Yesterday I just updated the mailcow as it was a new Moorch 2025 Update. I understood that SOGo is now just / and /admin has to be admin panel.

I am using Nginx reverse proxy where earlier I was redirecting my domain to /SOGo with the following (after update I commented the redirection so that it directly goes to default.):

location / {
#rewrite ^/$ /SOGo/;
proxy_pass http://127.0.0.1:18080;
proxy_set_header Host $http_host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Proto $scheme;
client_max_body_size 0;

proxy_buffer_size 512k;
proxy_buffers 128 2048k;
proxy_busy_buffers_size 2048k;

}
}`

Yes, mailcow.conf contains http port binding to 18080 which had been working since many years.

But after update, nginx restarting, docker restarting, server itself restarting process, the mailcow UI is not coming on the screen. Neither the default UI not the /admin panel nothing is coming on the screen.

if anyone can help on this.

Have something to say?

Join the community by quickly registering to participate in this discussion. We'd like to see you joining our great moo-community!

Cisco30 I know that too. I do not have any additional ADDITIONAL_SAN in mailcow.conf, and set ENABLE_SNI=n.

The thing is domainname.com/rspamd working fine, it’s just mailcow UI and admin panels are stuck.

    I simply rolled back the update with –legacy option and everything came back to normal. Could not find the root cause why the default mailcow UI or /admin /domainadmin was not working.

    4 days later

    I tried again the Moorch 2025 update to assess the issue with –stable update. But still everything stays blank and no page comes on the screen.

    Do I have to run generate_config.sh again? if anyone can help.

    I guess you are creating a loop here. You rewrite / to /SOGo and the March update does rewrite /SOGo to /.

      esackbauer No earlier config was to route default UI to SOGO e.g. rewrite ^/$ /SOGo/; in proxy pass. I just commented it so that it routes to default / where it goes to whatever it presents by proxy_pass http://127.0.0.1:18080;.

        4 days later

        I experience the same issue. I have quite a standard setup, just some adjustments in postfix. Cannot find out why the page is blank. / /admin /domainadmin - all the same. Just blank. Log in nginx-mailcow:

        nginx-mailcow-1 | <IP> - - [05/Apr/2025:20:59:27 +0200] “GET / HTTP/2.0” 200 20 “-” “Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/135.0.0.0 Safari/537.36”

        What can cause this?

          kingfisher

          Probably you updated from a much older release? Could be that the MariaDB upgrade did not go through properly.
          See this issue and the linked one regarding MariaDB:
          mailcow/mailcow-dockerized6429

            I updated from the most recent release, as i always keep Mailcow up to date. Usually i do not update at the forst day but after a week or so.

            What i did is export the database (with Adminer), stopped the mysql container, removed the contet fo the mysql volume, started the mysql container and imported the database export with Adminer. Everything went smooth, the Mariadb log reported nothing unusal. I cannot see a relation to the database why my Mailcow UI does not work anymore in the browser. E-Mail flows are still working. Nothing suspect in any logs.

            How would i debug this animal further?

              kingfisher What i did is export the database (with Adminer), stopped the mysql container, removed the contet fo the mysql volume, started the mysql container and imported the database export with Adminer.

              Why did you do that?

                  I also always updated my mailcow on the same day when it shows it has a new update. So in my case also it was recent fresh update too. I also did not see any unusual in logs so I just rolled back it to legacy update. it’s good to know that I am not alone with this issue as @kingfisher user is also have the same setup like mine and having the same issue. If you get any solution to this issue please share the same with me too bro.

                  DocFraggle

                  @esackbauer mentioned mailcow/mailcow-dockerized6429

                  I looked into index.php and have added this line at the beginning:

                  var_dump($_SERVER['DOCUMENT_ROOT']);

                  /web was printed

                  until i added it under the line

                  require_once $_SERVER['DOCUMENT_ROOT'] . '/inc/header.inc.php';

                  After that it was not printed anymore.

                  I looked into index.php and have added this line at the beginning:

                  var_dump($_SERVER['DOCUMENT_ROOT']);

                  /web was printed

                  until i added it under the line

                  require_once $_SERVER['DOCUMENT_ROOT'] . '/inc/header.inc.php';

                  From that line on /web was not printed anymore.

                      No, it was just a message to describe that i try to solve the thing :-) index.php is “producing” the white page, because no content is beeing generated. I wanted to understand why. Better a Mailcow developer should have a look.

                        kingfisher … haha thanks for the update. I will not try to update it now. Was just thinking to update it and your message came. Thanks a lot.

                          I am wondering how to debug this. I have another Mailcow with the same version which updated fine. The problematic one is from 2022. The good one from Mid 2024.

                          I compared a lot of settings, including database contents, with no obvious signs of a problematic source. I tried to understand what index.php does, but without a deeper understanding of the architecture, how everything is connected, I’m not very comfortable going in that direction. A bit out of ideas now.

                          I do not want to go back to legacy …