Cloudron Forum

rcdncn

It's appearing in the Nextcloud log. But isn't this the error received by collabora to Nextcloud when the request is placed?

rcdncn

I have a Nextcloud instance running on a non-Cloudron server, but use Collabora within Cloudron to open documents and convert. Currently, I will get the following error in my Nextcloud logs:

[richdocuments] Error: Failed to convert preview: Client error: `POST https://[servername]/cool/convert-to/png` resulted in a `413 Request Entity Too Large` response:
<html>
<head><title>413 Request Entity Too Large</title></head>
<body>
<center><h1>413 Request Entity Too Large</h1>< (truncated...)

	GET /core/preview?fileId=11028&x=32&y=32&mimeFallback=true&v=687e6d&a=0
	from 70.54.87.2 by kv_pdacadmin at Sep 12, 2025, 12:46:02 PM

Some research suggests the it might be the nginx at the cloudron side that may be the issue, and that client_max_body_size may need to be increased. What's odd is that I can open large files for editing, so I think that Nextcloud is sending other requests to Collabora (thumbnails, pdf conversion?) that are failing.
I can't see any error logs in nginx on Cloudron, but I can see the access.

Am I on the right path? Is there a documented way to alter the nginx infor for a single app that is persistent between updates?

rcdncn

In the log for Nextcloud, I can see that it was simply restoring:


at the top of the app page, I see:  Error : Task Error - Task 15416 crashed as it ran out of memory

This is running on an AWS Lightsail machine with 2 vCPUs and 4GB RAM. This hasn't been an issue before.

rcdncn

According to the app event log, I was able to restore to a working copy with 5.1.2, and an overnight update initiated automaticaly the next morning setting it to 5.2 at which point the app was listed as 'not responding'. I've disabled auto-update and started another restore to a 5.1.2 image to see what happens.

rcdncn

My Nextcloud was not running today, and I could see that there was an upgrade (5.3.0) yesterday. I tried to restore to that previous state (5.2.0), and it fails (Task Error: Task 15381 crashed as it ran out of memory).

I worry that the issue is likely due to my Nextcloud storage being filled up (out of space). When trying to restore, there is not enough space. I can't seem to get it started again.

What are my next steps? Restoring the backup also seems to be restoring the files that are filling up my drive.
I do have my data on a separate volume. I would be willing to migrate my data to a larger drive, but I need to get the instance up and stable first.

rcdncn

@rcdncn

An update. I goofed and it's working now. My first try I navigated to app/data/.htaccess and there was no

<IfModule mod_php7.c>

I noticed that there is both and htaccess and a .htaccess and I was in the incorrect file.

rcdncn

@nebulon I just tried this myself and when I refreshed the browser the the php value did not update. I don't have to restart any services for this to take effect?

rcdncn

Thank you for the prompt response. I've filed a ticket.

rcdncn

So... I messed up.

I accidentally deleted the custom path in Resources > Storage, and Nextcloud happily tried to flood the Nextcloud partition with my data.

When I ran out of space, the action failed and the Cloudron reset. I was able to get into it via terminal and delete the migrated data, but the Nextcloud Repair tab is displaying the following error:

An error occurred during the data migration operation: Addons Error: Error setting up postgresql. Status code: 500 message: could not write init file

My only option is to 'Retry Data Migration', and I can't 'Restart' or 'Enable Recovery Mode' the instance. Without restarting the instance, how can I stop this action??

rcdncn

Thanks @girish this totally worked.

Steps:

Option 1: Create a new skeleton folder

add directory /app/data/skeleton using web terminal
chown www-data:www-data skeleton
add files to the directory, also chowning themto www-data:www-data
open config file with vi /app/config/config.php
insert line 'skeletondirectory' => '/app/data/skeleton'on a new line (i to enter insert mode)
press esc, then :wq and <enter> to write and save
new users will receive the contents of /app/data/skeleton on first log-in

Option 2: NO files added on first log-in

If you just want to remove the default skeleton (no files added on first launch), do the following:

open config file with vi /app/config/config.php
insert line 'skeletondirectory' => ''on a new line (i to enter insert mode)
press esc, then :wq and <enter> to write and save

rcdncn

It's possible to change the skeleton location using this variable in config/config.php

'skeletondirectory' => '/path/to/nextcloud/core/skeleton',

Is there a writable space for this to point to?

Cloudron makes it easy to run web apps like WordPress, Nextcloud, GitLab on your server. Find out more or install now.

Cloudron Forum

rcdncn

Posts

Option 1: Create a new skeleton folder

Option 2: NO files added on first log-in