https://github.com/sivel/ansible-memory-profiler

Docker image to aid in memory profiling an ansible playbook using cgroups
https://github.com/sivel/ansible-memory-profiler

Last synced: 19 days ago
JSON representation

Docker image to aid in memory profiling an ansible playbook using cgroups

• Host: GitHub
• URL: https://github.com/sivel/ansible-memory-profiler
• Owner: sivel
• License: gpl-3.0
• Created: 2018-04-30T16:45:04.000Z (over 6 years ago)
• Default Branch: master
• Last Pushed: 2018-04-30T21:21:26.000Z (over 6 years ago)
• Last Synced: 2024-10-03T23:41:11.413Z (about 1 month ago)
• Language: Python
• Size: 18.6 KB
• Stars: 1
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# ansible-memory-profiler

This Docker image was created to aid in memory profiling an Ansible playbook using

the `cgroup_memory_recap` callback plugin on a system without access to cgroups

## Using

```

docker run --rm -ti --privileged \

    -v /path/to/playbook_dir:/playbook \

    -v /path/to/ansible/source:/ansible \

    sivel/ansible-memory-profiler \

    ansible-playbook -v -i /playbook/hosts /playbook/playbook.yml

```

## Notes

### ansible directory

This image assumes that `/ansible` will be mounted, and should be the path to the Ansible source. It will be used following the [Running from Source](http://docs.ansible.com/ansible/latest/installation_guide/intro_installation.html#running-from-source) instructions. If `/ansible` is not mounted, you will presented with an error.

### playbook directory

The `/playbook` directory can be whatever you want, I've just used it as an easy to follow convention.

### cgroup_memory_recap callback plugin

This is the same callback plugin that will be included in Ansible 2.6.0, but bundled here for use in earlier Ansible versions.

## Why?

I develop on a Mac, and as such I don't have direct access to cgroups, but I do have access to Docker. This image gives easy access to use cgroup based memory profiling locally, with little hoops to jump through.