Same media constantly "Finished Processing" over and over

Plex Media Server
1

Server Version#: Version 1.23.2.4600 (Docker Container, Host is Ubuntu 18.04)
Player Version#: N/A

I’m having a bizarre issue with Plex in the last few weeks. Both the TV and Movie scanner are on the latest updated scanners.

When scanning, especially movies, the same Media gets reprocessed over and over again. I cannot figure out what is triggering this. I looked through the logs and see errors such as the following:

SQLITE3:0x80000001, 28, double-quoted string literal: "2021-05-17 17:00:00"

File '/path/to/movie.avi' didn't exist, can't skip.

Noting item as needing media analysis, scanning=0, streams=0, missing channels=0, streamability=0, size=0, time=1.

Unknown metadata type:

The “time=1” intrigued me because I thought maybe something was modifiying the timestamps of my media, but I traced a few items that constantly reprocess and they have no changed timestamps nor have they been updated or “touched.”

I don’t know if this is coincidence but the items that trigger all do appear to have multiple versions of that same movie/show that get merged into one item.

I’ve tried the following:
Clean bundles, optimize db, empty trash, rescan 2x
sqldump plex main db and blobs db and rebuild
delete a media item in quesiton, scan, empty trash, readd
Untick “use local assets” and scan, empty trash, scan 2x

It also appears to be affecting at least one other person:

I’ve attached my logs to this post as well.
Plex Media Server Logs_2021-06-01_14-25-17.zip (1.8 MB)

The other odd issue that started around the same time is even though I have “Prefer Local Metadata” unticked, Plex is picking up the titles in the file metadata. If I refresh metadata, it then correctly uses the Plex Agent matched title. Not sure if these are somehow related.

None of this has worked. I really, really don’t want to do “the Plex Dance,” would like to figure out the root cause of this. I’ve attached the plex logs. Thanks in advance for any assistance.
-Ed

1 Like
2

I’m having the same issue. Hoping a plex dev can take a look.

3

Don’t everyone answer at once :grin: Jk
But seriously any help from Plex devs would be appreciated. Also any details on what triggers the new scanners to “process” or “finish processing” new media would be appreciated. E.g. what triggers the time flag? File mtime? Folder mtime? Both? Etc. Thanks in advance again for any insight.

4

From a look at the logs it looks you have the same files in two different paths and both those paths are added to your library.

The two paths I spotted are:

/Users/andrewmcdonald/Downloads/My Drive_andrewmcdonald/Plex/Movies
/Volumes/Macintosh HD/Volumes/Plex Cloud Servers_andrewmcdonald/Shared drives/PlexCloudServers/Media/Movies

It keeps scanning in the one file thinking it’s new then when we go to add it we see the file already exists in the database and skip it. Get rid of one of the paths from the library.

1 Like
5

Thank you. These are two different paths though. Are you saying I’d you have a file with same hashed in two different paths, plex will fight back and forth? Is this a behavior of the new scanner? The old scanner didn’t do this. Thanks for the reply!

6

That seems a bit silly. 2 people’s collections combined into one library would cause these issues. Especially having 4k and non 4k combined. I want them combined so I don’t get dupes in my “continue watching”. It’s a cleaner set up than having its own 4k library. This way at home on my main device it auto plays 4k version and at work on work computer it plays the 1080p version all automatically.

7

Could you elaborate a bit one what the new scanner does differently so I can fix the problem correctly. This doesn’t happen if I revert to the old scanner only the new one. The versions of the like movies in the two paths have different filenames, but the file sizes are identical, so is it when hash matches occur or are you saying the new scanner doesn’t support more than one version of a movie period?

1 Like
8

There is no difference except the new scanner makes it visible. The code that checks for duplicates is the same code between both scanners.

9

Thank you for the reply. But something different is definitely happening as the legacy scanner can complete a scan in minutes, whereas the new one takes hours, and the only difference, at least in alerts, is the finished processing entries. What’s more odd is if a second scan is run directly after the first with the new agents, it seems to fly through. But several hours later, “finished processing” floods brings it to a crawl again.

Is there any scanner logic that can be shared. E.g. what is the scanner looking for specifically, for example:
Enumerate folders and files
Check file time stamp
Check file size
Check folder time stamp
Check file hash
Deltas? → process

I have noticed, at least for the ones spot checked, the filenames and dates are different but the sizes are identical. I thought this meant they can’t have the same hash but I could be wrong. Also this only happens on the new movie scanner, not the new tv scanner.
Ed

1 Like
10

The new scanner also seems to delete vob files from the library. For now I’ve reverted to the legacy scanner and once again, everything completes in minutes.

If @drzoidberg33 could clarify what you mean by this:
“It keeps scanning in the one file thinking it’s new then when we go to add it we see the file already exists in the database and skip it.”

I would greatly appreciate it. Again the files are in different paths and even different filenames in most cases in my spot checks. So is it a file size match (exact same byte size)? Or are you saying the new movie scanner will do this anytime there’s more than one version of a movie in the same library? Fwiw the new tv scanner agent doesn’t behave the same way.

1 Like
11

@drzoidberg33 even if you can point me towards any scanner documentation, it would be much appreciated. Really want to narrow this issue down to understand what is happening.

1 Like
12

I have the same exact issue. Anything clarification from the dev above?

13

You have files with the same hash (i.e. the same file) in two different paths. Removing the duplicate files/path will resolve the issue.

1 Like
14

Ok thank you!

15

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.