Posts by d4vid

    I have an OMV4 Installation with Nextcloud, MariaDB and Let's Encrypt Dockers based on Techno Dad Life installation guide.


    After updating my Nextcloud 16 Docker via the built-in update dialog in the Nextcloud UI, initially everything seemed fine. Then I got the message to exit "maintenance mode". After having done this by clicking in the UI, it seems not to be able to connect to the database anymore. The log file says:


    Unfortunately I realized too late, that I should have updated the Nextcloud Docker itself somehow.


    I hope so much that this can be solved. Thank you!

    I have following nextcloud subdomain config in the let's encrypt docker:



    But as a result I get following error message:


    Code
    2019/07/11 15:11:42 [error] 369#369: *5 nextcloud could not be resolved (3: Host not found), client: 192.168.0.1, server: get.acme.net, request: "GET /favicon.ico HTTP/2.0", host: "get.acme.net", referrer: "https://get.acme.net/"



    What could be the reason?
    Why is the "nextcloud" host not found?

    After re-checking the nginx error.log I see:


    Code
    2019/07/11 15:11:42 [error] 369#369: *5 nextcloud could not be resolved (3: Host not found), client: 192.168.0.1, server: get.acme.net, request: "GET /favicon.ico HTTP/2.0", host: "get.acme.net", referrer: "https://get.acme.net/"




    which relates to this line in the letsencrypt proxy configuration:

    Code
    set $upstream_nextcloud nextcloud;


    I also tried those with no success:


    Code
    set $upstream_nexcloud 192.168.0.111;
    set $upstream_nextcloud get.acme.net;


    The latter even seems to create a loop in the server ("*762 768 worker_connections are not enough while connecting to upstream, client: 192.168.0.1,").



    Why can the "nextcloud" variable not be found? I guess it's the name of the docker image, which should be there.

    I have the same issue.


    1. I updated and renamed "nextcloud.subdomain.conf.sample" - As a result the traffic is correctly forwarded to the Nextcloud docker.
    2. I updated "nextcloud/config/config.php" to trust my domain, added "trusted_proxies" and do the host/protocol overwriting:



    3. Now I had access-logs in the Nextcloud Nginx Log directory. But the browser screen always says "Bad Gateway".

    I thought so too, but I could only find a "config.sample.php" there.


    So maybe the best is to do a fresh re-install. Luckily I didn't have much modification done on the previous Nextcloud installation yet.

    For some reason my Nextcloud Docker image was not listed anymore in the "running" section.


    1. So I did a backup of the the "sharedfolders/AppData" and "sharedfolders/Nextcloud" folders.
    2. Started the Nextcloud docker again, with the according config, data and port parameters.


    Now, when I'm trying to access Nextcloud via https://192.168.0.11:444, I get following:


    Screen Shot 2019-07-10 at 15.22.40.png


    Digging deeper into the nginx logs it looks like it cannot connect to the DB. Probably because the IP of the DB has changed:


    Code
    [error] 330#330: *2 FastCGI sent in stderr: "PHP message: PHP Fatal error: Uncaught Doctrine\DBAL\DBALException: Failed to connect to the database: An exception occurred in driver: SQLSTATE[HY000] [2002] No such file or directory in /config/www/nextcloud/lib/private/DB/Connection.php:64


    How and where do I have to update the Nextcloud config, when I don't get the "Nextcloud Installer GUI" served?



    Thanks in advance.

    Thanks Morlan!



    Quote from Morlan

    as long as the certificates fail, the reverse proxy inside the container will not start.

    That brought some light in my not-understanding. :)




    Quote from Morlan

    The port forwards look fine. I think you can delete the extra traffic rules.


    It turned out the actual problem was deeply rooted in the overall home/wan network setup. Someone had done some weird wiring and setup there. So this had nothing to do with OMV.

    Okay, now I'm suspecting, that the port-forwarding may not work correctly. Or the let's encrypt nginx server is simply not running.


    First I thought the forwarding might be correct, because of this error message in the let's encrypt logs:



    Code
    Detail: Invalid response from
    http://omv.acme.net/.well-known/acme-challenge/zbe50NOV2kgDMC-8uzj7aTnWtximj9wVJv5LRAMnaGc
    > openmediavault - Page not found


    But that's because OpenMediaVault used to be running on port 80. Now, since I changed OMV to port 81, the error message changed to this:


    Should it work, if OVM is served at Port 80 and the Let's Encrypt connections forwarded from Port 80 to 90? And if yes - how is it supposed to be working?


    To @Morlan's question, here are screenshots from my dockerconfig:


    Screen Shot 2019-07-09 at 20.01.37.png
    Router - Port Forwarding:


    Screen Shot 2019-07-09 at 20.14.06.png
    The router also has a section "Traffic Rules" where I replaced those forwards (Not sure if that's a good idea?) and where I also explicitly opened 443 and 444 for SSL:


    Screen Shot 2019-07-09 at 20.17.54.png

    I have OpenMediaVault 4 with the Let's Encrypt Docker running.
    I have done the configuration as per Techno Dad Life's video (https://www.youtube.com/watch?v=pRt7UlQSB2g)


    All the port forwarding should be done correctly.
    But when trying obtain a cert, by starting the letsencrypt docker image, it results in an error:


    The issue is that the www directory "http://omv.acme.net/.well-known" is not available. I didn't see any instructions, that it should be created manually.


    At which point and how should it be created?
    Is there some nginx config wrong?