I have pushed out the update. The changelog is in github and maybe they are just busy with Nextcloud conference currently to update the website.

@girish said in Social Well Known:

I tested this out quickly. Indeed, the app auto creates an id of the form @nc.domain.com . After domain change, it doesn't adapt. But I don't think this is a bug in the app. Most of the federated apps (even outside nextcloud) cannot change their domain post installation.

image.png

I thought the same!

It seems to be a bit of a bugger that, on the other hand, I located the folder for well-known but nowhere does it show the URL in the files in the well-known folder for the app.

I have messaged the developer of the app to see if they could put in the description to only install on a URL that will not change or find a way around it but I don't think the developers of federated apps have it high on their priority to make domain name changes possible.

IMO, it's better to make these how to's SEO friendly with proper description and content. From what we have seen, the docs are not read as a manual like in the good old days.

People search -> hit forum post is the most common flow. The forum is waaaaaaaay more popular in SEO than the docs.

Whatever caused this, either a Cloudron update or some self-solving stuck/bug in NextCloud fixed it. It works now.

@3246 said in [💡 Guide] How to install Cloudron/Nextcloud with LUKS full disk encryption on Hetzner cloud server:

@Stardenver yes it is. I also have it running on baremetal at home/office.

My new set up has a main drive that's encrypted and another for local backups, which is also encrypted but unlocks after decrypting the main boot/data drive.

Happy to expand on this if folks find it useful 🙂

Thanks again for your setup guide. Just got a new server and its up and running. May I ask how you'd setup your system, so that additional drives are also encrypted and unlock after decrypting the main drive?

It was so long ago that I set it up now that I can't remember, but this is what I have in my domain section of the SnappyMail Admin Panel:

image.png

And in General -> Main I've got this:

image.png

And in Login, I've got this:

image.png

And under Extensions, I have:

image.png

Hopefully that helps?

The htaccess from /app/data/htaccess is linked into the app code folder https://git.cloudron.io/cloudron/nextcloud-app/-/blob/master/Dockerfile#L69

Instead of copying file by file you should tar the files as @robi suggested and upload it via file manager. You can then extract it (file manager has an Extract option).

meh.

but thanks for helping me find the guilty one 😉

@girish that's not the steps - that's an outcome 🙂

I didn't expect NextCloud instance without LDAP somehow got converted to the one with LDAP 🙂

@doodlemania2 I believe it's visible to all in the call that it's being recorded. Prefer that to separate stealth options. And, it's not as-if every company you call nowadays doesn't have a "we record your calls" disclaimer at the beginning, and they seem to be mostly unscathed.

remarkable behavior:
login via web does not work at the moment.
open the website via the nextcloud-desktop client does work and one is logged in.
but then, password-verification i.e. for module-update does again not work.

@nebulon your hint with the incompatible apps was true both problems.

the login problem. which went away after deleting dropit and the cron pbl. as well, after disabling Reader (ebook reader) and there was a recommended app, which also threw errors: `group-folders'
I leave it here for reference

thanks again

@jagan And I restarted the whole cloudron server and lo and behold, the app came back online.
I was watching the logs, it sputtered and crackled, but in the end, came through.
Thank you Cloudron for all this magic.

Per https://github.com/PaulLereverend/NextcloudDuplicateFinder/issues/111 and https://github.com/PaulLereverend/NextcloudDuplicateFinder/issues/110, it's not working properly

@robi you are welcome, glad you found it useful and thanks for your feedback.

I want to prevent unauthorised access to the volume when the server is offline as well as somebody who might take the server out of its rack and boot it. If the volume automatically mounts, they have full access to the data if they can circumnavigate the login cough recovery mode cough

That's covering for physical access to the machine, in what likely would be a read-only, forensic access scenario by a 'bad' actor. In my case, it's a virtualised server, so not as likely to matter, unless they move the VM to a dedicated slice to get busy with it.

Granted, this is somewhat unlikely but being paranoid is better than being sorry 😄

I am in the UK and storing professional data abroad (in nasty Europe no less - just kidding!) means encryption can cover the requirements nicely (as far as I know - I am not an expert in that field, so somebody please correct me if I'm wrong).

Latest update seems to have fixed this.

well, i got it fixed by deleting the collabora server and installing a new one under another subdomain, but I'm pretty sure it's just again a question of time until that breaks once more 😕

@marcusquinn Thanks.

In the recent Hub5 presentation, NextCloud's CEO showed clear commitment to data sovereignty for users. I think it is great how Nextcloud could as easily help individuals and families as much as businesses.