[critical] install.lock removed after cloudron reboot

[Crush33]

hello joseph
cloudron ask me to reboot whole OS for ubuntu security update
when I do, file /app/data/dolibarr/install.lock is missing after that reboot (which is critical for security reason).

I recreate them by hand (touch install.lock from ssh in OS, maybe I shouldn't to that...?).

On a later update ubuntu, they vanish again.
If I manually reboot (i.e command "reboot" in shell), they're gone too after the reboot.

I dunno where to look at to see exactly what's happening. Is it container side ? or OS side ?
Provided above logs are from container side and there is nothing about the lock file...

[james]

Hello @Crush33
The Cloudron Dolibarr app does move the /app/data/dolibarr/install.lock file on each startup of the app temporary inside the docker container to /tmp/install.lock to run upgrade steps.
After the upgrade steps are completed, the /tmp/install.lock is moved back to /app/data/dolibarr/install.lock.

Maybe you have seen this behavior at the exact moment the app started.
But, the file should be moved back into place after the app is fully started.

[Crush33]

hello,

got same behavior tonight, after ubuntu reboot for security updates (clicked from notification panel, red button "Reboot", all my 26x Dolibarr instances restarted without putting the install.lock file back in place.

I recreate all of them at one using this in shell on cloudron :

#!/bin/bash

BASE_DIR="/home/yellowtent/appsdata"

for appdir in "$BASE_DIR"/*; do
    if [ -d "$appdir/data/dolibarr" ]; then
        LOCKFILE="$appdir/data/dolibarr/install.lock"
        if [ ! -f "$LOCKFILE" ]; then
            echo "Create $LOCKFILE"
            touch "$LOCKFILE"
        else
            echo "$LOCKFILE already exist"
        fi
    else
        echo "no dolibarr in $appdir"
    fi
done

dunno what I should do to see why it doesn't work appropriately

thanks for any help !

[robi]

Maybe the upgrade is failing, check logs?

The code could also be adjusted to copy vs move the file so it stays available. But, it's probably moved for a reason I am not aware of.

[joseph]

@Crush33 Can you give us access to your instance (possibly a cloned instance of your existing dolibarr?) . If so, can you write to me on support@cloudron.io and I can try to take a look. There is possible some bug in the package causing this.

[sebastienserre]

Hello,
Is a solution has been found because I'm experiencing the same issue.
Video => https://www.loom.com/share/a526c2ccf8ff46108d791ae6e4d862e8
Logs => https://file.thivinfo.com/Cloudron/82a18dde-001d-4c51-a69b-5ac33376e6b5 (1).log

[sebastienserre]

Fixed by setting my own cron job https://forum.cloudron.io/topic/14965/which-user-for-cron-task/4?_=1773687776018

[sebastienserre]

Hello

I thought this issue has been fixed, but we always have the issue. (which is a security one)
I do not know why the Cloudron Dolibarr app runs the cron as root and not as www-data user.

[james]

Hello @sebastienserre
I was not able to reproduce this with a fresh instance of Dolibarr.
Maybe this is due to your other issue you have posted with the custom cron job?

[tobiasb]

This can happen in this step:

=>  PREV_DOLIBARR_VERSION=`php /app/code/htdocs/install/get-version.php`

Log:

2026-02-07T13:35:00Z ==> Starting Dolibarr
2026-02-07T13:35:03Z #0 /app/code/htdocs/core/class/conf.class.php(561): DoliDBMysqli->query()
2026-02-07T13:35:03Z #1 /app/code/htdocs/install/get-version.php(28): Conf->setValues()
2026-02-07T13:35:03Z #2 {main}
2026-02-07T13:35:03Z PHP Fatal error:  Uncaught Error: Call to a member function query() on false in /app/code/htdocs/core/db/mysqli.class.php:355

A shell script with set -e stops when If an error occurs. The script does not have an error handling in such cases, therefore the lock file is still in /tmp.

With error handling

trap 'echo "Houston we have problem"; [[ -f /tmp/install.lock ]] && mv /tmp/install.lock /app/data/dolibarr/install.lock; exit 1' ERR