cat2devnull

Just thought I would post some more information on this error. I have had the same problem with a system always powering down ~3-4min after being turned on. Jul 15 21:08:33 Tower elogind-daemon[1185]: Power key pressed. Jul 15 21:08:33 Tower elogind-daemon[1185]: Powering Off... Jul 15 21:08:33 Tower elogind-daemon[1185]: System is powering down.. Jul 15 21:08:33 Tower elogind-daemon[1185]: Suspend key pressed. Jul 15 21:08:33 Tower shutdown[7531]: shutting down for system halt This is on a brand new vanilla USB that had just been created by the USB Creator for both 6.12 and 7.0-beta2. There was no issue in memtest or when booting Windows. After a bit of digging it seems to be due to USB. I had a USB keyboard and LCD monitor (USB powered) connected. I'm not sure if they were pulling too much power or the keyboard was sending a power off command over USB. Anyway, just unplugging the devices fixed the issue. Edit: Seems to only happen if the keyboard is plugged into USB on system boot. Hmmm...

So you will need to setup 2 domains nextcloud1.mydomain.com and nextcloud2.mydomain.com Then point both of those externally at you external routers IP (which hopefully has DDNS). Port forward 80/443 (HTTP/HTTPS) to the address of a reverse proxy such as nginx. Also point your internal DNS server for both at nginx. Configure nginx to use Let's Encrypt for certs and point the two domains at their respective servers. Nginx will then look at the request header of incoming requests and based on the domain will forward the request for both servers. When you setup like this, your devices (phones, laptops, etc) will work seamlessly on your lan or WAN. Your other option is to setup a service like Tailscale so all your devices appear to be on the LAN 24/7 and then you don't need to setup any external DNS CNAMEs or run nginx.

If you want access from outside then you need to open port 443 (HTTPS) and point it at your NGINX instance. NGINX will look after the security side of things. Take a look at some of the IBRACORP videos to get an understanding of how it all works together. You will need DDNS and a domain name but it's totally worth the effort. I suggest that you watch a bunch of videos from IBRACORP and SpaceInvaderOne and get your head around how all this technology works. A few hours of research will save you a lot of time in the long run. Generally this is done on a per computer and per browser basis, so will be dependent on what you run. Most browsers have an option when they encounter a self signed certificate to "trust" it in the future.

Running NGINX is for local or remote access. If you truely only ever intend to access your NextCloud instance from a machine or two on your LAN then the solution is to add the lsio cert to the trust store on your machine and your done. Otherwise install NGINX and let it manage the HTTPS certs. I made a post for Vaultwarden to explain how it works for both LAN and WAN connections but you can just replace Vaultwarden for NextCloud. This is the better way to set things up. Trust me, it won't be long before you find other uses for NGINX. I initially set it up for Vaultwarden but now it manages NextCloud, Plex and a bunch of other services.

The idea with NextCloud and other applications such as Bitwarden, is that you access the app via a reverse proxy such as NGINX. This can then manage certificates automatically from Let's Encrypt.

So you have to get NextCloud to rescan the existing data folder and add all the files to the database. php occ files:scan --all

Great to hear. Just go the the latest supported versions, either 25 or 26. I would upgrade the docker to the latest 19.x release first and then upgrade NC to 19.x. You can find the correct docker tag by using the search function and searching for "19." Rinse and repeat until you are up to date. It will take a number of iterations since you can only go one major release at a time. After each upgrade check if NC reports any issues under Administration Settings -> Overview. You may need to run a few database commands by hand after some of the updates. Just remember in future to upgrade both docker and NC a bit more often so you don't get out of sync again. 😀 Not that you need this anymore but yes. You just reinstall, point the data folder at the appropriate directory and then run "php occ files:scan --all" and it will rescan the folder and add all the files to the database.

Since you are only using it for file storage. If the downgrade doesn't work then just install a fresh copy of 26.0.2 and move your files across. That should take less than an hour of work and has the advantage that you will be on the latest version. SpaceInvader did an updated NextCloud install video in 2022 which is very easy to follow.

Great so you're config and version files both seem to be in sync with version 18.0.3 so that's a good start. I think your upgrade then downgrade of the docker has introduced this PHP issue. You must be on a version (I think you said 7.4.33) that is not compatible with 18.0.3. According to the PHP release history, 7.4.33 was the last 7.4 release and came out in Nov 2022. It is probably too new for NC 18.0.3 which is from years earlier. You could try downgrading to 18.0.3-ls76 which is the linuxserver.io version for NC 18.0.3 but I don't know if that specifies an older PHP as well or if it will just pull the latest. But it's worth a shot. If not the next step would be trying to do a downgrade of PHP within the docker but I have no idea how to do that.

Ok so we at least we have a GUI again, even if it won't let us do anything because it thinks we are mid upgrade. Go into the docker shell and run the command "occ status". You should see something like; root@1384826c6432:/# occ status - installed: true - version: 25.0.5.1 - versionstring: 25.0.5 - edition: - maintenance: false - needsDbUpgrade: false - productname: Nextcloud - extendedSupport: false Then check what the version.php thinks; cat /config/www/nextcloud/version.php Also check what version the config file thinks; cat /config/www/nextcloud/config/config.php | grep version All of these should be in agreement. I suspect that they are not because of a previous failed install. I thought step 3 is where it turns on maintenance mode. So it would be good to know what mode "occ status" thinks it is in. You could try to repair the database; occ maintenance:repair As for the actual error; Could not resolve OCA\Settings\AppInfo\Application! Class OCA\Settings\AppInfo\Application! Class OCA\Settings\AppInfo\Application does not exist in /config/www/nextcloud/lib/private/AppFramework/Utility/simpleContainer.php:106 I found a bit of stuff here, but I have never really done any PHP so I can't say if this is a problem because you are still running an incompatible version of PHP or if you have a half upgraded/corrupted copy of NextCloud. I've never had an upgrade fail. If you only use NextCloud for file storage then you might find it easier to restore from a backup or just rebuild the docker from scratch. If that's not an option then in theory assuming your database is intact and your config folder is correct (see my earlier comment about 'occ stats', version.php and config.php needing to be in sync) then you should be able to create a new NextCloud2 docker using an appropriately old version from linuxserver.io Start by downloading an old version without pointing it at your appdata folder and make sure you have one that is using an appropriate 18.x version of NC and a suitable PHP7.x Once you have that, then point that new docker at your old appdata folder and it should fire back up.

You can see all the available tags on their page. Just edit the NextCloud docker on Unraid and change the repository line to read lscr.io/linuxserver/nextcloud:<tag> where <tag> is the version you want to roll back to. I'm not sure what was the last docker version to ship with PHP7 so maybe try 24.0.12 The documentation says they moved to PHP8 on 10/10/22 but it's unclear if that was in all releases or only docker 25.x and 26.x NextCloud 18.x is really old. Without a really compelling reason to be on an unsupported platform from 2021, I would recommend upgrading to a version that is still getting security patches. Even after upgrading to 18.0.14 you won't be able to go back to the "latest" docker version because you still won't be running a version that can support PHP8.

There are plenty of posts over the last couple of months about this. Basically you have been updating the NextCloud docker but not updating NextCloud itself. There are a few options, you could downgrade the docker, get NextCloud running again on PHP7 and then update via the GUI. The other option is to upgrade via the command line within the docker shell. sudo -u abc php /config/www/nextcloud/updater/updater.phar occ upgrade When upgrading you can only update one version at a time. So the question is what version are you actually on, and what version is it trying to upgrade to? This seems to be an issue that a few users have been hitting recently. It may be the case that if you upgrade to the latest docker, hit the PHP issue and then try to upgrade NextCloud it gets into a weird state where the upgrade process fails part way through and the versioning gets out of sync. Eg, it is actually on v22 but thinks its on v23 so attempts an upgrade to v24 (if that makes sense). You can check as per my earlier post here.

I didn't see anything concerning in the logs. Are you getting errors access nextcloud from just the web interface or also the native client? Might be worth deleting any old browser cookies/cache and trying again.

Great stuff, starting to get somewhere. This is just because we had to hand edited the file so we broke the hash. Not a big deal. Ok, sorry that's on me. I gave you the wrong command because I didn't read the upgrade docs carefully. updater.phar prepares the system for an upgrade and performs the steps listed; Check for expected files Check for write permissions Enable maintenance mode Create backup Downloading Extracting Replace entry points Delete old files Move new files in place Once that is done, it is ready to perform the upgrade which uses the "occ upgrade" command. So you're making progress. Looks like you are on 24.0.12. FYI running "docker exec -it nextcloud updater.phar" from Unraid is the same as running "updater.phar" from the NextCloud docker directly. 2 steps forward 1 step back So first just make sure you are on php 8 as 24.0.12 can't run on php 7. Run "php -v" from within the docker. The log files of interest are; /config/log/php/error.log /config/log/nginx/access.log /config/log/nginx/error.log This is probably a permissions issue... But that's an educated guess. See if there are any errors in the logs. Sorry this is taking so much back and forth but I have never hit this type of corruption during an upgrade and I'm not an expert. I know just enough to be dangerous. If anyone else on this forum knows what their doing, feel free to jump in.