Re-request deleted media

[narrow wedge]

https://github.com/sct/overseerr/issues/617

This issue pretty much sums it up

Basically, when media is deleted on Plex, jellyseerr traps the content in a deleted state which doesn't allow for re-requesting. This means a simple Plex drive disconnection could cause the entire jellyseerr library to come up as deleted and then non-requestable. Clearing db would lose all the requests + logged in Plex users

[sly hearth]

Deleted media can be re-requested

Can you show me what you mean

Because deleted media are allowed to be re-requested.

And the only thing that even puts deleted badge is media availability sync job so if that doesnt run it wont.

Also about the disconnect part, thats gonna be fixed nest version where that job wont run if plex is disconnected

See. Deleted media is requestable

Could you show us your logs

[narrow wedge]

Thanks for the quick reply, I'll pull up an example and see what logs can get collected

In the meanwhile, is it possible to reset all deleted statuses?

Even via manual db manipulation (i know that's probably not the best idea)

[sly hearth]

Just run a full scan

Yhat should make everything available again

[narrow wedge]

It's hard to find a good example as my deleted filter shows nothing, but then if i manually look for some shows they'll be in the deleted state

[sly hearth]

Just run the full scan first

See if it makes them available

[narrow wedge]

This is the issue - I genuinely deleted the items - so they got marked as deleted with a full scan

Then when trying to rebuild the library, I had to manually go to eevery deleted item in jellyseerr and clear the data on each for it to re-request

Basically Jellyseerr is tied to plex watchlist
When the item is deleted, plex watchlist wont re-trigger the items

[sly hearth]

If you deleted them they should be requestable tho. Full scan would not mark them as deleted.

Only media availability scan eould

So run that job

[narrow wedge]

Plex watchlist sync doesn't re-request deleted media in my experience

[sly hearth]

Then find one and send example plus logs

[sly hearth]

Plex Watchlist is broken on latest.

Its not bc they're deleted

Its bc plex api changed

The fix is available in next release

[narrow wedge]

I was not on latest - I was on an older version , only moved to latest this morning but either way this behaviour seems to be the same

Because when plex watchlist was pulling new entries, the behaviour was the same

media in deleted state > Plex watchlist runs > deleted media doesnt get re-requested despite being on the watchlist

[sly hearth]

Like I just said

Its broken in latest

[narrow wedge]

What is broken, pulling new plex watchlist items?

[sly hearth]

And older versions

[sly hearth]

Need logs for this

And an example of item that cant be re-requested

Without logs cant tell

Do send when can c:

[sly hearth]

In my experience I can re-request even at deleted state

Also run a media availability sync job and send logs too

[narrow wedge]

Plex watchlist seems to be working fine on the version I was on at least ( it was a lot older but I lost the yaml now so can't say what version it was)

[sly hearth]

It cant get new

Even on older

[narrow wedge]

The strange thing is my deleted list is clear, but there are soem shows that are in deleted state

[sly hearth]

Plex api changed

[sly hearth]

Idk what you mean by deleted list

Without examples I cant tell 😅

[narrow wedge]

I just added this to my plex watchlist

I don't undertand what's broken here..

  jellyseerr:
    cap_drop:
      - ALL
    deploy:
      resources:
        limits:
          cpus: '2'
          memory: 4096M
    environment:
      GID: '568'
      GROUP_ID: '568'
      NVIDIA_VISIBLE_DEVICES: void
      PGID: '568'
      PORT: '30042'
      PUID: '568'
      TZ: Europe/London
      UID: '568'
      UMASK: '002'
      UMASK_SET: '002'
      USER_ID: '568'
    group_add:
      - 568
    healthcheck:
      interval: 30s
      retries: 5
      start_interval: 2s
      start_period: 15s
      test:
        - CMD
        - wget
        - '--quiet'
        - '--spider'
        - http://127.0.0.1:30042/api/v1/status
      timeout: 5s
    image: ghcr.io/fallenbagel/jellyseerr:latest
    platform: linux/amd64
    ports:
      - mode: ingress
        protocol: tcp
        published: 30042
        target: 30042
    privileged: False
    restart: unless-stopped
    security_opt:
      - no-new-privileges=true
    stdin_open: False
    tty: False
    user: '568:568'
    volumes:
      - bind:
          create_host_path: False
          propagation: rprivate
        read_only: False
        source: /mnt/Applications/jellyseerr/config
        target: /app/config
        type: bind
volumes: {}
x-notes: >
  # Jellyseerr


  ## Security


  **Read the following security precautions to ensure that you wish to continue
  using this application.**


  ---


  ### Container: [jellyseerr]


  #### Running user/group(s)


  - User: 568

  - Group: 568

  - Supplementary Groups: apps


  ---


  ## Bug Reports and Feature Requests


  If you find a bug in this app or have an idea for a new feature, please file
  an issue at

  https://github.com/truenas/apps
x-portals:
  - host: 0.0.0.0
    name: Web UI
    path: /
    port: 30042
    scheme: http

image: ghcr.io/fallenbagel/jellyseerr:latest

This is the 'deleted' filter

No results, but if I browse enough I'll eventually find a show that is actually in the deleted state

So not sure If a db cleanup is needed

I would reset my db but then all the plex users would need to sign in again

We can see adding to plex watchlist > jellyseerr caught it > request made it to my backend

[sly hearth]

Its broken for other users

[sly hearth]

Ah that you mean

[narrow wedge]

Strange, plex watchlists was never an issue with me

[sly hearth]

Without examples csnyvtell

Also logs

[narrow wedge]

The main issue I have is the deleted filter doesnt show all the deleted shows and I have to find them manuallya nd reset them for the watchlist to re-request

[sly hearth]

Also this will only show deleted requests. Not deleted media.

[sly hearth]

Requests count as requested manually

Not warchlist

[narrow wedge]

Ahh that's useful

Is there a table that contains deleted media that I could just clear on the db

My jellyseerr might not be in a good state

[sly hearth]

Without clearing it'd be more useful if you could share a broken media so we can figure out whats going on but yes in db there is a media table that should have that

[narrow wedge]

Cool I will see if I can find a show that's deleted state, see if it can be re-requested without removing it (didnt work in the past but lets test the theory) and then go through from there

just gathering a few examples now to test

[sly hearth]

Its in a requestable state I.e the request button. So watchlist should in theory re-request it

Request button only shows when its in a requestsble state 🤔

[narrow wedge]

I did so much manual clean-up, trying to find the right example here

The issue was with titles that are "deleted" but have a prior request attached to it

You know what I'll delete a title that I've requested and we can use that

[sly hearth]

If an item gets deleted it should also mark the request as deleted. If its not there is some issue there yeah

[sly hearth]

Lets see

Need reproducible steps so we can check too 😅

[narrow wedge]

okay so here is what I did

Added Shrek to my watchlist on Plex -> Jellyseerr picked it up - > It downloaded onto plex server -> Completed status on jellyseerr -> I manually deleted it from plex -> Ran media availability sync -> Jellyseerr says it has been deleted

Now I ran plex watchlist sync again - Shrek is in my plex watchlist

But the state wont move to re-requested

(screenshot is the same before and after plex watchlist sync)

[sly hearth]

Show me without the modal opened too

[narrow wedge]

request button is there - but plex watchlist sync wont re-request the show

I would have to either click request again, or i would have to use 'Clear data' for plex watchlist sync job to request the media again

[sly hearth]

Ill check

And try and reproduce this

[narrow wedge]

Now when there's 100+ entries that get marked as deleted due to various reasons (plex cleanup, readldebrid cleanup etc) , this is tedious as I would then have to manually goto every show / movie and clear data for plex watchlist sync to request again

hopefully that is enough information to help you reproduce this

I think to put it simply, what I'm looking for, is for jellyseerr to look at the plex watchlist and be like "oh well the media is deleted, but its still in their watchlist. lets request it again"

for reference , my image tag was
fallenbagel/jellyseerr:2.7.2
and now is currently
ghcr.io/fallenbagel/jellyseerr:latest which should be 2.7.3
And Plex watchlist works fine for requests on both

And this bug/issue has always existed but I would just manually clear data wherever possible to get plex watchlist to re-request deleted media

[narrow wedge]

Would you be able to point me to the table that contains deleted media entries so that I could reset my jellyseerr to allow my plex watchlists to go through

[covert cosmos]

I'm in the same position. I've played around with Agregarr (which creates fake media so that Plex will display "new media"). However, when a media file is deleted, and a person then requests it, it won't be processed. It has the tag "deleted" in Jellyseerr, and I have to clear the data for that movie for the request to work.
I'm only at around 30-40 movies that I have been testing out with Agregarr. Is there a way to filter all media for that deleted tag so I can clear them?

To be clear, the way it works is how I think it should work. I just want to fix the mess that I made with Agregarr

I personally don't want this function. It would be a pain to delete media from the server:
"I think to put it simply, what I'm looking for, is for jellyseerr to look at the plex watchlist and be like "oh well the media is deleted, but it's still in their watchlist. Let's request it again."

[narrow wedge]

I think a simple tick-box in the settings/config should be sufficient under the plex watchlist settings to allow for either behaviour if possible

Something like this lol

I've paired this with debrid so all my content is symlinked hence idm deleted content coming back - no local storage used for media

[covert cosmos]

Yes, that is true, a tick-box would be good in this case for you.
Maybe add it as a future request?

[narrow wedge]

Tbh even a button that clears media deleted history would work too for manual cleanups here and there, just to remove those tags and allow plex watchlist to apply again

Let's see once it's reproduced by @sly hearth

[narrow wedge]

@covert cosmos I made a shell script to delete the entries in the meantime whilst this is reviewed

Enter the path to your db.sqlite3 file:
/mnt/Applications/jellyseerr/config/db/db.sqlite3

Listing deleted entries (status=7):
15414|tv|40143|7
15417|movie|245703|7
15418|movie|617126|7
15420|movie|426063|7
15422|movie|1242898|7
15425|movie|950396|7
15429|movie|1078605|7
15430|movie|701387|7
15432|movie|533533|7
15433|movie|1234821|7
15434|movie|696506|7
15435|movie|1054867|7
15436|movie|241554|7
15675|tv|245703|7
15677|tv|241554|7
15678|tv|216082|7
15679|tv|211288|7
15691|tv|43168|7
15693|movie|20982|7
15694|movie|118406|7
15696|movie|317442|7
15697|movie|50723|7
15698|movie|16910|7
15699|movie|16907|7
15701|movie|17581|7
15702|movie|18861|7
15709|movie|810693|7
15710|tv|13916|7
15711|tv|31911|7
15773|tv|285356|7
15775|movie|360920|7
15776|tv|14658|7
15777|tv|294867|7
15778|tv|276349|7
15779|tv|126027|7
15780|tv|152483|7
15782|tv|226285|7
15783|movie|982843|7
15785|tv|253391|7
15801|tv|236565|7
15802|tv|259886|7
Total deleted entries: 41

Do you want to backup and delete these entries? (y/n): y

Backup created: /mnt/Applications/jellyseerr/config/db/db.sqlite3.backup
Deleted 41 entries from media table.```

Stop the container and then run against your db

root@narrow wedge-Server[/mnt/Applications/jellyseerr/config]# cat cleanup_deleted_media.sh

echo "Enter the path to your db.sqlite3 file:"
read DBPATH

if [ ! -f "$DBPATH" ]; then
  echo "File not found: $DBPATH"
  exit 1
fi

echo "Listing deleted entries (status=7):"
sqlite3 "$DBPATH" "SELECT id, mediaType, tmdbId, status FROM media WHERE status = 7;"

COUNT=$(sqlite3 "$DBPATH" "SELECT COUNT(*) FROM media WHERE status = 7;")
echo "Total deleted entries: $COUNT"

read -p "Do you want to backup and delete these entries? (y/n): " CONFIRM

if [[ "$CONFIRM" =~ ^[Yy]([Ee][Ss])?$ ]]; then
  cp -p "$DBPATH" "${DBPATH}.backup"
  echo "Backup created: ${DBPATH}.backup"
  sqlite3 "$DBPATH" "DELETE FROM media WHERE status = 7;"
  echo "Deleted $COUNT entries from media table."
else
  echo "No changes made."
fi```

This can easily be added as a button in the ui if the team agrees with the implementation

Obviously more checks can be added to ensure the backup is made before it executes the deletions (or abort if backup file didnt create) but yeah I manually backed up before running anyways

[covert cosmos]

Wow, nice work!
Yeah I can just create a manual backup before running this. Will try it in a few days and get back to you with the progress.