Just small notes about some parts of nomad and workflow with it.
Nomad has great documentation, most of the time it is everything you need when interacting with Nomad
Intro https://www.nomadproject.io/intro/index.html
Documentation https://www.nomadproject.io/docs/index.html
Nomad is a tool built by HashiCorp for efficient cluster management, jobs sheduling and other types of infrastructure operations.
Installing process is pretty obvious, as it built in Golang the easiest way will be just to to grub single executable binary from https://github.com/hashicorp/nomad/releases
You should add Nomad ex. binary to your $PATH in your preffered shell.
As described in docs primary configuration structure, most of the time when using Nomad you interacts with them. Job are written in DSL language built by Hashi team called HCL.
Docs about HCL can be found here
Job syntax validation can be done by
$ nomad validate <path_to_job>Each job consists of Tasks and TaskGroups
Task is a single action that should be performed. It can be starting nginx, or building your API docker image etc.
TaskGroup is a set of tasks that should work together.
Sample task group can be called web where you start your api and mongo for example.
Nomad provided REST API for fetching, creating jobs, their statuses etc.
There are some packages with wrapper around API. Main official package in Golang as it a core of nomad and third party in Java, Ruby etc.
Can be used by cluster monitoring solutions.