@nebulon said in Passing the Healthcheck on custom app:

if you push your current code to some public repository

I have solved the problems I was encountering. Phew.
Now in a basic working state. Not pretty but functional.

I will share in case it can be useful to others or there is advice on how to do it better.

I tried to create a project on https://git.cloudron.io to upload
But I got invalid namespace
Is creating projects on git.cloudron.io restricted ?

@girish phew, thank you

Thanks @mehdi and @nebulon

My problem solved by deleting the package-lock.json. After that I can update this app and all works good.

Thank you very much.

@marcusquinn : I tried out out Kasm.
Generally a good system and an easy install (if you put on its own box not with other stuff).
However, even after increasing the cores and RAM that can be allocated to an app/image, I found it a little slow. Maybe it's the VPS I put it on, despite the VPS having 32Gb RAM and 8 cores.
My expectation level has been set at a middle level remote desktop Workspace from AWS. Although I am trying to get off AWS, I have to say that workspace performs well. Better than Kasm.
So very interested to see what performance a Cloudron Desktop App will be. When it's ready.

@roru2k20 just git clone on a computer you have cloudron-cli installed then follow instructions in readme (skip docker build/push)

@rmdes groom seems even better than boringproxy. Would be nice if it had a webUI, would make a nice App for us.

@msbt thank you
I've decided to 'step over' adjusting them in the config file (because I can't) and adjusted the app code to read the config file
e.g. process.env.CLOUDRON_MONGODB_HOST as it is nodejs app
Not there yet but making progress

Right, just use the turn addon - https://docs.cloudron.io/packaging/addons/#turn

@alphagroup if it's an actual clone, you can just install the initial app, create a backup and clone the app to a new (sub)domain for each customer. If they don't have access, you might want to check out the Cloudron CLI or API and build some Dashboard around that 😉

@klawitterb Oh I see, good point, that part escaped my notice. Mediafiles are directly served by nginx and only the rest is forwarded to Gunicorn.

#serve media files location /media/ { alias /app/data/mediafiles/; } # pass requests for dynamic content to gunicorn location / { proxy_set_header Host $http_host; proxy_pass http://localhost:8080; # -> this gos to Gunicorn }

So with the Cloudron Nginx, static files would still be served from Gunicorn. SO I guess, it is worth the effort 😉
Thanks for that.

@privsec why? The two are separate things.

Cloudron devs maintain the packaged updates, app maintainers the code.

@nebulon Great answer, thank you

@jodumont I believe the problem may come from the --daemon flag that you pass in the start.sh. Can you try removing that ?

@grienauer
When I did my testing for the Valheim Server App I wrote this script to make testing easier.

Note tho, this only works when not using any addons, but localstorage works with the local mount.

#!/bin/bash # Function to get the user to input the needed vars get_vars () { read -p "Cloudron E-Mail: " EMAIL read -p "Cloudron Username: " USERNAME read -s -p "Cloudron Password: " PASSWORD read -p "Cloudron URL: " CLOUDRON_URL read -p "DOCKER REPOSITORY URL: " DOCKER_REPOSITORY_URL read -p "DOCKER USERNAME: " DOCKER_REPOSITORY_USERNAME read -p -s "DOCKER PASSWORD: " DOCKER_REPOSITORY_PASSWORD printf "EMAIL=$EMAIL\n USERNAME=$USERNAME\n PASSWORD=$PASSWORD\n CLOUDRON_URL=$CLOUDRON_URL\n DOCKER_REPOSITORY_URL=$DOCKER_REPOSITORY_URL\n DOCKER_REPOSITORY_USERNAME=$DOCKER_REPOSITORY_USERNAME\n DOCKER_REPOSITORY_PASSWORD=$DOCKER_REPOSITORY_PASSWORD\n " > .env } # check if .env file exists and read it or check if vars exsist if [ -f ".env" ]; then export $(egrep -v '^#' .env | xargs) &> /dev/null else echo ".env File missing - asking for required vars" get_vars fi if [ -z "$EMAIL" ] || [ -z "$USERNAME" ] || [ -z "$PASSWORD" ] || [ -z "$CLOUDRON_URL" ] || [ -z "$DOCKER_REPOSITORY_URL" ] || [ -z "$DOCKER_REPOSITORY_USERNAME" ] || [ -z "$DOCKER_REPOSITORY_PASSWORD" ]; then echo "some vars are empty - asking for required vars" get_vars fi echo "=> Login Docker" docker login --username $DOCKER_REPOSITORY_USERNAME --password $DOCKER_REPOSITORY_PASSWORD $DOCKER_REPOSITORY_URL echo "=> Login Cloudron" cloudron login --username $USERNAME --password $PASSWORD my.$CLOUDRON_URL set -x # Get the ID and VERSION from the CloudronManifest.json ID=$(jq -r ".id" CloudronManifest.json) VERSION=$(jq -r ".version" CloudronManifest.json) echo "=> Create Test Data dir" mkdir -p ./cloudron_test/data ./cloudron_test/tmp ./cloudron_test/run echo "=> Cleanup Test Data" rm -rf ./cloudron_test/data/* ./cloudron_test/tmp/* ./cloudron_test/run/* echo "=> Build test image" cloudron build --set-repository $DOCKER_REPOSITORY_URL/$ID # docker build -t $DOCKER_REPOSITORY_URL/$ID:$VERSION . if [[ RUN_IMG = "" ]]; then read -p "Do you want to run the image created image?: (y/N)" RUN_IMG fi if [[ "$RUN_IMG" = "y" || "$RUN_IMG" = "Y" ]]; then echo "=> Run `test` tag of build image" docker run -ti --read-only \ --volume $(pwd)/cloudron_test/data:/app/data:rw \ --volume $(pwd)/cloudron_test/tmp:/tmp:rw \ --volume $(pwd)/cloudron_test/run:/run:rw \ $DOCKER_REPOSITORY_URL/$ID:$VERSION \ bash fi echo "=> Running tests" cd test echo "=> Running ncu - updating deps" ncu -u npm -i echo "=> Installing package.json" npm install if [[ -f ./node_modules/.bin/mocha ]]; then ./node_modules/.bin/mocha ./test.js -b fi if [[ -f ../node_modules/.bin/mocha ]]; then ../node_modules/.bin/mocha ./test.js -b fi # remove installed test instance echo "Cancel now if you wish to keep the app installed" sleep 10 cloudron uninstall --app test.$CLOUDRON_URL

@girish hah, I didn’t realize there was an update! Thanks!!! I’m glad I wasn’t crazy there and it was a platform issue. I redid my package like 4 times thinking I was overriding the environment vars somehow

@girish For reference, these are the constraints if we're trying to support Tasmota-based IoT devices using BearSSL:

Your SSL/TLS server must support TLS 1.2 and the RSA_WITH_AES_128_GCM_SHA256 cipher - which is the case with the default Mosquitto configuration
 
The server certificate must have an RSA private key (max 2048 bits) and the certificate must be signed with RSA and SHA256 hash. This is the case with default LetsEncrypt certificates.

The docs are obviously dated, because I don't know that those LE defaults still apply.

@jodumont If you run the node file for cloudron directly (add it to your PATH) you don't need the set-executionpolicy bit. Node is separate from powershell and as such it can run without that.

@nebulon Thanks thought that might be the answer.

Thanks, that's perfect.

@jimcavoli I think that makes sense. Let me see if I can get the recvmail addon working again, should be straightforward. For your suggestion of having a "selector" in the email UI, we did exactly that for sendmail addon in the previous release. So, we can do the same for recvmail as well I guess. Let me investigate.

@girish Now It's working, thanks for your help