seems to be be in series but wanted to confirm.

This is my understanding also as having parallel operations could be problematic as you could violate provider request speed limits.

[Edrock200 53]

33 minutes ago, Happy2Play said:

From my understanding yes.

This is my understanding also as having parallel operations could be problematic as you could violate provider request speed limits.

Thank you for the response. I know plex does in parallel. Series is fine too if the API could be modified to allow "priority" or manual scans to move to the top of the queue. E.g. pause current scan, do manual, proceed. Does 4.6.4 change this logic at all?

[Happy2Play 9785]

6 minutes ago, Edrock200 said:

Does 4.6.4 change this logic at all?

Not to my knowledge, but @Lukewould have to comment on this.

[Luke 42086]

It's mostly correct. It takes the path that was modified and finds the first ancestor parent that exists in the library, and then scans from there.

[Edrock200 53]

17 hours ago, Luke said:

It's mostly correct. It takes the path that was modified and finds the first ancestor parent that exists in the library, and then scans from there.

Thank you! Is my second theory correct, that scans queue and execute is series?

[Edrock200 53]

Also, just to make sure I understand what you mean @Luke

If I have a Movies library with folders:

/path/to/Movies1/

/path/to/Movies2/

And I trigger an API scan request for a path that has never been scanned in before such as /path/to/Movies2/New Movie XYZ/

Are you saying that Emby will only scan the root of /path/to/Movies2 and not /path/to/Movies1, even if Movies1 is part of the library? If so, I can create a work around for my issue. Thanks again for any insight.

[Luke 42086]

Yes it tries to do a targeted scan when possible.

[Edrock200 53]

Thanks! I think I know of a way to speed up the scans now. Appreciate the info. Now onto testing.