#Windows Container Can't Reach TMDB (Connection Refused)

82 messages · Page 1 of 1 (latest)

orchid swift
#

I tried everything from updating to changing dns servers and all, same result:

2024-06-21T07:48:29.311Z [debug][API]: Something went wrong retrieving movie {"errorMessage":"[TMDB] Failed to fetch movie details: connect ECONNREFUSED 127.0.0.1:443","movieId":"670"}
2024-06-21T07:48:29.348Z [debug][API]: Something went wrong retrieving movie {"errorMessage":"[TMDB] Failed to fetch movie details: connect ECONNREFUSED 127.0.0.1:443","movieId":"38365"}
2024-06-21T07:48:29.352Z [debug][API]: Something went wrong retrieving movie {"errorMessage":"[TMDB] Failed to fetch movie details: connect ECONNREFUSED 127.0.0.1:443","movieId":"216015"}

One thing that happened recently that I think may have caused it, is Virtualbox. I'm not sure why but when I sat a shared folder as read-only (that isn't even on the same drive) permissions got messed with on all drives... I'm not sure how or why it did this and to what extent things got affected, but I'm hoping someone may have a suggestion. The only thing that returned a normal address is launching a container (via docker run) and then doing a dnslookup api.themoviedb.org 1.1.1.1. But when switching to that dns or Google's the result is the same and it resolves to localhost/127.0.0.1. I have another container that seems to write and do everything normally but their written data is on another drive that I changed the owner of back to me. The overseerr container connects to Sonarr, Radarr, and Plex ([Plex Scan]) as normal and commuinicates with them as expected, and I checked that I wasn't banned from tmdb and I wasn't (and even changed my home ip just to be sure), and phone connted to my home wifi doesn't face issues... however there's something odd, Firefox gives off the same error: connection refused, however Chrome doesn't (I mainly use FF) have any issues just like my phone. I'm honestly confused at this point

orchid summit
#

Worked for me: Settings > Jobs & Cache > Find "Media Avaliability Sync'> Run ... for some reason it looks like imdb changes the id.. so,ed to resync.

sacred hill
#

You can add this to your docker container

#
dns:
   - 1.1.1.1
#

If docker compose ^

#

If docker cli

#
--dns 1.1.1.1
orchid swift
orchid swift
# sacred hill ``` --dns 1.1.1.1 ```

yeah I did try with many DNS services, all had the same effect of the connections being refused to 127, I'm sure it's a bigger issue but not sure how to even pinpoint where it could be from

sacred hill
orchid swift
#

Yes

#

I restarted my machine so I don't the command but I basically ran docker run --dns <every dns> .... for windows from the docs, just added the dns part

#

could be the issue from tmdb or is it something internal inside the container/docker?

sacred hill
#

Stop overseerr. Remove the container

#

Recreate with the dns flag passed in

sacred hill
#

Could you show me the full command you ran

#

Also could you try

#

docker exec -it overseerr sh

#

See which ip it resolves to

orchid swift
#
docker exec -it overseerr sh
/app # nslookup api.themoviedb.org
Server:         208.67.222.222
Address:        208.67.222.222:53

Non-authoritative answer:
Name:   api.themoviedb.org
Address: 127.0.0.1

Non-authoritative answer:
Name:   api.themoviedb.org
Address: 2600:9000:2764:ae00:c:174a:c400:93a1
Name:   api.themoviedb.org
Address: 2600:9000:2764:8400:c:174a:c400:93a1
Name:   api.themoviedb.org
Address: 2600:9000:2764:8c00:c:174a:c400:93a1
Name:   api.themoviedb.org
Address: 2600:9000:2764:b800:c:174a:c400:93a1
Name:   api.themoviedb.org
Address: 2600:9000:2764:b000:c:174a:c400:93a1
Name:   api.themoviedb.org
Address: 2600:9000:2764:1400:c:174a:c400:93a1
Name:   api.themoviedb.org
Address: 2600:9000:2764:fe00:c:174a:c400:93a1
Name:   api.themoviedb.org
Address: 2600:9000:2764:a00:c:174a:c400:93a1
orchid swift
sacred hill
orchid swift
#

yeah that's why I'm confused

sacred hill
#

What happens when you try cloudflares dns

#

Check your docker daemon.json

orchid swift
#

pinging thru some DNSs sometimes give the normal addresses, but when switching to said DNSs it doesn't work, still resolves to 127

sacred hill
orchid swift
sacred hill
#

Hmmm

orchid swift
#

there's last-start-linux-daemon.json & daemon.json

#

but only last-start-linux-daemon.json gets modified, daemon.json hasn't changed since a year or so

sacred hill
#

Could you do this

#

Inside the container

#

Cat

#

/etc/hosts

orchid swift
#
127.0.0.1       localhost
::1     localhost ip6-localhost ip6-loopback
fe00::0 ip6-localnet
ff00::0 ip6-mcastprefix
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
172.17.0.2      476b1f1b0cbc
sacred hill
#

Also check your windows hosts

#

If there's any mention of tmdb

#

Windows?

#

And get the proper ip

#

This will narrow down whether docker or whole system dns messed up

orchid swift
#

huh, it says on windows

  nslookup api.themoviedb.org
DNS request timed out.
    timeout was 2 seconds.
Server:  UnKnown
Address:  fe80::1

DNS request timed out.
    timeout was 2 seconds.
DNS request timed out.
    timeout was 2 seconds.
sacred hill
#

Yeah your host machine

#

Is not resolving tmdb

#

So check your windows host

#

Also check what dns your host machine is using. Maybe set your host to use cloudflare. Like from advanced network settings

orchid swift
#

the relevant for docker in hosts file:

# Added by Docker Desktop
192.168.1.15 host.docker.internal
192.168.1.15 gateway.docker.internal
# To allow the same kube context to work on the host and the container:
127.0.0.1 kubernetes.docker.internal
# End of section

Everything else is related to blocks for adobe and unrelated stuff

sacred hill
#

Your host machine is not resolving tmdb api

orchid swift
#

hmm, I wonder why chrome works unlike Firefox

sacred hill
#

Chrome doesn't use OS dns afaik

#

It uses its own

#

I think

orchid swift
#

I just changed firefox to CF and now it works... wth

#

why would windows + google dns have issues, I will reset the dns to CF

sacred hill
#

Yeah this means

#

Your router

#

Wait

#

Wait

orchid swift
#

I'm not sure... don't know ipv6 much

#

It's somewhat common to use fe80::1 as link-local address to provide default gateway functionality to a segment.Sha.

#

hmm I think yes according to google

sacred hill
#

Oh its a link-local address

#

I would say chrck your windows dns settings. Not just from the advanced network place but from the new way to set dns

#

See if there're discrepancies there. See if its messed up

#

In either of the locatoons

orchid swift
#

I just changed to from the new w11 ui settings and the issues persists

#

will check up the old ui on control panel

#

yeah both match up to CF

#

hmm some long time ago I did some flush command that fixed a minecraft server connection issue...

sacred hill
#

I would say restart

#

See if that fixes it

orchid swift
#

I reset (basically every ipconfig command) my network and restarted, same issue

#

I will have to tackle this tommorow, it all started from virtualbox I'm sure, but I'm not sure how it messed everything up this bad
Thanks for all the help fallenbagel, makes me wanna contribute to the repo :p