Cloudron Forum

I bet at some point we save more time as Cloudron users by packaging apps ourselves and documenting the process, maybe improving it via LLM, than waiting for the roadmap

@plusone-nick said in Focus on Business Apps:

imo the main issue in regards to new apps is the developer experience/on-boarding

Cloudron has had their own "standardized" method for packaging apps for many years now and have basically failed in facilitating a thriving marketplace for Open Source and new custom apps. Way before Supabase, Coolify and others "We"/Cloudron have had: Server/Runtime, Auth, DBs, Backups, DNS & so on readily available with absolutely no streamlined method to build an app using Add-On pieces and simply launch it.
...

And don't get me started with the fact that we still don't have a single "no code/low code" APP/PAGE builder! No lie it makes me lose sleep at night...We can design a UI with Penpot, build databases with several options, build workflows with n8n, do AI stuff with Typebot/Open WebUI but I only have fucking wordpress to build a page!? (excuse my french & no offense WP but really!? it's basically 2025!!)

...

I know, I know before anyone says it! Sounds like a "skill issue" this is all just linux, docker and open standards, languages, protocols so you should know all of this already, its ASSumed that you do...and this is where the problem/current attitude toward "new apps" derives and resides.

...
Don't mistake my passion for hate or anger

TLDR:

1. App/Page Builder
2. Better Docs/Dev Onboarding
3. Compose & Swarm Implementation/Unified Dashboard

I pray this does not continue to fall on deaf ears...Especially since I am willing and able to contribute...
Everyone is here for a mutually beneficial relationship, let's keep that the focus or part ways amicably.

For points 1 and 2, I totally feel you. Anyway there is some hope and I want to share my experience below.

as an experienced developer and devops person, I have to admit it requires some skillset/experience and patience in order to be able to make the effort to package an app. As a busy parent, I had cooked a simple python app, very minimal, and packaged it for alpine, to boost my productivity. Then I wanted to host it on my Cloudron and it was not so easy to make it work, due to permissions issues, various errors (exec format etc), failed healthcheck... Despite having a working Dockerfile and docker-compose.yml in local, it took me 2 weeks of spare time to adapt my app for Cloudron. Mostly because my spare time is very limited and my knowledge of Cloudron packaging is limited too, and also I had a goal to only use ChatGPT.

The base Cloudron image is mostly ubuntu based while I wanted to keep alpine as a base image, and despite this difference I could manage to yet install my app on my Cloudron instance while still keeping it minimal

It took me a few ChatGPT prompts + a few evenings to go over all the issues I faced, yet I succeed! I believe the DevEx is quite horrible at this time, but with patience and help from ChatGPT , and good documentation I'm convinced I could tackle any such challenge as long as the effort is realistic and the app is worth it. The more we cook app, the easier it becomes, and the funnier it is.

I will try to share a bit about this effort soon , so at least people can benefit from the learning. What is interesting, when building your own app, with a minimal setup, is that it is possible to avoid to rely too much on Cloudron "magic" as I don't care at all about the base image, etc.

I had to admit that the information I needed to aggregate for the initial prompt to ChatGPT was huge and sourced from many different parts (docs, git...) and is not an effort everyone is willing to take, so it is worth cooking some ChatGPT app for this I believe.

Anyway the good news is that it is possible to have sort of a recipe to repeat this success, i.e by feeding ChatGPT with examples of Cloudron/Dockerfile/start.sh of similar apps; i.e apps made with similar tech stack; + relevant excerpts from the documentation about packaging apps for Cloudron and Cloudron requirements. And in total it took maybe a few hours of prompting to ChatGPT and git/docker push/cloudron install/update to make the whole thing work.

I hope to build maybe one or two more apps for Cloudron, for my own needs, this time I'll focus on apps I didn't build myself like maybe soulseek and others. Maybe in the end I'll have a good tutorial to help people contribute to Cloudron with sensible apps.

In 2023, Uptime kuma was updated to support Chromium for one of the watches, which is reflected in the Cloudron package update here:
#4141 Added support for /snap/bin/chromium (Ubuntu's default Chromium path) (https://forum.cloudron.io/topic/5388/uptime-kuma-package-updates/77)

And Chromium appears effectively as one of the possible watches:
but I couldn't make this work as I don't know the path to chromium on the app as packaged in Cloudron.

Could it be that this was not mentioned yet and not implemented yet as part of Cloudron package? Or am I missing something?
I don't see any mention of Chromium in the packaging: https://git.cloudron.io/packages/uptime-kuma-app/-/blob/master/Dockerfile?ref_type=heads
And at runtime, when trying to configure Chromium in Uptime Kuma, the app complains the path to chromium is not found,
In the settings, trying to configure a path shows this if no valid path is given, I guess no default chromium instance is setup, but I might be wrong.

This could potentially be delivered as part of the package, as done for Changedetection maybe? https://git.cloudron.io/packages/changedetection-app/-/blob/master/Dockerfile?ref_type=heads

What do you think?
Thanks in advance for the great and hard work

@girish About "app is down", I'm interested in a way to tune "restart automatically" for app being down. It's not the first time I notice an app is down or not responding because it crashed during the night and in the morning I have to manually restart it.
it happened for instance that my wordpress blog was stopped due to high traffic because of being on the frontpage of hacker news... and so was down for several hours... what a pity

@joseph : for now I'm using a cron job (code below) and Cloudron API to check every 10 min which app is down and restart it, is it possible to mimic some basic autoRestart as part of Cloudron platform instead?
That would be nice to free ourselves from maintaining scripts like below even if I truly love scripting and fixing problems myself

#!/bin/bash
API_TOKEN="<REDACTED>"
CLOUDRON_URL=${CLOUDRON_WEBADMIN_ORIGIN}
APPS="<APP_ID> <APPNAME.MYDOMAINE.TLD>
<ANOTHER_APP_ID> <APPNAME.MYDOMAINE.TLD>"
echo "$APPS" | while read id url; do
  status1=$(curl --max-time 15 -s -o /dev/null -w "%{http_code}" "https://$url")
  status2=$(curl --max-time 15 -s -o /dev/null -w "%{http_code}" "https://$url")
  if [ "$status1" -ne 200 ] && [ "$status2" -ne 200 ]; then
    curl --max-time 5 -s -X POST "$CLOUDRON_URL/api/v1/apps/$id/restart" -H "Authorization: Bearer $API_TOKEN" \
    && echo "$url restarted ✅" \
    && php /app/data/scripts/send_email.php <EMAIL>  "Cloudron Healtcheck" "Service at $url was automatically restarted!"
  fi
done

I do agree that as a paid customer I feel voting in the forum is a waste of time and also more and more of my questions in various topics are just not answered or the priority/ backlog is not visible to us so it feels like it's going nowhere.
I've seen several topics and complains like this very one but staff gives an impression of being too busy to improve and that's a red flag which becomes very visible to customers, so I Believe more transparency and action is needed or making it more easy for existing customers to contribute more apps and influence the strategy through a serious app voting system.
I stopped suggesting apps to the wishlist because the existing apps in wishlist are waiting for too long without any follow up like nobody from the staff cares anymore. And I want to think I'm wrong.

Your post seem to imply we ignore the hard work, but what I think is rather the customers are concerned with staff priorities and worry of the status quo or regression in quality.
There are currently thousands of topics in the wishlist section and no way to see clearly if any will be on the roadmap in the future. If you then compare with how many apps are in the app store it means it would take decades before current wishlist can even partially be addressed.
The staff also mentioned they have more work so cannot deliver interesting apps or apps requiring more complex setup, which raise the natural concern of the customers who see bugs being introduced and dont see more popular apps be included.
So yeah Cloudron has an incredible value but wouldn't be nice to be able for customers to have more influence at least the apps to be picked? Or at least to cleanup the wishlist from the apps that are waiting for too long in the forum to at least not create false hopes.
And as you say packaging Cloudron apps is a challenge but then what is done to simplify this or to call for contributors?
What is done to motivate more customers? Even the referral program was cancelled.
There is of course lot of work around quality and maintenance of apps. This is I believe understood by the customers but yet doesn't invalide the complaints.

For me, I got into Cloudron thanks to

, both in French. Anyway I contributed the reddit thread. It is normal to not please everyone in forums, but yet it is important to take those opportunities to share some good words to the public about Cloudron.

You can backup to a private cloud with Obsidian Git plugin or Syncthing (not a plugin, but available on any machine and even on Cloudron) already , I do it for years and because I also sync many other things with Syncthing, I feel I do not need a specific app for Obsidian sync. Just Syncthing works well for tons of stuff like even replace Dropbox to sync GBs worth of Video/Pics/Music/Docs..., etc.
Once you give Syncthing (which is free) a try, you won't need much of any other sync app

I'm also highly interested in such features, to the light of all backup failures faced by few folks including me recently and reported in the support and discussion forums. Could be indeed something as simple as using rclone.

I confirm the notification persists also for me (Ubuntu 22.04) despite 2 successive reboots.

I'm just migrating from Contabo Storage to Hetzner because of issues with Contabo... and I hope I won't face your issues folks ! I'll keep you posted

I'm frustrated with the lack of such a basic funny utility
So if you are in the mood for workarounds without the need for fully packaging a Cloudron app, here my hacky solution cooked in 1h:

1. Create a LAMP app in Cloudron, something like yt-dlp for a name sounds good xD
2. Mount a volume to your created app. It's recommended but optional, it depends if you want Cloudron to backup your downloaded videos (by default it backups /app/data) ?
3. Open a terminal from Cloudron WEB UI into your freshly created yt-dlp app, and then download/execute this script (https://gitea.zoemp.be/sansguidon/snippets/raw/branch/main/cloudron/awesome-installs/yt-dlp.sh) from my Gitea instance

curl -O https://gitea.zoemp.be/sansguidon/snippets/raw/branch/main/cloudron/awesome-installs/yt-dlp.sh
bash yt-dlp.sh

-> The script installs yt-dlp, prompts for htaccess password and for downloads location (by default /app/data/downloads), writes a config.ini file, configures htaccess,... and creates a /app/data/public/index.php which will be a very minimalistic "web interface to yt-dlp"
4. You might need to follow instructions here https://github.com/yt-dlp/yt-dlp/wiki/FAQ#how-do-i-pass-cookies-to-yt-dlp and download your YouTube cookies from Firefox/Chrome/Brave (then remove the extension) because some YouTube downloads require authentication.
5. Point your browser at your Cloudron app and upload your cookies txt file then try filling a few youtube links in the textarea.

Screenshot of this minimalistic web app. Feel free to take ownership of the code and customize at need.

I've also lost quite some time kicking accounts and repositories,
in the end I've combined some of the options below (don't forget the [service] section marker and also restart after the changes !)

[service]
DISABLE_REGISTRATION = True
REGISTER_MANUAL_CONFIRM = True
EMAIL_DOMAIN_ALLOWLIST = XX_your_domain_here_XX,cloudron.local
DEFAULT_USER_IS_RESTRICTED = True

It's overkill if all are used at the same time, but depending on the needs and paranoia, it can be useful

FYI the migration to Hetzner Storage Box coming from Contabo Object Storage was worth it and fixed all failures related to backups whether it was in term of stability or duration.
I didn't encounter issues yet, despite I'm using encrypted backups and i've plenty of apps including Nextcloud but I don't have as much data on it.

Hey, I've an update to report on this:
I had recently lot of crashes of my Shaarli instance that I couldn't explain nor troubleshoot. I was also convinced I should migrate away from Shaarli to fix my suffering. But with patience and being stubborn, come great success! A Shaarli plugin was it seems causing lot of troubles: https://github.com/ilesinge/shaarli-related, this plugin was causing memory exhausted issues on many pages of Shaarli.
Disabling it solved all my php memory exhausted issues.

Hope it helps!

Related GH issues:

• https://github.com/shaarli/Shaarli/issues/2130#issuecomment-2917570457
• https://github.com/shaarli/Shaarli/issues/953#issuecomment-2917561291

Looks like it's solved! Thanks @vladimir.d (https://git.cloudron.io/packages/uptime-kuma-app/-/commit/42c2f7a6d689ab08a1c05317fede770e8d72d6f0)

Hey all,

https://github.com/slskd/slskd/ is a modern and secure alternative web front to work with the Soulseek network.
I wanted to get rid of Soulseek/Nicotine/Seeker on my mobile phone and just use Cloudron for connecting to Soulseek. And now it works.

This is something I'm still fine tuning so is still in early stage and not clearly documented, yet feedback is welcome so I can rework the defaults correctly.
Anyway I'm sure some people wanted to use Soulseek on their Cloudron so it is now paving the way to this
https://gitea.zoemp.be/sansguidon/soulseek-cloudron

I have yet to adapt the whole packaging/deployment to not use my own git server and also to use the Cloudron base image , this is why this is labelled WIP

For most app upgrades installed on my instance, the full changelog is shown in the notifications tab, but not for Calibre-Web.

Is it possible to fetch the full changelog and include it in the notification area as for other apps?

I tend to think the right solution for now is in the middle of prioritizing the right apps for Cloudron core team, while also having more community contributors and docs to make it easy to package new apps with sane defaults without help of Cloudron staff, relying on scripts or templates to rely on to avoid too long feedback loops between Cloudron staff and users.

I'm myself interested to help packaging a few apps for Cloudron as I have the skills, I love Cloudron, and I'm Developer/DevOps/SRE very focused on automation/docs and eliminating chore and repetitive tasks. I'm building a small app for my needs right now and I'll likely package it for Cloudron as a custom app, then I bet it will build me some more experience to contribute back to package other apps.

Before finishing this post, it would be very customer-friendly to maybe have a known roadmap of such apps Cloudron staff is busy packaging and the ones for which the staff sees an interest but lacks time, and where community effort is welcome. Having maybe the staff delegating the packaging for some apps to its enthusiastic users will spread the knowledge and increase the speed to ship.

I do agree that I'd love to see at least 2/3x more apps being packaged in Cloudron every year