topic Re: Send Command to a linux host using workflows in Automations

Send Command to a linux host using workflows

Jorge3229 — Wed, 06 Nov 2024 16:55:25 GMT

Good day everyone, I am currently working with EdgeConnect because I want to send commands to a Linux host to restart a process. I remember seeing an EdgeConnect application for workflows, is that correct?

If that app no longer exists, how can I send that command to my host through a workflow?

Re: Send Command to a linux host using workflows

DanielS — Thu, 07 Nov 2024 03:19:53 GMT

Hello @Jorge3229 using ansible does the trick.

https://xxxxxxxx.apps.dynatrace.com/ui/apps/dynatrace.hub/browse/all?search=ansible&details=dynatrace.redhat.ansible

with this you can create an action to launch a job template in Red Hat Ansible Automation Controller.

Re: Send Command to a linux host using workflows

christian_kreuz — Fri, 15 Nov 2024 08:44:15 GMT

You can take a look at Kubernetes Automation: https://docs.dynatrace.com/docs/platform-modules/automations/workflows/actions/kubernetes-automation

This leverages EdgeConnect on Kubernetes, and you can leverage several actions (e.g., apply configuration or restart deployment) to perform certain tasks.

In theory you can also try to use a Kubernetes job/workload to execute a command on another host.

Please also refer to @DanielS post about using Red Hat Ansible, that's a perfect alternative.

Re: Send Command to a linux host using workflows

PacoPorro — Fri, 15 Nov 2024 09:32:22 GMT

You can use a server in your host to allow restarting the process.
I do prefer the Ansible solution as it is more scalable than this approach.
But something like this should work.

import ssl from http.server import HTTPServer, BaseHTTPRequestHandler import subprocess class RequestHandler(BaseHTTPRequestHandler): def do_POST(self): if self.path == '/restart-process': # Add authentication check here if needed self.restart_process() else: self.send_error(404, "Not Found") def restart_process(self): try: # Replace this with your actual restart command subprocess.run(["systemctl", "restart", "your-process-name"], check=True) self.send_response(200) self.send_header('Content-type', 'text/plain') self.end_headers() self.wfile.write(b"Process restarted successfully") except subprocess.CalledProcessError: self.send_error(500, "Failed to restart process") def run_server(): server_address = ('your-linux-host.internal', 443) # Use the appropriate hostname and port httpd = HTTPServer(server_address, RequestHandler) # SSL wrapping context = ssl.SSLContext(ssl.PROTOCOL_TLS_SERVER) context.load_cert_chain('server.crt', 'server.key') httpd.socket = context.wrap_socket(httpd.socket, server_side=True) print(f"Server running on https://{server_address[0]}:{server_address[1]}") httpd.serve_forever() if __name__ == '__main__': run_server()

server.crt and server.key used for HTTPS
The restart_process method uses subprocess.run to execute the restart command. Replace "your-process-name" with the actual process name