#Slow Interface and Requests timing out

116 messages · Page 1 of 1 (latest)

tranquil lotus
#

Since I pulled a new image (hotio:nightly) yesterday, the Seerr interface has been very slow and auto approve requests are failing regularly. Please note I've been running hotio:nightly since it became the new merged Seerr beta. I haven't had any issues like this until now. Log attached. You'll find most of what I saw towards the end of this log.

clear nest
#

only thing alarming i see is timeouts to your arr exceeding the 5000ms and causing a error. are the arr's reachable?

tranquil lotus
#

Yes. Everytime I go in and do a test they pass.

clear nest
#

has anyone else on hotio reported issues?

#

run this and see what the time is to reach radarr docker exec seerr curl -w "\nTime: %{time_total}s\n" http://radarr:7878/api/v3/system/status?apikey=YOUR_API_KEY

tranquil lotus
tranquil lotus
# clear nest run this and see what the time is to reach radarr `docker exec seerr curl -w "\n...
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0{
  "appName": "Radarr",
  "instanceName": "radarr",
  "version": "6.0.4.10291",
  "buildTime": "2025-11-14T23:25:20Z",
  "isDebug": false,
  "isProduction": true,
  "isAdmin": false,
  "isUserInteractive": true,
  "startupPath": "/app/bin",
  "appData": "/config",
  "osName": "alpine",
  "osVersion": "3.22.3",
  "isNetCore": true,
  "isLinux": true,
  "isOsx": false,
  "isWindows": false,
  "isDocker": true,
  "mode": "console",
  "branch": "master",
  "databaseType": "sqLite",
  "databaseVersion": "3.50.4",
  "authentication": "external",
  "migrationVersion": 242,
  "urlBase": "",
  "runtimeVersion": "8.0.12",
  "runtimeName": "netcore",
  "startTime": "2026-02-10T06:08:04Z",
  "packageVersion": "release-1a57dc4",
  "packageAuthor": "[hotio](https://github.com/hotio)",
  "packageUpdateMechanism": "docker"
100   821    0   821    0     0   250k      0 --:--:-- --:--:-- --:--:--  267k
}
Time: 0.003200s
clear nest
#

response time is fine so something else is deff going on

tranquil lotus
#

only thing I can think of is maybe this is happening when the network is being stressed from lots of requests and subsequent downloads happening at the same time.

Also to note, Seerr and Radarr/Sonarr are in different networks. Seerr in custom bridge network and Sonarr/Radarr in macvlan. Though as I'm talking about it, maybe worth me putting Seerr in the same macvlan network.

clear nest
#

your response is 3.2 milliseconds way under the threshold

tranquil lotus
#

i'll have to run this test tonight when my server is in "prime" time

clear nest
#

its generally recommended to keep all your apps on a custom docker network tho your approach should work as well.

safe lodge
#

We recently added a timeout

#

To service api calls if it exceeds too much

#

It will timeout

#

For some reason the radarr/sonarr calls are taking too long

#

That shouldnt happen

#

Can you press test on services

#

And send logs too

#

So we can see whats wrong w your network

clear nest
#

ahh i missed that one. ill let fallen take it from here 🙂

safe lodge
#

Im also seeing this

#
info][Plex Scan]: Recently Added Scan Complete 
2026-02-10T03:10:07.561Z [debug][API]: Something went wrong retrieving movie {"errorMessage":"[TMDB] Failed to fetch movie details: Request failed with status code 404","movieId":"1483978"}
#

404 on tmdb

#

So there's a combination of network issues happening including dns

#

Can i see your compose?

light axle
#

Please note I've been running hotio:nightly since it became the new merged Seerr beta
It's not, seerr have this own image why use hotio image ?

tranquil lotus
# safe lodge Can i see your compose?
  seerr:
    container_name: seerr
    image: ghcr.io/hotio/seerr:nightly
    networks:
      - swag
    ports:
      - 5055:5055
    volumes:
      - ${APPDATA}/seerr:/config
    environment:
      - fpm=false
      - branch=dev
      - TP_HOTIO=true
      - TP_THEME=dark 
    extends:
      file: .docker-compose.extends.yaml
      service: common
light axle
#

you use nightly we can't know which version you are running on

#

if you could get us the commit you are running on would be nice

tranquil lotus
#

i have to use nightly to use the new seerr

#

will do

light axle
#

seerr have develop tag

#

it's exactly the last commit of seerr

tranquil lotus
#

But I was migrating from Overseerr and was told I had to run hotio:nightly to migrate and not lose anything

#

e3dc1c302d18bc9618dcd2eb0f99c1c59b438df2

light axle
#

Overseerr migration is handled by seerr script

tranquil lotus
#

Ok well about a month ago i saw people saying to just switch from ghcr.io/hotio/overseerr:latest to ghcr.io/hotio/seerr:nightly and everything would just work

#

and it has

#

until yesterday

#

but it looks like maybe I have had these underlying issues and didn't know about it

safe lodge
#

Can yiu share your arr compose

light axle
#

I don't think it comes from your third party image, just saying I don't understand why people use this

tranquil lotus
# safe lodge Can yiu share your arr compose
radarr:
    image: ghcr.io/hotio/radarr:release
    container_name: radarr
    networks:
      vlan140:
        ipv4_address: 10.4.140.213
    ports:
      - 7878:7878
    volumes:
      - ${APPDATA}/radarr:/config
      - /mnt/fast_storage:/fast_storage
      - /mnt/storage:/storage
      - /optane/scripts/theme.park/98-themepark-radarr:/etc/cont-init.d/98-themepark
    environment:
      - TP_HOTIO=true
      - TP_THEME=dark 
    extends:
      file: .docker-compose.extends.yaml
      service: common
safe lodge
tranquil lotus
#

still want that test log?

light axle
#

can you do docker compose pull on seerr ?

safe lodge
#

As in api calls to arr

tranquil lotus
light axle
#

ahah

#

nightly sucks 🙂

#

you don't have the last commit

#

I guess it rebuild every X hours

#

it's why develop is better it rebuilds on new commit

tranquil lotus
light axle
#

They are running the CI with qemu

#

it's why they their build is slow

tranquil lotus
#

Better?

light axle
#

it's the latest version at least (if you want that)

safe lodge
#

@tranquil lotus could you add the dns block

dns:
   - 9.9.9.9
   - 8.8.8.8
   - 1.1.1.1
#

Into your compsoe

#

And then check if browsing around is slow

#

Also press test in services and send logs

tranquil lotus
#

Will do

#

First time browsing to movies and series tabs was slow to load posters but now subsequent scrolling down to load more posters is snappy. Tests are passing. But will clear log, test again and post log here shortly.

#

Hmmm, deleting the seerr.log and rebooting doesn't generate a new log file?

safe lodge
safe lodge
#

Try enabling settings > networks > force ipv4 option

#

And restart

tranquil lotus
#

When I do service tests I'm not seeing those in the logs

#

FYI I don't have DNS Cache enabled in case I should

safe lodge
#

Thats mostly if you wanna avoid dns spam

tranquil lotus
#

off topic, but is there a quick way to disable watch list for all users? When I select all users and do bulk edit it's greyed out

#

[error][Plex.TV Metadata API]: Failed to retrieve watchlist items {"errorMessage":"Request failed with status code 401"}

safe lodge
tranquil lotus
#

got it. I kind of want to just disable watch list for all users

tranquil lotus
#

Interface is still super slow and requests showing as failed with timeout of 5000ms exceeded even though they are in fact being added to radarr.

#

what tag version did you add the timeout? Im curious if I downgrade back to before that change if things "work" again

#

just so I can isolate that it is indeed the change the brought this issue to light

clear nest
#

It’s not the version as there’s been no other reports, and for myself it’s still been fine. I suspect a network issue going on.

tranquil lotus
#

possible. Just strange that none of my other 50+ containers seem to be having an issue. I think I'll just let this go for now though because I have like 50 new server parts in my office ready to rebuild my whole server stack at home which will include many underlying changes so we'll see where the dust settles when that's all done 😄

safe lodge
tranquil lotus
safe lodge
warm summit
#

Sorry, to ping an old post but I am experiancing something very silimer, Radarr and Sonarr keep timing out but the log show no info other than "request timed out"

warm summit
#

nope using seerr/seerr

#

2026-02-14T19:51:39.528Z [error][Radarr]: [Radarr] Failed to retrieve root folders: timeout of 5000ms exceeded
2026-02-14T19:51:40.580Z [error][Radarr]: [Radarr] Failed to retrieve root folders: timeout of 5000ms exceeded

safe lodge
#

Could just be thay container cant reach them

#

So currently we timeout after 5seconds of trying to reach

#

So please show us your compose + explain your setup

#

So we can go through the troubleshooting steps together

warm summit
safe lodge
tranquil lotus
tranquil lotus
#

Looks like moving from image: ghcr.io/hotio/seerr:nightly to image: ghcr.io/seerr-team/seerr:preview-servarr-timeout-increase-more is a no go. Essentialy takes me to the setup wizard.

safe lodge
#

Between hotio and our image

#

Please send compose

tranquil lotus
# safe lodge Please send compose
  seerr:
    container_name: seerr
#    image: ghcr.io/seerr-team/seerr:preview-servarr-timeout-increase-more
    image: ghcr.io/hotio/seerr:latest
    networks:
      vlan140:
        ipv4_address: 10.4.140.214
    ports:
      - 5055:5055 
    volumes:
      - ${APPDATA}/seerr:/config
    environment:
      - fpm=false
      - branch=dev
      - TP_HOTIO=true
      - TP_THEME=dark 
    extends:
      file: .docker-compose.extends.yaml
      service: common
safe lodge
tranquil lotus
#

ok i'll change it for this test.

#

First test request since the updated image run went through successfully. Interface is still a bit laggy but at least the requeset didn't fail

safe lodge
#

Should make it a bit betrer

tranquil lotus
#

I took that out because last time we tested it didn't seem to make much difference but I'll add it back in and try again

safe lodge
tranquil lotus
#

Correct, and yes adding that back in makes the interface much snappier

safe lodge
#

You can stick to that preview image for now. We plan to add the timeout as a configurable field for next version

tranquil lotus