Nextcloud: Adding local server yields error 400

[Jimmyson]

I am looking to test the PhotoPrism Backup functionality with my local Nextcloud server. Both services are running in Docker containers, and both are exposed and accessible. I am utilizing HTTP rather than the HTTPS, as the endpoints are both contained within my local network.

When adding the Nextcloud server to PhotoPrism, I am informed that a connection could not be established.
Looking over the Logs, I am presented with the following information. I attempted a connection twice

2020-09-10 11:17:18 DEBU POST /api/v1/accounts (400) [55.223016ms]
2020-09-10 11:17:18 DEBU api: abort /api/v1/accounts with code 400 (Could not connect, please try again)
2020-09-10 11:17:18 ERRO could not connect
2020-09-10 11:17:09 DEBU GET /api/v1/status (200) [43.564µs]
2020-09-10 11:16:36 DEBU POST /api/v1/accounts (400) [53.83915ms]
2020-09-10 11:16:36 DEBU api: abort /api/v1/accounts with code 400 (Could not connect, please try again)
2020-09-10 11:16:36 ERRO could not connect
2020-09-10 11:16:07 DEBU GET /api/v1/status (200) [50.171µs]

Using Cyberduck, I am able to connect to both PhotoPrism and Nextcloud using WebDAV, confirming that both instances are accessible.

[lastzero]

Did you try connecting without providing the exact path, just http://celluloid:8080/? We're performing auto-detection for common WebDAV path patterns. Not sure if your browser might filter or block passwords sent via http, I haven't seen a warning like in your screenshot before when testing locally. In general, you should always use HTTPS if possible, especially on servers that are exposed to the public internet.

Also note that Nextcloud WebDAV support might be broken under certain conditions, see WebDAV: Uploading to Nextcloud may result in 0 byte files #443.

[Jimmyson]

I have tried that address too, and still receiving the error.
My server is not exposed to the Internet, but my internal network is resolving the calls.
I am not utilizing HTTPS for any of my Docker Instances, as (from my understanding) it requires a Domain, and I am not sure how to setup HTTPS for a local, internal server. Reverse Proxies are a thing, but I am not understanding the documentation to set it up...
The web browser I am utilizing is Firefox Dev 81.0b9

[lastzero]

Are you sure the Nextcloud server can be properly reached from within the Docker container using the host name? Name resolution can be different in some cases. Do you see anything in the logs of your Nextcloud server so that you know the connection works on the network level? Any helpful error messages?

[Jimmyson]

Are you sure the Nextcloud server can be properly reached from within the Docker container using the host name? Name resolution can be different in some cases.

I am a little unsure on what you mean here.

Do you see anything in the logs of your Nextcloud server so that you know the connection works on the network level? Any helpful error messages?

Nothing in my Nextcloud logs are indication connection errors. Attaching to the Docker container does not log any GET or POST HTTP requests coming from Photoprism. Although, I am able to see requests from Firefox and also Nextcloud itself.

[lastzero]

It seems the PhotoPrism container cannot reach the Nextcloud instance for some reason, maybe because the host name celluloid only works on your host and not from within the PhotoPrism container? Is it configured in your local /etc/hosts file? What is the IP address of celluloid?

[Jimmyson]

My system hosts file contained a record for 127.0.1.1 celluloid
My system's IP address is a static IP set by my router for 192.168.1.4. I added such record to the hosts file, and no luck.

Nextcloud Hosts:

root@99a60b068cc4:/var/www/html# cat /etc/hosts
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
172.18.0.4      99a60b068cc4

Photoprism Hosts:

root@338d3676cf17:/photoprism# cat /etc/hosts
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
172.18.0.3      338d3676cf17

Looking over the Docker Network setup I have a couple of interfaces setup.

jimmyson@celluloid:~$ docker network ls
NETWORK ID          NAME                     DRIVER              SCOPE
4e560fc93795        bridge                   bridge              local
93af30e04fa7        host                     host                local
b452fee5e1f0        jimmyson_default         bridge              local
332995b3c2e4        jimmyson_gitea-network   bridge              local
e7c1c66af704        none                     null                local

Inspecting the Network interfaces of both Nextcloud and Photoprism, they are sharing the same Network Interface...

jimmyson@celluloid:~$ docker inspect photoprism -f "{{json .NetworkSettings.Networks }}"
{"jimmyson_default":{"IPAMConfig":null,"Links":null,"Aliases":["photoprism","338d3676cf17"],"NetworkID":"b452fee5e1f0c309865bebdda82ca90ac03b980c56b8d2481b9bdca46846a96e","EndpointID":"9a7f4d19bb89b9a80909522b556364e4cb3d3653cf2c605e91b184c62e2b8e45","Gateway":"172.18.0.1","IPAddress":"172.18.0.3","IPPrefixLen":16,"IPv6Gateway":"","GlobalIPv6Address":"","GlobalIPv6PrefixLen":0,"MacAddress":"02:42:ac:12:00:03","DriverOpts":null}}
jimmyson@celluloid:~$ docker inspect nextcloud -f "{{json .NetworkSettings.Networks }}"
{"jimmyson_default":{"IPAMConfig":null,"Links":null,"Aliases":["99a60b068cc4","nextcloud"],"NetworkID":"b452fee5e1f0c309865bebdda82ca90ac03b980c56b8d2481b9bdca46846a96e","EndpointID":"f60226cf3d58c236199cf09ee6c20c3fbeeb2a2bb473f3f288a06187ddfe3644","Gateway":"172.18.0.1","IPAddress":"172.18.0.4","IPPrefixLen":16,"IPv6Gateway":"","GlobalIPv6Address":"","GlobalIPv6PrefixLen":0,"MacAddress":"02:42:ac:12:00:04","DriverOpts":null}}

[lastzero]

Without spending much time going through your complete networking setup, it seems obvious that PhotoPrism gets HTTP error 400 (invalid request) when trying to connect to 127.0.1.1 (localhost) via WebDAV as it will connect to itself.

Docker Compose typically takes care of networking so that you don't need to enter any IP addresses into /etc/hosts. Also you typically don't let it access your host for security reasons, it's a container ("jail") after all.

Here, for example, the photoprism service may connect to photoprism-db simply by using the name as host:

https://dl.photoprism.org/docker/docker-compose.yml

[lastzero]

Any progress / feedback on this? Was my suspicion going in the right direction?

[Jimmyson]

I haven't had any luck so far with the WebDAV access issue, but have not touched it for the last week.

[Elmardus]

I have the same issue, adding either the full path or the docker internal hostname, I both receive the same 400 error:

time="2020-11-10T15:58:15Z" level=error msg="could not connect"
time="2020-11-10T15:58:15Z" level=debug msg="api: abort /api/v1/accounts with code 400 (Could not connect, please try again)"
time="2020-11-10T15:58:15Z" level=debug msg="POST /api/v1/accounts (400) [235.410629ms]"

With the following Service URLs:
https://nextcloud.domain.tld/remote.php/dav/files/Username/
or:
http://nextcloud-app/remote.php/dav/files/Username/

[lastzero]

Can the domains nextcloud-app and nextcloud.domain.tld be resolved from inside the container? Do they maybe point to localhost or 127.0.0.1?

[Elmardus]

They resolve properly in the container, I used getent hosts nextcloud-app and getent hosts nextcloud.domain.tld, they resolve to the nextcloud-app container IP and my public IP respectively.

[Elmardus]

Anything else I can try? I would love to use Photoprism, but whatever I try I can't link up Nextcloud.