-
-
Notifications
You must be signed in to change notification settings - Fork 2.9k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Issue]: Authentication/Login Errors #11620
Comments
This comment has been minimized.
This comment has been minimized.
If anyone is looking into it and has questions I will probably respond quicker with a discord ping @jkboy. |
try to clear cache/delete data apps |
On the server? Do I just go in and RM the temp folder? |
no, on the clients, web browser, and for phones, at least in android i know we can delete app data in the settings. same might be doable for ios. At worst just reinstall apps on phones, result will be the same |
I've reinstalled multiple times and it didn't fix the issue. I've also used incognito mode on my browser to ignore caching and cookies and it still triggers the issue |
also do you any sort of caching on your reverse proxy? |
Unsure on the reverse proxy part but I know I deleted all device instances |
i would try to clear any cache on the reverse proxy if there's some. Else I don't know, for me the update did caused the same issue as you but clearing the cache on clients, and restarting everything fixed it. Worth mentioning i don't do any caching on my reverse proxy |
I'm having the same issue. I can temporarily fix the issue by clearing the browser cache (on firefox, clicking the lock, clear cookies and site data), and I'm able to log in. But, once I log out and leave the page, when I return, the login form won't load again. From the web developer console (see attached), near the top, it appears to say something about already being logged in, which I am not. |
I've the same issue with the Android Jellyfin app, both when using my nginx reverse proxy as well as when using the direct ip-address:port number. The pattern is follows: I've no problems logging in from the chrome browser on the same mobile device. Using chrome on my desktop pc is also going fine. See below a part of my docker stack, which is running within a VM hosted via Proxmox.
|
I've also experienced this issue here. I tried to install it in LXC container, and I had to use double SSH tunnel for configuration (since the default installation doesn't support IPv6 connection which needs to be explicitly enabled in the config, and I rely on IPv6 connection to the server). The strange thing is that when I go to the login page through the SSH tunnel, it works as intended, it logs me in, all is good. But as soon as I try the IPv6 address of the server for the login page (mind you, I've enabled it in the settings after I've installed and configured it), the login doesn't work, and says that I entered a wrong password. The very same password that works in the SSH tunnel. Another strange thing is that if I install this software directly into the host OS, without container virtualization sitting in between the server's host OS and the software, this issue disappears. EDIT: It looks like I was accessing a different server by accident (didn't realize it was a different IPv6 address - that happens when there is no DNS involved), so please don't take my previous post into account. I realized that I simply have to use the correct IPv6 address, and it will eventually become functioning once I set up a reverse proxy on the VM running the container. |
I also don't have any caching setup so I don't think that is it. |
@thornbill hope this helps. tried logging on using methods that worked and ones that didn't so you could see both. |
Did this start when you deleted devices/sessions? |
I deleted the devices to attempt to fix the problems. I assumed that if I deleted the devices from the device section they would relink and it maybe would clear some kind of security problem. What started it was 3 incorrect login attempts that followed what everyone else has been saying. Then I came home and deleted all of the devices in an attempt to fix it. Has it messed something up? |
I was trying to login from my phone and it wasn't working and I had been recommended to clear all devices I think by someone in the discord. Is there anything I can do to fix it? I also upgraded nginx/swag at the same time. I also can't access the server at all from my phone now. It doesn't query device based clients like the windows client or my android based client. On my PC not being able to access might have been caused around then. |
I also ran into some authentication issues recently. Not sure if adding my info helps here. Basically, in Jellyfin iOS I have it set up for 192.x.x.x:8096 local access when I am on the home network, and my https://remoteurl.net for remote access when I am not home. I switch them based on which network I'm connected to. When first adding each server (signing into the same user account), everything behaved as usual. But as soon as I switched from the remote connection to the local connection, I was getting the same blank screen as above, and the logs were showing that same Invalid Token message. After this point, the https://remoteurl.net connection always worked as usual, and the 192.x.x.x:8096 connection always gave this blank screen. The only remedy was to delete the app and reinstall. Deleting the servers from the list and re-connecting, or clicking "Reset Application" did not fix the problem. My current workaround is to have a 2nd user account - one for the home connection and one for the remote connection. This has worked without error so far. |
In my case I was using the Authorization header at the reverse proxy and clearing it before sending the request along to the application (Jellyfin). Jellyfin has started using the Authorization header, so setups like mine will be broken. Hosting Jellyfin at a subdirectory i.e. As a work-around I'm thinking of hosting Jellyfin on a subdomain with a secret path as I don't like the idea of having Jellyfin's login be exposed to the Internet and I would prefer an extra layer of control at the reverse proxy layer. |
I did manage to get the issue to go away by clearing browsing data. However, I have not figured out how to clear this information on the android app. If anyone knows that portion feel free to let me know! Thanks for all of the help surrounding this issue! Edit: I have since found that you can clear the cache and data from android settings for jellyfin! Thanks for the help everyone! |
I also have the same problem with the Android App since upgrading to 10.9.1. |
After updating from 10.9.0 to 10.9.2, the issue #11620 is solved for me on the Android app when connecting via the ip-address:port. Nice! The issue remains when connecting via my configured nginx address. The following logs appear on the server:
I'm using nginx proxy manager and added the config as stated on the website In more detail:
|
Observed pattern: Docker config:
Docker runs in a VM: Open Media Vault 7.0-32 (Sandworm). The VM is hosted in Proxmox. |
After upgrading to 10.9.2, this issue appears to be resolved for me as well! |
I followed this instruction to get the reverse proxy working again#11463 (comment) I am not understanding why this change is needed after upgrading from 10.8.x to 10.9.x (i've never filled this in in 10.8.x), but it's working for me. |
@pimw1 usually this is an issue if you haven't properly configured your |
Thank you for your suggestion. I've just tested this:
The issue reappears with these settings. When changing it back to the settings below, it works again. |
The above problem seems to be fixed for me in 10.9.2. I will report back if I run into any other issues. |
This fixed my issue with the android client being able to only connect once, everything seems fine now for me on 10.9.2. |
I have the same issue, e.g.:
I also fixed it by adding my external URL "all=https://jellyfin.mydomain.tld:sslport" to the "Fire and Proxy Settings" in Jellyfin under Dashboard->Networking I don't know if that is a good solution as I wonder if all traffic is going "outside" even if I'm just sitting in my local network?
I have the feeling it really did! I can't remember though if I did it in 10.8.13 or 10.9.1. I am now running 10.9.2 okay, with the above "fix". I hope you guys can narrow the problem down. Good luck and thanks for this great piece of software. |
After installing 10.9.2 the issue went away and all is working again as always, perfectly, thanks a lot, you ROCK! :-D |
I ran into this by deleting a user and recreating them. But on devices they didn't logout of they kept trying to send the old session ID and then would get the "blank" page response. My solve was adding a custom override to "force" a logout; For iOS app; delete, re-install
|
I'm having the same issue on a LG webOS (app version 1.2.2), jellyfin server version 10.9.3 However ... hilariously enough, I can use the search function and find the content and play it just fine. |
Please describe your bug
I upgraded to Jellyfin 10.9.1 yesterday and made sure my reverse proxy containers (swag, duckdns) were also up to date. This morning I went to help a friend with a password that seemingly didn't work. I attempted logging in from my mobile app and it didn't work so I logged back into admin to change his password and then logged out to attempt to use the new password. From there I could no longer access the server from my phone. I then got home did some config file fixing on Swags end as when that updates it tends to cause edited files to not be changed. I found that at that point I could login to my PC using my domain. Attempted again to login with my mobile app and it hung. I looked at the logs and when this hanging happens (which happens on my pc now too) it says that the tokens are invalid invalid authentication or invalid token etc. Thankfully it still seems to work for my home instance using the local IP address. Every time I opened the clients that seemingly were breaking auth it would crash swag so that I couldn't login to my nextcloud either. There are waaaaaayyyyy more logs was trying to get in all day if they would help lemme know.
Reproduction Steps
Jellyfin Version
10.9.0
if other:
10.9.1
Environment
Jellyfin logs
FFmpeg logs
No response
Please attach any browser or client logs here
No response
Please attach any screenshots here
examples of what the browser looks like after authentification fails
Code of Conduct
The text was updated successfully, but these errors were encountered: