I had a horrible experience with nextcloud on a pi. I have a great experience on a good server. It stores files. It does that very well for me. Clients work reliable.
Nextcloud apps are sometines good and sometimes not. It can do everything. One should let navidrome serve music and not nextcloud. Mealie is for recipes. Jellyfin for videos and immich for images. Paperless for documents.
Nextcloud is file storage, backup, syncing and maybe collab. That’s what I expect and that’s what I get.
I wanted to spin up OCIS but for some reason ran in to difficulties with the Docker container. I forget what the issues were, but I already had a solid Nextcloud instance running so I didn’t dig very hard. Would like to revisit it some day.
However, since then Owncloud has been bought out, causing some worry.
I find it really weird that something as simple as the basic functionality of nextcloud seemingly can’t be implemented in a stable and lightweight manner.
Nextcloud always seems one update away from self destruction and it prepares for that by hoarding all the resources it can get. It never feels fast or responsive. I just want a way to share files between my machines.
There are other solutions, I know, but they’re all terrible in their own way.
I use owncloud infinite scale and overall its rock solid. The downside is the lack of plugins. Nextcloud has been nothing but trouble for me and every update was a mess so I decided to try OCIS and for my need I was extremely satisfied.
Now, I admit, I’m not one to get carried by the drama in the FOSS sphere (still use Gitea) but I do agree there is an history to the separation of owncloud and nextcloud that can make some people uncomfortable. Having a choice is good I believe.
There was something wonky with the mapping of OIDC attributes to user properties, so I decided to look at the seahub source and see if it would be easy to fix.
Turns out, the whole thing is held together with hope and spit. Literal beginner code.
Any tips for speeding it up? Loading can be painfully slow at times. I was reading that it may be the database (I use MariaDB which in theory shouldn’t limit it with 32gb RAM and an R7 1700x).
I reverse proxy over tailscale to a VPS because I have double NAT… The connection to the VPS is direct with wireguard at least, no relay node. Adds ~30ms latency. But even when I connect direct locally it’s not substantially faster.
I’ll check my config.php for caching. I don’t recall adding anything for it so if it’s not on by default then that’s a likely reason. Thanks!
Yeah for my instance I remember I had to do a few performance optimizations. Also make sure you are up to date because nextcloud has gotten performance upgrades.
I initially set up Nextcloud with MariaDB on spinning disk but it was slow even completely empty. I moved that container to ssd & performance was a lot better. The web UI may still have some slow loading parts but I can’t say for sure since rarely use it. Caldav+carddav+Nextcloud client are how I usually interact with it.
Nobody cares about your home services unless they can use them to send spam or mine bitcoin. Owncloud is a funny name because it seems to imply it’s for personal use: your own cloud. I didn’t know until I found myself in one, but apparently a lot of schools use Owncloud.
phpinfo is just a bad idea. It’s a built in facility that dumps everything without knowing whether it’s sensitive or not, right into the current page, making it trivial to add this vulnerability to your own application or library that an unsuspecting developer will include into their application. There’s not even a single security warning in the documentation. Here’s practically the same problem from 21 years ago: nvd.nist.gov/vuln/detail/CVE-2002-1725
The only thing that makes this case worse in docker is that more info is in ENV variables. The vulnerability has nothing to do with containers though, and using ENV variables to provide sensitive data is in general a bad decision, since they can be leaked to any process with /proc access.
Unfortunately, ENV is still a common way which people use to pass data to applications inside containers, but it is not in any way a requirement imposed by the tech.
Absolutely not. Many applications used ENV variables for sensitive stuff even before. Let’s remember that the vulnerability here is being able to execute phpinfo remotely.
Containerization can do good for security, in general.
This is just a bad practice that was popularized by CI/CD solutions and later on by containers. I’m not saying containers aren’t good for security, what I’m saying is that they’re misused and abused and that images shouldn’t even be a thing. Isolation is great, blindingly trusting images made by someone and/or having people that don’t have any basic knowledge of security nor infrastructure suddenly being able to deploy complex solutions with a click ends up in situations like this.
OK, but how do you solve the problem? Trusting an image is not so different than downloading a random deb and installing it, which maybe configures a systemd unit as well. If not containers you still have to run the application somehow.
Ultimately my point is that containers allow you to do things securely, exactly like other tools. You don’t even have to trust the image, you can build your own. In fact, almost every tool I add to my lab, I end up opening a PR for a hardened image and a tighter helm chart.
In any case, I would not expose such application outside of a VPN, which is a blanket security practice that most selhosters should do for most of their services…
My point was that “random deb” and/or “random web application” are way less likely to come with unsafe default ENV based configuration files and usually go with the config files securely stored in system directories with the proper permissions enforced during installation or simple .php config files that won’t get exposed and that will require the user to configure in a proper way (like WordPress does by providing wp-config-sample.php but not the final wp-config.php file required to run it). Those are the solutions people used before the containerization hype and things were just fine.
My second point: containers “lowered the bar”, allowing for almost anyone to be able to deploy complex solutions and this was / is bound to be a disaster. No matter how safe Docker and others become we can’t just expect people who know almost nothing about computers and networking to be able to safely deploy things. Even the ones that know a lot, like developers, sometimes use Docker to deploy things they wouldn’t be able to deploy otherwise and fall to the pitfalls of not understanding networking and computer security.
In any case, I would not expose such application outside of a VPN, which is a blanket security practice that most selhosters should do for most of their services…
Well, me too, however I understand that some people might want to expose it publicly because they might want to access their instances from public or work machines where they can’t install a VPN. For those cases I would recommend 2FA and only allowing specific IP addresses or ranges access to the thing - the office static IP, their ISP or eventually only the user’s country.
It's probably best to look at what the devops industry is embracing, environment variables are as secure as any of the alternatives but poor implementations will always introduce attack vectors. Secret management stores require you to authenticate, which requires you to store the credential for it somewhere - no matter what there's no way to secure an insecure implementation of secrets access
They are not as secure, because there are less controls for ENV variables. Anybody in the same PID namespaces can cat /proc/PID/environ and read them. For files (say, config file) you can use mount namespaces and the regular file permissions to restrict access.
Of course you can mess up a secret implementation, but a chmod’d 600 file from another user requires some sort of arbitrary read vulnerability or privilege escalation (assuming another application on the same host is compromised, for example). If you get low-privileged access to the host, chances are you can dump the ENV for all processes.
Security-wise, ENV variables are worse compared to just a mounted config file, for example.
In general, a mounted file would be better, because it is easier to restrict access to filesystem objects both via permissions and namespacing. Also it is more future proof, as the actual ideal solution is to use secret managers like Vault (which are overkill for many hobbyist), which can render secrets to files (or to ENV, but same security issue applies here).
I’m using Kubernetes and many of the apps that I use require environment variables to pass secrets. Another option is the pod definition, which is viewable by anybody with read privileges to K8s. Secrets are great to secure it on the K8s side, but the application either needs to read the secret from a file or you build your own helm chart with a shell front end to create app config files on the fly. I’m sure there are other options, but there’s no “one size fits all” type solution.
The real issue here is that the app is happy to expose it’s environment variables with no consideration given to the fact that it may contain data that can be misused by bad actors. It’s security 101 to not expose any more than the user needs to see which is why stack dumps are disabled on production implementations.
The problem is in fact in the applications. If these support loading secrets from a file, then the problem does not exist. Even with the weak secrets implementation in kubernetes, it is still far better than ENV variables.
The disappointing thing is that in many “selfhost” apps, often the credentials to specify are either db credentials or some sort of initial password, which could totally be read from file or be generated randomly at first run.
I agree that the issue is information disclosure, but the problem is that ENV variables are stored in memory, are accessible to many other processes on the same system, etc. They are just not a good way to store sensitive information.
owncloud.com
Hot