Issue
A Linux device shows as offline, not checking in, or missing from recent scans in the Automox console, even though the device itself is powered on and reachable. This is usually caused by the Automox agent service (amagent) not running on the device.
Environment
- Automox agent on Linux (Red Hat, CentOS, Debian, Ubuntu, Fedora, and other systemd-based distributions)
- Agent runs as the
amagentsystemd service, with a legacy/etc/init.d/amagentscript kept for backward compatibility on older init systems
Resolution
-
Check whether the service is running:
Automox documents theservicecommand as the standard way to check agent status on Linux, since it works whether the distribution uses systemd or a legacy SysV init script:sudo service amagent status
On distributions that use systemd directly, the equivalent command also works:
sudo systemctl status amagent
Both commands report the same underlying
amagent.serviceunit. See Installing the Automox Agent on Linux for the full reference list of Automox's Linux agent commands. -
Confirm with the process list if the status output is unclear:
ps aux | grep amagent
No matching process (other than the
grepcommand itself) confirms the agent is not running. -
Start the service if it is not running:
sudo service amagent start
or, on systemd distributions:
sudo systemctl start amagent
If the service was never enabled to start automatically at boot, enable it as well:
sudo systemctl enable amagent
- Re-check status to confirm the start succeeded, then re-run the status command from step 1. A successful start typically shows the device back online in the Automox console within a few minutes; running a manual scan from the console confirms check-in immediately.
-
If the service fails to start or immediately stops, check the agent logs before re-installing:
- Primary agent log:
/var/log/amagent/amagent.log - Command execution log:
/var/log/amagent/command.log
See Location of Files Required By Automox for the full list of Linux agent file paths. A locked database file (for example, a stale
osquery.db/LOCK) is a known cause of the service failing to stay running after install — see Agent Install Failure on SUSE Linux for the process-cleanup and database-reset steps that resolve it. - Primary agent log:
-
Restart the service instead of stopping/starting separately when recovering from a hung state (for example, after an agent update):
sudo service amagent restart
Notes
- Automox's Watchdog Service does not apply here: it only monitors the Windows Tray UI process and does not monitor or restart the
amagentservice on any platform, including Linux — see Understanding the Automox Watchdog Service. On Linux,systemdrestarting a crashed unit depends on the unit's ownRestart=policy, not a separate watchdog process. - If the agent won't uninstall with an error that
amagent.serviceisn't loaded, that's a separate, known packaging issue — see Agent Will Not Uninstall Linux Due to Service Not Running.