Skip to content

Connect to shell of containerized application like you were used to before Docker

License

Notifications You must be signed in to change notification settings

DocX/aws-ecs-utils

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 

Repository files navigation

AWS Container Service (ECS) CLI utils

Connect to shell of containerized application like you were used to before Docker.

Installation

  1. (Prereq) Install awscli (https://aws.amazon.com/cli/) and jq (https://stedolan.github.io/jq/)
  2. Make sure you have aws configured with stored authentication
  3. Download bin folder from this repository and copy content to your /usr/local/bin
  4. Add path to your ECS PEM file to ECS_PEM_FILE environment variable

Commands

Conventions

All commands expect your aws command is configured with default AWS region. Also majority of commands expect ECS_PEM_FILE ENV variable to be pointing to PEM file for SSH access to ECS cluster hosts.

ecs-run-shell

Start new ECS task with overridden CMD to sleep 30m and connect to its shell. Stops tasks after shell is exited.

ecs-run-shell cluster-name task-definition-name[:version]

This is useful to run your app framework console (like rails console) within environment and with ENV variables of your ECS service or task while not risking failure of already running containers.

If task definition defines more than 1 container, all of containers are started but it connects you to first one.

ecs-connect

Connect to running container of specific service.

ecs-connect cluster-name service-name

This is useful to investigate inner state of processes inside container or run processes with available environment variables from task definition.

If your service is running more than 1 task, first task provided by aws interface is used to connect to.

If your task has more than 1 containers, first container is taken.

Warning: Be aware that you are inside container that runs the main process already. Starting heavy processes may cause the whole container will run out of memory and be killed.

About

Connect to shell of containerized application like you were used to before Docker

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages