Anubis error - 500 Internal Server Error

[Beckid]

I have Anubis installed. It's up and running, but it fails to work properly.

I have this line uncommented at the top:

- "AUTH_REQUEST_ANUBIS_UPSTREAM=http://127.0.0.1:8923" # scheme + ip/hostname/domain (+ port) to internal anubis, without path, this is unrelated to OIDC

And here is my Anubis section in my NPMplus docker compose file:

# See the readme on how to configure your proxy hosts to use anubis and what needs to be changes in the botPolicy to make it work
  anubis:
    container_name: npmplus-anubis
    image: ghcr.io/techarohq/anubis:latest
    restart: unless-stopped
    network_mode: host
#    ports:
#      - "127.0.0.1:8923:8923"
    environment:
      - "TZ=America/New_York" # needs to be changed
      - "TARGET=http://192.168.1.3:81" # This is the IP to my local NPMplus instance. I tried HTTPS as well and didn't work, I see Anubis for a second and then it goes to a 403 forbidden page, cloudflare, but only when I enable https. http goes to a 500 error message page
      - "POLICY_FNAME=/etc/botPolicies.yaml" # required so the status codes can be changed
    volumes:
      - "/volume2/docker/npmplus-anubis/anubis.yaml:/etc/botPolicies.yaml:ro" # since a file is mounted here, you first need to create it before deploying this container, you can download the default file from here: https://raw.githubusercontent.com/TecharoHQ/anubis/refs/heads/main/data/botPolicies.yaml

 

As the last comment in the Anubis section above mentions creating the botfile, I did that. I also downloaded the default .yaml file in the link provided.

NEXT STEPS:

I login to my npmplus web UI, and I pick a host to test with NPMplus/Anubis. For example, let's pick the first host. I edit the first host, go to the bottom section to the dropdown list for Auth Request and I select Anubis.

NEXT UP, ERROR MESSAGE IN LOG:

Everything should be working, but when I go to test.mydomain.com, I can see the Anubis spash screen for a half of a second, and then it goes to a 500 Internal Server Error white screen.

I check the npmplus-anubis log in my Docker and this is what I see:

{"time":"2026-05-18T11:45:40.778267665-04:00","level":"INFO","source":{"function":"github.com/TecharoHQ/anubis/lib.(*Server).issueChallenge","file":"github.com/TecharoHQ/anubis/lib/anubis.go","line":137},"msg":"new challenge issued","subsystem":"anubis","host":"127.0.0.1:8923","method":"GET","path":"/.within.website","user_agent":"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/148.0.0.0 Safari/537.36","accept_language":"en-US,en;q=0.9","priority":"u=0, i","x-forwarded-for":"2600:1700:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX","x-real-ip":"2600:1700:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX","challenge":"019e3bc3-XXXX-XXXX-XXXX-XXXXXXXX"}
{"time":"2026-05-18T11:45:41.181007909-04:00","level":"INFO","source":{"function":"github.com/TecharoHQ/anubis/lib.(*Server).issueChallenge","file":"github.com/TecharoHQ/anubis/lib/anubis.go","line":137},"msg":"new challenge issued","subsystem":"anubis","host":"127.0.0.1:8923","method":"GET","path":"/.within.website","user_agent":"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/148.0.0.0 Safari/537.36","accept_language":"en-US,en;q=0.9","priority":"u=1, i","x-forwarded-for":"2600:1700:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX","x-real-ip":"2600:1700:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX","challenge":"019e3bc3-XXXX-XXXX-XXXX-XXXXXXXX"}

 

And if I actually go to the Anubis page: http://192.168.1.3:8923, I see this error message: https://imgur.com/XJlg7n0

Any help would be appreciated. Thank you!

[Beckid]

UPDATE: Duhh... the TARGET in my Anubis section needs to be http://192.168.1.3:8923 which is the ANUBIS docker install. NOT NPMplus.

It's working slightly better. Now when I visit one of my hosts, I am stuck on this screen and it won't proceed:

[Zoey2936]

NPMplus/compose.yaml

Line 150 in b849444

# - "TARGET= " # important to be set to a space

it need to stay one single space

[Beckid]

OMG, like LITERALLY a space 😂 😂

When I read the word "space" I thought it meant it meant something else, and English is my first language LOL! 😂

Maybe update that line to say something like "don't touch this, leave this line alone."

I feel like an idiot now 🤦‍♂️

It's all working now!!

Thank you, @Zoey2936 !!

[Zoey2936]

I've updated the wording a bit

[Beckid]

Anubis is working. I see the splash screen now and then it forwards me on to my host.

Two final things @Zoey2936 . I'm not sure if you can help, but I figured I would ask.

1. If I go to http://192.168.1.3:8923/ I still see this error. Do I just disregard it:

 
 

2. I run a self-hosted docker container of Vaultwarden (it's a password manager using Bitwarden's code.) I enabled Anubis for my vault.mydomain.com. I can't login to the Bitwarden browser extension after enabling Anubis. If I disable Anubis protection on my vault.mydomain.com it works just fine.

Any idea on this? I'm assuming I'm going to have to whitelist something somewhere. Maybe you could point me in the right direction. Thank you!

[Zoey2936]

For 1) this should be fine, anubis is not made to be opened directly if in subrequest mode
for 2) Yes, this is to be expected, since the extension probably behaved like an API client and not like a browser anubis will block it, I don't use anubis currently myself, so I can't help you much, but maybe you can try to open vaulwarden once normally with the browser, so that bitwarden maybe keeps the anubis allow cookie

[Beckid]

Great, thank you @Zoey2936 !

Yeah, you're probably right about the API. You are correct: I can login to Vaultwarden once, and the browser plugin works, but unfortunately the extension logs out every 30 days. I would have to disable Anubis once a month. I will probably just leave Anubis disabled on this host.

In one of my other hosts behind Anubis, it's blocking images in the web UI for my UGreen NAS. Looks like I'll have to disable Anubis for this host as well.

For the most part it works great! Just a couple of hosts broken behind Anubis.

[Beckid]

For 1) this should be fine, anubis is not made to be opened directly if in subrequest mode for 2) Yes, this is to be expected, since the extension probably behaved like an API client and not like a browser anubis will block it, I don't use anubis currently myself, so I can't help you much, but maybe you can try to open vaulwarden once normally with the browser, so that bitwarden maybe keeps the anubis allow cookie

@Zoey2936 I found the solution to fix the Vaultwarden/Bitwarden browser extension to work when it's being protected behind Anubis! Here is the resolution, I am cross-posting this for anyone that finds this through Google: #2664 (reply in thread)

The Fix: Use Custom Locations to Bypass API Paths

In NPMplus, go to your Vaultwarden proxy host → Custom Locations tab and add separate locations for the API paths that don't have Anubis selected.

Add these locations, each pointing to your Vaultwarden upstream, with Auth Request set to "None":

Location Path	Forward to	Auth Request
/api/	http://vaultwarden:8080	None
/identity/	http://vaultwarden:8080	None
/icons/	http://vaultwarden:8080	None
/notifications/	http://vaultwarden:8080	None

Of course, set the "Forward To" that matches your setup. Vaultwarden is installed on my NAS, along with all my other Docker containers. So, my Forward To is set to http://192.168.1.3:8000 (port 8000 is what my Vaultwarden install is set to.)