Volatile

Ah, okay. thanks. I never worked with docker before and have to read through this. May take a while, but for what are weekends anyway.

I have a problem accessing the newly installed Jitsi (conferencing). What I did: Jitsi is installed via APT into the system, not a customer, and an A entry to the DNS has been made, pointing to the (sub)domain which should deliver the GUI, something like: meetings.my-domain.com. After installation, one is asked to provide this name by the installer, and a set of configuration files, including LetsEncrypt stuff in generated, i.e. (probably more) /etc/sysctl.d/20-jvb-udp-buffers.conf /etc/sysctl.d/50-IPv6.conf /usr/lib/sysctl.d/50-pid-max.conf /etc/sysctl.d/99-nc-kernel.conf /usr/lib/sysctl.d/99-protect-links.conf /etc/sysctl.d/99-sysctl.conf /etc/sysctl.conf Now, when I open meetings.my-domain.com, I get however the Froxlor default "this domain is not configured" page. So I created the subdomain in the admin panel, but now I get of course the index.html of the resp. folder. I guess, I have to change a virtual host file, so Froxlor doesn't deliver the "not configured" page, but I don't know which file exactly and how to do it properly so it doesn't get overridden by Froxlor the next time I (re)configure something. Additional Info: This is the quickstart of steps, in case someone needs to know: https://jitsi.github.io/handbook/docs/devops-guide/devops-guide-quickstart/ I didn't do the optional FQDN step, because i wasn't sure if this breaks Froxlor or anything else, and I didn't secure the domain yet, but this only prevents anyone from create a conf. room, which isn't the issue atm. So, If anyone maybe installed Jitsi before or knows what to do, any help would be appreciated. Thanks.

I can't create subdomains as admin, at least I never saw the button for that. well, the point is, I am the only one having several domains for different purposes. I also host a small webpage for a friend, who then should also have access to the webmail (service) like roundcube. I could put the services in my (me us customer) home dir, but no one except me is doing all the maintenance here anyway. I just use two customers to have my webpages hierarchically divided from that of my friend, otherwise I would ditch customers completely and do this all as "admin" if I could. It's me, after all But thanks so far. I guess I do it with softlinks then. It costs to much time to just get it done otherwise.

Hi again, sorry, but I may misunderstand open_basedir and path settings. So I want to install basic services like nextcloud in a general place, where several subdomains can point to. Until now, I used softlinks, but I think, It would be better to do this via path. I added ":/var/services/" to "Settings » Webserver settings » Paths to append to OpenBasedir". Then, in the "Create subdomain" -> Path = "/var/services/nextcloud/" and in the drop down of "OpenBasedir-path" I selected "Path from field above". I expected that this would set to folder at (root)/var/services/nextcloud, as it should be seen as valid destination, but instead, the folder is created at (root)/var/customers/webs/$USER/var/services/nextcloud/ Before I mindlessly try every possible constellation and break the server, I just ask, what I do wrong/misunderstood with the path settings and how to point to subdomain correctly. BTW I also tried to go down directories with /../ but it's converted to single dot.

Ah, well, I just read now in the signature, you are the lead developer. So I guess, no forwarding required. 😁 Thanks again for the help!

Just to inform, uninstalling gnupg2 via APT resolved it and now the extension is being found. so it was a conflict with the system package. Thanks again and maybe forward my suggestion about the warning indication. See ya.

Thanks so far. Okay, so I understand, that the yellow indication is not related to the PHP version, but the missing Module. Don't get this wrong or as an offense, but because of the layout (how it is written), it's not obvious for someone who doesn't know this. For an "outsider", duo to perception, the yellow "not okay" is being applied to the PHP version, not the missing module described below it. If I do a hardware restart, the webserver will forcibly be restarted as well. I'll take a look at some log-files. If it's not a php-version issue, maybe the module isn't started correctly or something like that. Maybe a conflict with the gnupg2. I try to uninstall that and see what happens. If someone has another suggestion where to look, feel free to post. Thanks!

I rented a new root-server and currently setting up the basics, including Froxlor, like on my old system. The server came preinstalled with a Debian 12[Bookworm] (netinst, minimal) I installed Froxlor via APT acc. to the webpage. Now, in the web-interface of the Froxlor installer, I get two messages: 1.) [yellow exclamation mark] Your system is running with PHP 8.2.26 Now the yellow indicates it not a problem, but also not good. There is no indication what the problem is. PHP version to high? to low? my pixel just run out of blue ink and it's in real green? Here it would be nice (for a future version), if it writes, what it warns for (As it internally checks for something to decide the color, that shouldn't be a problem) 2.) The following php extensions could not be found but are recommended gnupg So I installed gnupg2 via APT as well as gnupg via pecl, added the "extension=gnupg" line in "php.ini", restarted the whole server via ACPI and checked with "php -m" if the extension is running; and it's listed. Yet, Froxlor claims the module is still missing. What comes to mind: - Do I have to register gnupg somewhere else for froxlor to find? - Is it, that I installed Froxlor via APT before installing gnupg and the check was done during install time? - Is the PHP version [warned about] insufficient to work with gnupg and Froxlor? I know it's not required (it writes it on the install page), but now that I have a clean new system, I don't want to start with holes in the first service I setup Anyone have a suggestion and run into the same problem?