[GH-ISSUE #22] Feature Request: Agent Health Check CLI Argument #13

New issue

Open

opened 2026-03-02 02:12:43 +03:00 by kerem · 2 comments

kerem commented

2026-03-02 02:12:43 +03:00

Owner

Originally created by @UranusBytes on GitHub (Oct 28, 2022).
Original GitHub issue: https://github.com/amidaware/rmmagent/issues/22

I have the scenario where the installation of the agent is delegated to others (I document what to do; others execute the install on their servers). When something "doesn't work right", I've written a debug script the user can run to gather info (what is contents of config file, can it ping the mgmt console, etc... ) But it would be extremely helpful if the agent supported a CLI flag to do a health check. None of the "mode" options seemed right, as they didnt print results.

I could see a flag like "-testhealth" that would:

Verify the agent can connect to the console API
Verify through the console that it is registered, assigned to a client/site, and that it is "healthy"
Output a Good/Bad result to CLI (extra bonus if it is script parsable and/or provides an exit code on success/failure)

Originally created by @UranusBytes on GitHub (Oct 28, 2022). Original GitHub issue: https://github.com/amidaware/rmmagent/issues/22 I have the scenario where the installation of the agent is delegated to others (I document what to do; others execute the install on their servers). When something "doesn't work right", I've written a debug script the user can run to gather info (what is contents of config file, can it ping the mgmt console, etc... ) But it would be extremely helpful if the agent supported a CLI flag to do a health check. None of the "mode" options seemed right, as they didnt print results. I could see a flag like "-testhealth" that would: - Verify the agent can connect to the console API - Verify through the console that it is registered, assigned to a client/site, and that it is "healthy" - Output a Good/Bad result to CLI (extra bonus if it is script parsable and/or provides an exit code on success/failure)

kerem commented

2026-03-02 02:12:43 +03:00

Author

Owner

@silversword411 commented on GitHub (Oct 29, 2022):

The installer will usually tell you what problem it encounters.

What problem scenario have you had in the past where it didn't tell you?

@silversword411 commented on GitHub (Oct 29, 2022): The installer will usually tell you what problem it encounters. What problem scenario have you had in the past where it didn't tell you?

kerem commented

2026-03-02 02:12:43 +03:00

Author

Owner

@UranusBytes commented on GitHub (Oct 30, 2022):

Out of ~300 agents deployed, we have ~10 that are supposedly working (individual system owners are installing; we don't have access until RMM agent installed), but are not showing up in the console. We have only troubleshoot one so far.

For this scenario, somehow the RMM agent was downloaded, the systemd service was registered and started, but no config file at /etc/tacticalrmm was generated (no file exists). The agent was producing no error logs (either in journald or in /var/log/tacticalrmm.log) of any form (only the generic "Agent has started..."). Our normal troubleshooting steps of "Have you verified the agent service is running" and "Can you ping/curl the console" didn't point us to the problem.

I suggest the "health check" because there are other scenarios I can see happening in the future, such as an agent installing but a firewall/network preventing reaching the API, or I can see an agent fully registering but the system being shutdown for an extended period (weeks+) and when it comes back online it no longer being registered in the console. A simplified "health check" that verifies all of the aspects of an agent being fully operational, but from the agent's perspective, would prove very useful.

And to explicitly say this, thank you for such a great tool!

@UranusBytes commented on GitHub (Oct 30, 2022): Out of ~300 agents deployed, we have ~10 that are supposedly working (individual system owners are installing; we don't have access until RMM agent installed), but are not showing up in the console. We have only troubleshoot one so far. For this scenario, somehow the RMM agent was downloaded, the systemd service was registered and started, but no config file at /etc/tacticalrmm was generated (no file exists). The agent was producing no error logs (either in journald or in /var/log/tacticalrmm.log) of any form (only the generic "Agent has started..."). Our normal troubleshooting steps of "Have you verified the agent service is running" and "Can you ping/curl the console" didn't point us to the problem. I suggest the "health check" because there are other scenarios I can see happening in the future, such as an agent installing but a firewall/network preventing reaching the API, or I can see an agent fully registering but the system being shutdown for an extended period (weeks+) and when it comes back online it no longer being registered in the console. A simplified "health check" that verifies all of the aspects of an agent being fully operational, but from the agent's perspective, would prove very useful. And to explicitly say this, thank you for such a great tool!