Nextcloud with Letsencrypt using OMV and docker-compose - Q&A

Zitat von Soma

First, the mariadb "latest", which is implied in your yml runs on mariabionic, which is incompatible (will give you errors on NC) with the "latest" version of NC (v21).

Is this something unique to SBC's?

I've always used mariadb:latest and have never had a problem and just tested it in a virtual machine and had it going in 5min. :latest definitely pulled the ubuntu version

version: "2"
services:
  nextcloud:
    image: ghcr.io/linuxserver/nextcloud:latest
    container_name: nextcloud
    environment:
      - PUID=1000
      - PGID=100
    volumes:
      - /NAS/AppData/nextcloud:/config
      - /NAS/Server/Media/.Nextcloud:/data
      - /etc/localtime:/etc/localtime:ro
    depends_on:
      - mariadb
    ports:
      - 450:443
    restart: unless-stopped
  mariadb:
    image: ghcr.io/linuxserver/mariadb:latest
    container_name: nextclouddb
    environment:
      - PUID=1000
      - PGID=100
      - MYSQL_ROOT_PASSWORD=password
    volumes:
      - /NAS/AppData/mariadb:/config
      - /etc/localtime:/etc/localtime:ro
    restart: unless-stopped

root@omv5-test:~# docker exec -it nextclouddb bash
root@e4cc58e9ffc0:/# cat /etc/os-release 
NAME="Ubuntu"
VERSION="18.04.5 LTS (Bionic Beaver)"
ID=ubuntu
ID_LIKE=debian
PRETTY_NAME="Ubuntu 18.04.5 LTS"
VERSION_ID="18.04"
HOME_URL="https://www.ubuntu.com/"
SUPPORT_URL="https://help.ubuntu.com/"
BUG_REPORT_URL="https://bugs.launchpad.net/ubuntu/"
PRIVACY_POLICY_URL="https://www.ubuntu.com/legal/terms-and-policies/privacy-policy"
VERSION_CODENAME=bionic
UBUNTU_CODENAME=bionic
root@e4cc58e9ffc0:/# 

I do agree though, it sounds like he's running those commands on the host rather than bashing into the container (OP, see example above on how to bash into the container)

Zitat von Soma

That's a good question

When I was on NC v20, there was a warning on the "Settings-> General View" about the upcoming v21 beeing incompatible with mariaDB 10.1 (which is used in the bionic version of linuxserver).

Maybe it's changed now but I really didn't saw it.

Hence the post on page27 to change the DB to webhippie (mariaDB 10.4)

For new ymls, it's better to apply the tag mariadb:alpine (mDB 10.5) since it will be used as "latest" soon

Alles anzeigen

Oddly enough, I just noticed my own nextcloud (my actual nextcloud install on my server) wasn't running. I'll check this real quick. Maybe it's incompatible with old databases, but with new ones it's fine?

Regardless, at least on x64 it does appear to work with latest.

As I understand it the mariadb:latest image will move from Ubuntu to Alpine. So I will stay on latest and see what happens when the base changes "soon".

I think there's been a change w/ the swag container because I can't get nextcloud back up (Looks like it was updated 4 days ago... I've been working a lot so I've not checked my nextcloud). I'm getting my certificate OK, and I've edited all my configs just as before.. but I just get a message this page cannot be managed via https when I go to my subdomain

I'll look at it later.

Zitat von Soma

If you haven't updated the image in a long time, you might have missed this warning on the nextcloud image:

Don't know if is what's giving you problems but,

First I heard of that... what needs to be changed there (I'm guessing that is the issue)..

Nevermind, I see what you're saying, remove it.

Zitat von Soma

Is it working now?!?

No, actually was able to test it just before I left this morning.. gonna work on it in a few after I eat.

Zitat von Soma

Well, if I recall correct, you use a different approach to the "sub-subdomain" that is created from "swag (letsencrypt)".

Haven't you edited the files in order to have access to nextcloud via "yoursubdomain.duckdns.org" instead of "nextcloud.yoursubdomain.duckdns.org" ???

If this is the case, you have to check all files you've changed (me thinks).

Yes true, but the image that "linuxserver/..." uses has, as base, the same version of mariadb (10.1)(maybe I'm wrong, have to check the "dockerfile").

armhf: docker-mariadb/Dockerfile.armhf at master · linuxserver/docker-mariadb (github.com)

Yeah, it seems that it only get's v10.4 for platforms other than armhf

Alles anzeigen

I've not really had time to look at it to be honest as work has been a little on the nutty side (moreso than usual).. I think I'm off Monday.. I'll look further at it then.. but you may be on to something there.

First sorry for the long post, just wanted to be thorough... this isnt' really my forte, so hopefully I don't look to stupid

So I got back to this today.. I think this goes along with this thread, if anyone feels differently I'll move it. I got duckdns working, but truth be told, I've been wanting to try a few things with my own domain in the future. So I figured now is as good a time as any... and purchased a cheap domain name from domain.com (just the domain, nothing extra.. like $4 for the year). In doing some reading, I decided to use cloudflare and subfolders. If someone has a better suggestion that may make this go easier, I'm all ears.

So first, I set up my domain to point at the cloudflare nameservers (not sure if it's important but also note I'm using the free cloudflare option, not paying) I'm also pretty sure I needed to setup CNAME and A records on cloudflare

I have 2 records on cloudflare I'm pretty sure are the relevant ones:

A mydomain.xyz my.public.ip on cloudflare (its proxied)

CNAME www mydomain.xyz (also proxied)

ken@laptop:~$ dig ns mydomain.xyz

; <<>> DiG 9.16.1-Ubuntu <<>> ns mydomain.xyz
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 28593
;; flags: qr rd ra; QUERY: 1, ANSWER: 2, AUTHORITY: 0, ADDITIONAL: 1

;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 65494
;; QUESTION SECTION:
;mydomain.xyz.        IN    NS

;; ANSWER SECTION:
mydomain.xyz.    3604    IN    NS    marge.ns.cloudflare.com.
mydomain.xyz.    3604    IN    NS    ben.ns.cloudflare.com.

;; Query time: 0 msec
;; SERVER: 127.0.0.53#53(127.0.0.53)
;; WHEN: Tue Apr 20 07:15:19 EDT 2021
;; MSG SIZE  rcvd: 99

ken@laptop:~$ 

ken@laptop:~$ dig CNAME www.mydomain.xyz

; <<>> DiG 9.16.1-Ubuntu <<>> CNAME www.mydomain.xyz
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 11996
;; flags: qr rd ra; QUERY: 1, ANSWER: 0, AUTHORITY: 0, ADDITIONAL: 1

;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 65494
;; QUESTION SECTION:
;www.mydomain.xyz.        IN    CNAME

;; Query time: 72 msec
;; SERVER: 127.0.0.53#53(127.0.0.53)
;; WHEN: Tue Apr 20 07:20:32 EDT 2021
;; MSG SIZE  rcvd: 48

ken@laptop:~$ 

Since I'm not getting an answer on my CNAME, I suspect that is part of the problem.. but I'll come back to that.

My docker-compse:

version: "2"
services:
  nextcloud:
    image: ghcr.io/linuxserver/nextcloud:latest
    container_name: nextcloud
    environment:
      - PUID=1000
      - PGID=100
    volumes:
      - /srv/dev-disk-by-label-D1/AppData/nextcloud:/config
      - /srv/dev-disk-by-label-D1/Disk_1/Media/.Nextcloud:/data
      - /etc/localtime:/etc/localtime:ro
    depends_on:
      - mariadb
    #ports:
      #- 450:443
    restart: unless-stopped
  mariadb:
    image: ghcr.io/linuxserver/mariadb:latest
    container_name: nextclouddb
    environment:
      - PUID=1000
      - PGID=100
      - MYSQL_ROOT_PASSWORD=something_super_secret
    volumes:
      - /srv/dev-disk-by-label-D1/AppData/nextclouddb:/config
      - /etc/localtime:/etc/localtime:ro
    restart: unless-stopped
  swag:
    image: ghcr.io/linuxserver/swag
    container_name: swag
    cap_add:
      - NET_ADMIN
    environment:
      - PUID=1000
      - PGID=100
      - URL=mydomain.xyz
      - SUBDOMAINS=www,
      - VALIDATION=dns
      - DNSPLUGIN=cloudflare
      - EMAIL=myemail@aol.com
    volumes:
      - /srv/dev-disk-by-label-D1/AppData/swag:/config
      - /etc/localtime:/etc/localtime:ro
    depends_on:
      - nextcloud
    ports:
      - 457:443
      - 91:80
    restart: unless-stopped

Deployed... everything worked fine except swag couldn't get a cert, which was expected. Made an edit to /AppData/swag/dns-conf/cloudflare.ini and added in my email and Global API from cloudflare... Also at this point, I copied nextcloud.subfolder.conf.sample to nextcloud.subfolder.conf as mentioned in the first post of the how-to Restarted swag... congratulations... got a cert. Then I remembered my CNAME isn't getting an answer, yet I got a cert? I thought maybe I was expecting to much to get an answer on the CNAME output.

Next, I move to editing nextclouds config.conf

<?php
$CONFIG = array (
'memcache.local' => '\\OC\\Memcache\\APCu',
'datadirectory' => '/data',
'trusted_proxies' =>
array (
0 => 'swag',
),
'overwritewebroot' => '/nextcloud',
'overwrite.cli.url' => 'https://mydomain.xyz/nextcloud',
'trusted_domains' =>
array (
0 => 'mydmoain.xyz/nextcloud',
),
'instanceid' => 'oc1km0frotj4',
);

and restarted the nextcloud container. Container logs show everything starting normally... go to

https://mydomain.xyz/nextcloud and I get an error that it's being redirected to many times and to clear my cache. So I fire up a browser in privacy mode, same problem.

I remember I had this problem once before with Nextcloud and duckdns, when I hadn't setup my server locally first, so on a whim, I enable the ports under the nextcloud section of the docker-compose file. Redploy, Go to my.ip.address:450. It seems to work, but the sign in screen is unlike any I've encountered with Nextcloud before (has all the same info just doesn't look right .. see pic)

I've not tried setting it up yet, as I figured something has went wrong since I can't get to it from mydomain.xyz/nextcloud although everything seems to be setup properly... I still believe there's an issue with my CNAME since I'm not getting an answer on it, but in looking at some various sites, it looks right.

Hope you didn't fall asleep reading my novel.