Ansible managing Ansible

Archives Ansible Project
1

I’m thinking of running Ansible on every server to monitor server interfaces and connectivity to the Top of Rack switch. I envision the server sending pings every 30 seconds to the TOR, and based on a condition, modify interface config and routing table.

This can be done by a single Ansible server, but to eliminate the network and criticality of this running, I’m thinking of using a primary Ansible server to control the Ansible server on every other compute node. Basically the primary would ensure Ansible is installed on every compute node along with the playbook to run every 30 seconds. Triggered by API call to Tower or bash script.

General thoughts on this? Am I mad?

Thanks,
Jason

2

For Tower questions, please ask support@ansible.com, this list is for the open source project only.

Using Ansible as a monitoring system does strike me as inappropriate usage.