# Agentless architecture in Ansible **URL:** \ **Category:** Ansible Project\ **Created:** [June 10, 2016, 11:02am UTC](https://forum.ansible.com/t/agentless-architecture-in-ansible/21730 "2016-06-10T11:02:31Z")\ **Posts on this page:** 5\ **Page:** 1
**Author:** ![Kaushal\_Shriyan](https://avatars.discourse-cdn.com/v4/letter/k/838e76/32.png) [@Kaushal\_Shriyan](https://forum.ansible.com/u/Kaushal_Shriyan)\ **Post date:** [June 10, 2016, 11:02am UTC](https://forum.ansible.com/t/agentless-architecture-in-ansible/21730/1 "2016-06-10T11:02:31Z")
Hi, I am new to Ansible. I am not sure what is agent less in Ansible and how is it advantageous over agent based? I also read the whitepaper [https://www.ansible.com/benefits-of-agentless-architecture](https://www.ansible.com/benefits-of-agentless-architecture) and still have not completely understood. Any explanation with examples will be highly appreciable. Thanks and Regards, Kaushal ---
**Author:** ![Mark\_Janssen](https://avatars.discourse-cdn.com/v4/letter/m/e9c0ed/32.png) [@Mark\_Janssen](https://forum.ansible.com/u/Mark_Janssen)\ **Post date:** [June 10, 2016, 2:00pm UTC](https://forum.ansible.com/t/agentless-architecture-in-ansible/21730/2 "2016-06-10T14:00:09Z")
The name implies most of it… No agent needed… as opposed to some other tools like puppet which need a locally installed agent on every system. In practice, ansible just uses ssh/python/shell, and not a custom agent/client. Mark ---
**Author:** ![Adam\_Morris](https://avatars.discourse-cdn.com/v4/letter/a/ecae2f/32.png) [@Adam\_Morris](https://forum.ansible.com/u/Adam_Morris)\ **Post date:** [June 10, 2016, 5:35pm UTC](https://forum.ansible.com/t/agentless-architecture-in-ansible/21730/3 "2016-06-10T17:35:35Z")
Mark has it right... To my mind the biggest advantages of agentless over agent are... 1) bootstrapping, with puppet you have to install the agent before you can control the machine, with Ansible you just control it. 2) It uses the same mechanisms for access I would (ssh and sudo) so I don't need to worry about additional security issues (how secure is that agent?) 3) No overhead. If I am not running a task there is no extra process using resources, with an agent, even if it only uses a fraction of a percent of cpu it is still there. Adam ---
**Author:** ![Dick\_Davies](https://avatars.discourse-cdn.com/v4/letter/d/b9e5f3/32.png) [@Dick\_Davies](https://forum.ansible.com/u/Dick_Davies)\ **Post date:** [June 10, 2016, 5:49pm UTC](https://forum.ansible.com/t/agentless-architecture-in-ansible/21730/4 "2016-06-10T17:49:31Z")
Orchestration in Ansible is so simple compared to agent-based CM systems; when the play has gathered facts it has up to date information about all nodes to hand. Chef and Puppet need a fair bit of extra infrastructure and configuration to deliver that. One other one that I'm finding a big win for some clients - a constantly running agent can be a challenge in many workplaces if strict change windows etc. are in place. Puppet nodes could be prone to getting 'stuck' and recover at unexpected times (just after a reboot for example) and trigger service restarts etc. With an 'operator driven' tool like Ansible you know when you're about to make a change, you can track each scheduled piece of work to a play, to a commit, and roll it through environments just as you would code. None of those issues are insurmountable with other tools but they're an extra headache. ---
**Author:** ![Mark\_Phillips](https://avatars.discourse-cdn.com/v4/letter/m/73ab20/32.png) [@Mark\_Phillips](https://forum.ansible.com/u/Mark_Phillips)\ **Post date:** [June 12, 2016, 4:05pm UTC](https://forum.ansible.com/t/agentless-architecture-in-ansible/21730/5 "2016-06-12T16:05:32Z")
Hello, This is a bit old now (Dec '14) but still pertinent. A demo I gave to the DOXLON meetup… [https://youtu.be/A7-OW50jubY](https://youtu.be/A7-OW50jubY) –Mark