Cloudron Forum

Hello everyone,

I was asked to develop an account managing solution via LDAP for an institution with around 2000 users. In this scenario only a LDAP Directory Server is needed. External apps that consume the LDAP Directory are already there.
I immediately thought of Cloudron and its Directory Server for external Apps, because we also use it very successfully in our company. But in our company we use it for about 10 people and in this case it should manage 2000 users.
Does anyone have experience using Cloudron Directory Server with several thousand users as an LDAP provider for external apps and can give me an assessment of whether this is a good idea?
Or can someone from the Cloudron team tell something about whether the Cloudron Directory Server is designed to handle so many users or where problems might lie?

Thank you very much for your answers.
Kind regards
Philipp

Hello everyone,

I want to use Directus installed via Cloudron as data and content management system for a web project. I was wondering how many concurrent requests this setup could handle with a server that has multiple virtual CPU cores.
This led me to the question how scaling in Cloudron and Directus actually works.

The Directus backend runs with node and is single threaded. In the following GitHub discussion the tech lead of Directus, Rijk van Zanten, says that directus has no vertical node process scaling and that they use horizontal scaling of multiple container instances. https://github.com/directus/directus/discussions/9781#discussioncomment-1645326.

When I get the idea of Cloudron right, then there is one single container running for every app and there is no option to run multiple container instances of one app.

Actual Question

So am I getting it right that Directus on Cloudron always uses only one virtual CPU Core?
So when I have a server with 4 vCPUs and run Cloudron with only Directus installed on it, 3 vCPUs of them are needless?

Thanks for any help
Philipp

@girish Thank you very much, that solved the problem.

The 'change upload part size' hint would maybe be great for a Backup section in Troubleshooting Knowledge Base.
I also received an e-mail, where a backup section in troubleshooting was mentioned. But that section does not exist:

@girish thank you for your answer.

I did a little bit of load testing and that confirmed me again, that only one vCPU core is used.

But this is a problem for almost every node.js app of the cloudron App Store, isn't it? Do they all use only one vCPU core?

Then I came across this project pm2 https://pm2.keymetrics.io/, a process manager for node.js with integrated load balancer. I've never took a deeper look on how you package your Apps on Cloudron, but would it theoretically be possible to include pm2 in the Directus app setup for Cloudron or do you have any conditions that forbid "external" dependencies like pm2. Or are there any other obstacles that would speak against that?
I was thinking about playing a little bit around with it.

Thanks for your help again and best regards
Philipp

Hello everyone,

I was trying to move my Cloudron instance from an IONOS VPS Server to a bigger machine on netcup.
Sadly the app Baserow didn't work after the restore on the new server.

My prerequisites

• Ubuntu 20.04
• Cloudron Version 7.2.5
• Manual DNS Provider
• ionos-objectstorage as Storage Provider for backups

My steps

I tried to follow every step from the knowledge base article https://docs.cloudron.io/backups/#move-cloudron-to-another-server.

• Took complete backup & downloaded backup configuration file
• Installed Cloudron in correct version (7.2.5) on the new server and set ions-objectstorage as provider
• Set all DNS A Records to point to the new servers IP address
• Waited only a few minutes to let the DNS changes propagate
• Started restoring Cloudron with backup config file
• After a few minutes all apps worked fine except Baserow
• (in network settings new servers IP address was detected)

After that I sadly didn't inspect the Baserow app a little closer, but directly started a restore of this app.

• Since then the task of restoring the app seems to work fine until 75%
• The step "Restoring addons" fails after 30 minutes to hours (don't know exactly)

My suggestion was that the DNS change propagation wasn't finished as I started the restoring process and restore process failed because of this. (knowledge base: Note that if you do not switch the IP for the app domains, the restore of those apps will fail and you have to trigger the restore of the app from the Backups section when you are ready to switch the IP)

• So I changed all DNS A records back to old servers IP address
• After some time old servers IP address was detected
• But the Basrow app was in the same state: the restore process failed

I don't know what I can do next.
I'm thankful for any help.

@nebulon you were right.
With the new version 7.3 the restore worked immediately and now baserow is running again.
Thank you for the great support!