This is a User Script for unRaid to test containers have an internet connection via a container:vpn network. If the vpn is down (i.e. no internet) then run commands to restart it.

unRaidVPNTest

#!/usr/bin/php
<?php
#description=Container VPN Tester


$container_label = "label_here=value";
$restart_commands = [
    "php /boot/config/plugins/user.scripts/scripts/restartContainerVPN/script"
];


echo "\n\nChecking VPN connections...\n";
exec("docker ps --filter 'label={$container_label}' --format='{{.Names}}'", $containers);

foreach($containers as $container) {
    $status_output = [];
    exec("docker inspect --format='{{json .State.Health.Status}}' $container", $status_output);
    $status = str_replace(['"',"'"], "", $status_output[0]);

    if ($status == "unhealthy") {
        echo " - {$container} is UNHEALTHY!\n\n";
        echo "Restarting stacks...\n\n";
    
        foreach($restart_commands as $restart_command) {
            $restart_output = [];
            exec($restart_command, $restart_output);
            foreach ($restart_output as &$line)
                echo "{$line}\n";
        }
        break;
        
    } else {
        echo " - {$container} is {$status}\n";
    }
}
?>

Usage

This is an unRaid User Script. Install it from CA Apps.

Script Setup

- $container_label

The label you assign to the containers to check. Note that this can be set to any label you'd like.
If this contains '=' then the label must equal the value supplied. If not, then any container with the label regardless of value will be checked.

- $restart_commands

The command(s) to run. Separate each command by a comma.
The example provided will run a User Script called restartContainerVPN.

- Cron Schedule

Set the script to run every 15 minutes (or similar). On the main User Scripts page, set the schedule to Custom and enter the following cron: */15 * * * *

Docker Setup

Add your label to each container that you'd like to be checked.
Ensure each container has a health check that tests internet connectivity. After adding this, you should see the health status (in brackets) next to your uptime.

- Healthcheck via unRaid GUI:

Add this to extra parameters for each container. (Note - I haven't tested this)
--health-cmd='curl --fail http://google.com/ || exit 1' --health-interval=10s --health-retries=5 --health-timeout=10s --health-start-period=5m

- Healthcheck via Docker Compose:

Add this to each container:

    healthcheck:
      test: curl --fail http://google.com || exit 1
      start_period: 5m
      interval: 30s
      timeout: 10s
      retries: 5