I won’t update without first creating an image of the server to roll back to. Like others on here, the web updater almost always fails and goes into maintenance mode and I have to ssh in to fix it.
Having said that, functionally, I have no issues. Only when upgrading does the whole thing shit the bed.
I’ve hosted mine for years on my own bare metal Debian/Apache install and 28 is the first update that has been a major pain. I’ve had the occasional need to install a new package to enable a new feature, or needed to add new/missing indices to the database, but the web interface literally tells you how to do those things, so they’re not hard.
28 though broke several of the “featured” apps that I use regularly, like “Retention”. It also introduced some questionable UI changes that they had to fix with the recent .1 update. I’ll get occasional errors when trying to move or delete files in the web interface and everything. 28 really feels like beta software, even though we’re a point release in and I got it from the “stable” update channel.
I’m on my laptop so I thought I would elaborate on my first comment to give you things to watch out for if/when you update. I’ve been hosting mine with the zip file manually installed with my own Apache/PHP/MySQL/MariaDB setup for ages now without issue. It’s been rock solid except for, like I said, the occasional changes required to take advantage of new features such as adding new indices to the database or installing an additional php addon. Here’s the things that I noticed with updating to 28.
The 3 dot/ellipses menu was missing in the web interface and was replaced with dedicated buttons for “Download”, “Add to Favorites” and “Delete”. Shift clicking was also broken. This meant that when I, for example, take a lot of photos for a holiday, I can’t use the web interface to select a large range of multiple files and then move them all from “InstantUpload” into a more permanent album. I either had to use the mobile app, or do them one at a time. The ellipses menu, along with the options to bulk “move/copy” have been added back since then with the *.1 update, but shift clicking in the web interface to select a range of files is still broken.
The “Retention” app, which is listed as a “Featured” app doesn’t function any more. I used it to automatically delete backups of my Signal messenger, files in the “InstantUpload” folder that were over a year old, etc. You can enable it, but it doesn’t actually work and just throws errors in the log file, which is now reported in the “Overview” portion of the “Administration” page with a note of “X number of errors since somedate”, and prevents you getting the green checkmark. It’s probably safe to assume that other apps will also have issues because I had half a dozen get automatically disabled with the update.
Occasionally when I use the web interface to move or copy a file, I’ll get an error message that the operation failed. Sometimes this is true, sometimes it’s not and the operation actually succeeded. If it ends up being true and the move did actually fail, doing it again results in a successful move.
It seems like they’ve made some substantial under-the-hood changes to the user interface that shouldn’t have been shipped to the “stable” channel. It’s not completely broken, it “is” usable, especially after they restored my bulk move/copy button, but I still can’t use the Retention app, at least last time I looked, so I’ve literally got daily cron scripts to check those folders for old files and delete them, then trigger an occ files:scan of the affected directories to keep the Nextcloud database in sync with the changes. This however, bypasses the built-in trash bin so I can’t recover the files in the event of an issue. I actually considered rolling back to 27 for a bit, but decided against it, so if I were you, I would stick with 27 for a while and keep an ear to the ground regarding any issues people are having that are or aren’t getting fixed in 28.
Works great for me. I had it running in a snap for awhile, but now I just have it in a proxmox Debian container running a LAMP stack. I have over a terabyte of stuff saved and multiple computers syncing too, so its well used.
Installed it in k3s and then pulled up the Android app but all it does is say every single file is a duplicate and overload my notifications tray while not uploading anything
I gave up on owncloud just before it became nextcloud because it kept breaking every time I updated it.
Wallabag is similar for me now. I’m stuck on a slightly out of date version because I can’t get newer ones to run. Everything else I self host is painless though.
I didn’t realize that next Cloud was so bad, might I recommend people having issues try Seafile? Also open source and I’ve been using it for many years without issues. It doesn’t have as many features and it doesn’t look as shiny but it’s rock solid
I’m having a hard time believing that… There is a difference between being able to fix the update issues every time without problems or having no problems at all. But if so, neat.
I haven’t had any issues with Nextcloud yet. But any torrent client refuses to work. I’ve tried various qbittorrent containers, transmission, deluge briefly, they all work for a while but eventual refuse to do anything.
I had TOTP die for one user on my Nextcloud. I tried to disable it, but it “didn’t exist”. I tried to enable it, but it was already enabled. It would come up when I used occ twofactorauth:state user. I ended up fixing it by (force) disabling the app and re-enabling it. It didn’t break any other user’s TOTP and it fixed problem-user’s TOTP. No idea what went wrong, but I get these random issues with Nextcloud sometimes.
The plus side to this is I’ve learnt how to use Mariadb and I’ve gotten better at debugging things.
Well dang, I have Nextcloud installed as a snap (which has been perfectly stable for me when running on Ubuntu Server), but I was thinking of switching over to a docker installation; this thread doesn’t exactly fill me with enthusiasm for that idea…
Anecdotal, but Ive had a container running Nextcloud in an LXC on Proxmox along with PiHole, Step CA, Bacula, and quite a few other services and I’ve had zero downtime since June 2023. Even have Tailscale rigged to use PiHole as the tailnet DNS to have adblocking on the go.
Guess that restart: always value in the Compose config is pulling it’s weight lol
I ended up on the snap because I couldn’t get the AIO install working properly. My snap version has been super solid. I think I’m gonna stick with it for a while.
Why is your Collabora server on local host? Local host will always point to the device you are trying to access from. You need a publicly accessible URL
“Local host” in this instance refers to my desktop computer where all my super sweet Linux distros are saved. Nextcloud Office, while being an “app” appears to not have any function without the collabora; i.e. there will be no document viewer without the collabora “server” running next to NextCloud.
Or maybe it’s none of that. Coming from a Windows background, running docker is completely foreign.
Collabora need to be accessible at the URL you provide. As a example one might have nextcloud at nc.example.com and Collabora as cb.example.com. you would need to enter cb.example.com as the URL.
The easier way it to use the Nextcloud all in one image or the build in Collabora. Its not going to be as robust or fast but its much simpler.
I found a service that syncs our calendars self-hosted. That was the only thing that was missing. Can’t remember the name, works flawlessly and without any problems for a number of years now. If you are interested, I’ll look it up next weekend.
I want my docs and files on a self-hosted cloud (I can’t seem to get sftp, ftp, or sharing to work on windows 11 even after adding the missing features) , with the ability to at least open the contents without downloading them. I want to stop using google for calendars and notes, and it would be handy to have a self-hosted bulletin board I and my added users could write on.
According to the box, nextcloud does all these things, except that it doesn’t, without practically rewriting the code and somehow re-engineering linux to not be a fucking cunt.
I’ve tried and tried. It just won’t work. Maybe I need to get a different firewall program. I’m working in Pro, added the features, made the firewall exceptions, have my network setting as “private,” I’ve done everything. The host will be visible on the network, but logins time out or fail altogether.
Since writing my rant, I found HFS, which, though an OLD program, was stupid stupid easy to set up.
I also found Filebrowser, and though the config was way more of an asspain than it should have been, it’s fucking awesome. I’ve even moved on to trying to get HTTPS running for external connections using Win-Acme, but it isn’t going well.
Please do! I had spent solid day researching open source CALDAV server/clients to replace Google calendar for my boss. Almost no options on that front.
I have used Baikal for caldav for the server, with davx5 on Android. Was solid. Moved to NC for files, so went ahead with calendar sync on NC too. NC calendar sync has already worked well for me, no hiccups.
The only issue I’ve had with NC is auto upload of photos from my phone. It constantly has conflicts. Otherwise sync of regular files works great.
Add comment