Also noted a request for a List view here: https://forum.cloudron.io/topic/3465/dashboard-customize-buttons-on-the-sides-of-apps/4?_=1608519749716

@Mario I actually ran into this myself and had to combine my two private registries into one to work with Cloudron.

@girish Oh, absolutely with a custom app, I mentioned in the original post that I'm going to build my own Wordpress Stack to simply and only add this. But I saw benefit to other user's being able to script something post-update as that's literally the only thing my stack will do differently than the default Wordpress app (I'll have to integrate your updates into my stack manually each time so I still wish there was a way to run an external custom sh script post-installation).

As you can see my custom script simply uses the CLI to upgrade all databases with any new required formatting if and only if any updated Wordpress core or updated plugin require it. When you update from the Cloudron interface, it simply updates all the files and Wordpress has this really annoying tendency to not upgrade the databases post-upgrade invisibly. The script above is the only way to make sure file versions and database versions stay in sync 100% of the time every update.

This becomes nearly unavoidable if you want to support multisite in the future since the problem becomes more convoluted in that installation type so my script detects multisite installations and runs database upgrades accordingly with WP CLI commands or single site only commands if it's a single site.

I actually think the script below should be an optional "automatic upgrade plugin and themes checkbox when updating Wordpress core" option for Wordpress installations as well, here's the WP CLI code for it:

# Plugin updates wp plugin update --all wp plugin update --all --skip-plugins --skip-themes # Theme updates wp theme update --all wp theme update --all --skip-plugins --skip-themes

@hendrikvl Do you have any notes/instructions on how you changed the datadirectory ? I think we might have to leave a note in the packaging code not to overwrite data directory to not break your installation!

Indeed this is on our roadmap.

@msbt yeah, just a safety-fuse, I must spend most of my time online designing how things can scale, and most of that is systemising safety by design in structure and defaults.

I recently noticed a DMARC policy on one setup was generating mass emailing I hadn't considered, less haste, more speed 🙂

I'd like to add to this. The ability to select the pass length and whether we want special characters. Or if it's easier to do, allow us to add our own passwords instead of them being generated for us.

Adding to this: Maybe it's possible to implement a backup browser via the file browser, at least for the rsync format and even for encrypted backups?

@robi @girish thanks, I figured it might be something like that as I think it was just after a reboot that I noticed they had disappeared.

@jdaviescoates nope, just regular domains, not sure why but will just wait and see I guess

@nj Apart from what you mentioned, I think for 1) there is also the issue that we somehow need to update the 2FA inside the app's database when the cloudron 2fa changes. Recently, I saw that some apps like rocket.chat can pull 2FA from LDAP. I haven't looked into it closely but maybe some sort of standardization is happening in this space.

Can consider this for next release nevertheless. It's actually very easy to implement, the hard part is to not confuse end users. But really, all the hard work has to be done the Cloudron admin to communicate to their users.

Would be nice to have ClamAV available as an app for custom installations/API use. Looks like there is a docker image now:

https://hub.docker.com/r/mkodockx/docker-clamav/

Just to follow up, here's a sample of normal backups followed by a Cloudron upgrade, which itself triggered another backup run, and the corresponding relevant network and disk graphs:

Network Traffic.png Disk I_O.png

All in all, it's definitely fast-er but not insanely performant. CPU utilization vs load hints that it may in fact be down to inefficient utilization of cores to some extent, but there is definitely a fair bit more bottleneck coming from the network still.

CPU Utilization.png CPU Load.png

Nothing earth-shattering either way, and gains were more mild than I would have guessed, but all in all, not a bad outcome.

@robi Yes, more checks to keep the mounting "alive".

@fbartels We can always have a simple runner page preload indicating the rehydration of the app and how to avoid it.

Gamify it a bit..

Indeed, I have add a Media section now for next release.

+1 I often forget where Stop and Restart are found!

And it doesn't make sense to me that they are found in different places

Stop is hidden in Uninstall
Restart is in Repair

@jdaviescoates Having three items like so would be nice:

Capacity: xxxxxGB
Used: xxxxxGB
Remaining: xxxxxGB

@nebulon Ahh okay that makes it clearer why this style was chosen.

I can live with that explanation. 👍

@nebulon Great guys! You're doing amazing !!!