| # Recon |
| |
| Recon is a library to be dropped into any other Erlang project, |
| to be used to assist DevOps people diagnose problems in production |
| nodes. |
| |
| The source code can be obtained from [the github repo](https://github.com/ferd/recon). |
| A toy application to experiment with can be found at [recon_demo](https://github.com/ferd/recon_demo). |
| |
| ## Installing |
| |
| Add recon to your project by adding it as a dependency. With rebar3 this is done as: |
| |
| ``` |
| {deps, [recon]}. |
| ``` |
| |
| If you are using releases, make sure to add recon to this list of applications bundled in: |
| |
| ``` |
| {relx, [ |
| {release, {myrel, "0.1.0"}, |
| [myapp, |
| recon, % <-- here |
| sasl] |
| }, |
| ... |
| ]}. |
| ``` |
| |
| ## Running tests |
| |
| Call `rebar3 check` |
| |
| ## Using it |
| |
| Each module in the API describes how it should be used in a stand-alone manner. Insights about how to debug a production Erlang system, often relying on this library, can be found in the free online book [Erlang in Anger](https://www.erlang-in-anger.com/). |
| |
| The library also contains a few basic scripts that have proven useful in the past but may not be as much today in its `script/` directory. These include ways to visualize application dependencies, and a few short scripts to extract high-level information from crashdumps. |
