I’m still too container stupid to understand the right way to do this. I’m running it in docker under kubernetes and sometimes I don’t update nextcloud for a long time then I do a container update and it’s all fucked because of incompatible php versions of some shit.
I don’t remember much about how to use kubernetes but if you can specify a tag like nextcloud:28 instead of nextcloud:latest you should have a safer time with upgrades. Then make sure you always upgrade all the way before moving to a newer major version, this is crucial.
The solution for me is that I run Nextcloud on a Kubernetes cluster and pin a container version. Then every few months I update that version in my deployment yaml to the latest one I want to run, and run kubectl apply -f nextcloud.yml and it just does its thing. Never given me any real trouble.
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
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.
I have nextcloud running since nearly 5 years and it never failed once. Only dowtime is when the backup fails and somehow maintenance mode is still enabled (technically not a crash)
For those interested: Running in docker with mariadb in a stack, checking updates with watchtower everyday and pulling from stable, backups with borg(matic)
This is ultimately why I ditched Nextcloud. I had it set up, as recommended, docker, mariadb, yadda yadda. And I swear, if I farted near the server Nextcloud would shit the bed.
I know some people have a rock solid experience, and that’s great, but as with everything, ymmv. For me Nextcloud is not worth the effort.
I run it and mariaDB in docker and they run perfectly when left alone, but everything breaks horribly if I try to do an update. I recently figured out that you need to do updates for NC in steps, and docker (unRAID’s, specifically) defaults to jumping to the latest version. I think I figured out how to specify version now so fingers crossed I won’t destroy it the next time I do updates.
This is probably what I’m doing wrong. I’m using linuxserver’s docker which should be okay to auto update, but it just continuously degrades over time with updates until it becomes non-functional. Random login failures, logs failing to load, file thumbnails disappearing, the goddamn Collabora office docker that absolutely refuses to work for more than one week, etc.
I just nuke the NC docker and database and start from scratch every year or so.
You absolutely need to move from patch to patch and cannot just do a multiple version jump safely. You also need to validate the configs between versions, especially major release updates or you risk breaking. New features and optimizations happen and you also may need to change our update your reverse proxy configuration on update, or modify db table configuration (just puking this from memory as I’ve had to do it before). I don’t know that there’s automation for each one of those steps.
Because of that, I run nextcloud in a VM and install it from the binary package. I wrote a shell script that handles downloading, moving the files, updating permissions and copying the old config forward, symlinking and doing the upgrade. Then all I have to do is log in as administrator, check out the admin dashboard and make sure there aren’t new things I have to address in the status page. It’s a pain, but my nextcloud uses external db and redis and PHP caching so it’s not an easy out of the box setup. But it’s been solid for a long time once I adopted using this script.
There could probably be some additional refactoring here, but it works for my setup. I’m using default nginx paths, so they probably look different than other installs that use custom stuff like /var/www, etc.
Use it by putting it in a shell script, make it executable, then call it:
sudo scriptName.sh 28.0.1
Replace the version with whatever version you’re upgrading to. I would highly recommend never upgrading to a .0, always wait for at least a .1 patch. I left some sleeps in the when I was debugging a while back, those are safe to remove assuming it works in your setup. I also noticed some variables weren’t quoted, I’m not a bash programmer so there’s probably some consistency issues that could be addressed if someone is OCD.
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…
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.
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’m not self hosting an instance, but kbin is super fucking broken lately and it’s getting really frustrating. It’s been about a week. I submitted a ticket in their Git repo, but no response.
Weird. I’ve had a Pi-Hole + Unbound running on a Pi Zero since 2018 and it’s never had any issues. I expected the Zero to kinda suck but it has been nothing but smooth sailing. It gets USB power from my router and even if my router reboots the Pi also auto reboots itself.
I do next to no maintenance on it and it just keeps on chugging along. Maybe once every six months or so I SSH in and do a pihole -up and that’s it.
You are not logged in. However you can subscribe from another Fediverse account, for example Lemmy or Mastodon. To do this, paste the following into the search field of your instance: !selfhosted@lemmy.world
A place to share alternatives to popular online services that can be self-hosted without giving up privacy or locking you into a service you don’t control.
Rules:
Be civil: we’re here to support and learn from one another. Insults won’t be tolerated. Flame wars are frowned upon.
No spam posting.
Posts have to be centered around self-hosting. There are other communities for discussing hardware or home computing. If it’s not obvious why your post topic revolves around selfhosting, please include details to make it clear.
Don’t duplicate the full text of your blog or github here. Just post the link for folks to click.
Submission headline should match the article title (don’t cherry-pick information from the title to fit your agenda).
Always works great for me.
I just run it (behind haproxy on a separate public host) in docker compose w/ a redis container and a hosted postgres instance.
Automatically upgrade minor versions daily by pulling new images. Manually upgrade major versions by updating the compose file.
Literally never had a problem in 4 years.
I’m still too container stupid to understand the right way to do this. I’m running it in docker under kubernetes and sometimes I don’t update nextcloud for a long time then I do a container update and it’s all fucked because of incompatible php versions of some shit.
They have an "all in one" docker installer for the above because you are far from alone here.
Kubernetetes is crazy complex when comparing to docker-compose. It is built to solve scaling problems us self-hosters don’t have.
First learn a few docker commands, set some environment variables, mount some volumes, publish a port. Then learn docker-compose.
Tutorials are plenty, if those from docker.com still exist they’re likely still sufficient.
Yeah I’m only running it because truenas scale uses it
I don’t remember much about how to use kubernetes but if you can specify a tag like
nextcloud:28instead ofnextcloud:latestyou should have a safer time with upgrades. Then make sure you always upgrade all the way before moving to a newer major version, this is crucial.There are varying degrees of version specificity available: https://hub.docker.com/_/nextcloud/tags
Make sure you’re periodically evaluating your site with https://scan.nextcloud.com/ and following all of the recommended best practices.
The solution for me is that I run Nextcloud on a Kubernetes cluster and pin a container version. Then every few months I update that version in my deployment yaml to the latest one I want to run, and run
kubectl apply -f nextcloud.ymland it just does its thing. Never given me any real trouble.My Nextcloud has been flawless. The only issue I’ve had was NFS permissions. I have automatic update setup for docker so it stays up to date.
Care to share what broke?
Paperless often randomly stops accepting new documents. I have to wait several hours or restart it.
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
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.
I have nextcloud running since nearly 5 years and it never failed once. Only dowtime is when the backup fails and somehow maintenance mode is still enabled (technically not a crash)
For those interested: Running in docker with mariadb in a stack, checking updates with watchtower everyday and pulling from stable, backups with borg(matic)
++same
Docker:nextcloud+mariadb+caddy
This is ultimately why I ditched Nextcloud. I had it set up, as recommended, docker, mariadb, yadda yadda. And I swear, if I farted near the server Nextcloud would shit the bed.
I know some people have a rock solid experience, and that’s great, but as with everything, ymmv. For me Nextcloud is not worth the effort.
If all you want is files and sharing try Seafile
That’s what I’ve got running now, and for me Seafile is been rock solid.
I run it and mariaDB in docker and they run perfectly when left alone, but everything breaks horribly if I try to do an update. I recently figured out that you need to do updates for NC in steps, and docker (unRAID’s, specifically) defaults to jumping to the latest version. I think I figured out how to specify version now so fingers crossed I won’t destroy it the next time I do updates.
This is probably what I’m doing wrong. I’m using linuxserver’s docker which should be okay to auto update, but it just continuously degrades over time with updates until it becomes non-functional. Random login failures, logs failing to load, file thumbnails disappearing, the goddamn Collabora office docker that absolutely refuses to work for more than one week, etc.
I just nuke the NC docker and database and start from scratch every year or so.
Yeah I don’t like auto upgrades. Everyone says it’s fine but that’s not my experience.
My stuff isn’t public facing so I’m not worried about 0-days
You absolutely need to move from patch to patch and cannot just do a multiple version jump safely. You also need to validate the configs between versions, especially major release updates or you risk breaking. New features and optimizations happen and you also may need to change our update your reverse proxy configuration on update, or modify db table configuration (just puking this from memory as I’ve had to do it before). I don’t know that there’s automation for each one of those steps.
Because of that, I run nextcloud in a VM and install it from the binary package. I wrote a shell script that handles downloading, moving the files, updating permissions and copying the old config forward, symlinking and doing the upgrade. Then all I have to do is log in as administrator, check out the admin dashboard and make sure there aren’t new things I have to address in the status page. It’s a pain, but my nextcloud uses external db and redis and PHP caching so it’s not an easy out of the box setup. But it’s been solid for a long time once I adopted using this script.
Would love to take a look at that bash script (or at least a template of it) if you wouldn’t mind
Here you go:
https://pastebin.com/f5tL7xwx
There could probably be some additional refactoring here, but it works for my setup. I’m using default nginx paths, so they probably look different than other installs that use custom stuff like /var/www, etc.
Use it by putting it in a shell script, make it executable, then call it:
sudo scriptName.sh 28.0.1Replace the version with whatever version you’re upgrading to. I would highly recommend never upgrading to a .0, always wait for at least a .1 patch. I left some sleeps in the when I was debugging a while back, those are safe to remove assuming it works in your setup. I also noticed some variables weren’t quoted, I’m not a bash programmer so there’s probably some consistency issues that could be addressed if someone is OCD.
Thank you for taking the time ! This is a great resource
Sure! I’ll respond with a link in a bit.
For me everything works fine since years, EXCEPT collabora. I use onlyoffice now, it’s much faster and very stable
Not using Nextcloud. Found it a bit difficult to deploy and maintain than OwnCloud. Since then, I haven‘t had any problems with OwnCloud.
Wait so it’s not just that my vps only has 1gb of ram?
You guys with more ram still get crashes?
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…
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.
What issues did you have with the AIO docker?
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: alwaysvalue in the Compose config is pulling it’s weight lolI’m not self hosting an instance, but kbin is super fucking broken lately and it’s getting really frustrating. It’s been about a week. I submitted a ticket in their Git repo, but no response.
The most-recent release of lemmy dicked up outbound federation pretty badly on the instance I use.
For me it’s Pi-hole. For six months it runs fine, then dies so horribly I resort to snapshot rollback and we both pretend it never happened.
Weird. I’ve had a Pi-Hole + Unbound running on a Pi Zero since 2018 and it’s never had any issues. I expected the Zero to kinda suck but it has been nothing but smooth sailing. It gets USB power from my router and even if my router reboots the Pi also auto reboots itself.
I do next to no maintenance on it and it just keeps on chugging along. Maybe once every six months or so I SSH in and do a
pihole -upand that’s it.Give technitium a go, my woes diminished drastically with that.
!RemindMe 7d
Invidious. It’s to be expected for something like that though.