Giter Club home page Giter Club logo

aws-sts-assumerole's Introduction

assumerole: A bash script to easily assume AWS roles using Temporary Security Credentials and MFA

The script uses the standard AWS credentials file ~/.aws/credentials and it's own configuration file ~/.assumerole to assume a role on an account (defined in ~/.assumerole), using the AWS profile in ~/.aws/credentials referred to by the aws_profile property in ~/.assumerole.

Credentials are cached in ~/.assumerole.d/cache. When assuming a role using assumerole, and the cache exists, the cache is used and the validity of the cache is checked.

If the cache is valid, it is used. If the cache is not valid, the cache entry is cleared, and the user is asked to enter the MFA code.

An example to illustrate this.

The AWS credentials file ~/.aws/credentials contains:

[acme-bastion]
aws_access_key_id = AKIAXXXXXXXXXXXXXXXX
aws_secret_access_key = bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb
region = eu-central-1

The ~/.assumerole file contains this:

{
  "assume_roles": {
    "acme-sandbox-read": {
      "aws_profile": "acme-bastion",
      "aws_account": "123456789012",
      "aws_mfa_arn": "arn:aws:iam::210987654321:mfa/rtytgat",
      "aws_role": "read",
      "environment": [
        {
          "name": "MYVAR1",
          "value": "MYVALUE1"
        },
        {
          "name": "MYVAR2",
          "value": "MYVALUE2"
        }
      ]
    },
    ...
}

And running the command:

$ /usr/local/bin/assumerole
Select from these available accounts:
... 
acme-sandbox-read
acme-otheraccount-read
...

Account:   acme-sandbox-read
MFA token: 123456
export AWS_ACCESS_KEY_ID=XXXXXXXXXXXXXXXXXXXX
export AWS_SECRET_ACCESS_KEY=aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
export AWS_SESSION_TOKEN=aaaaaaa/a/very/long/string/aaaaaaaaaa

Does the following:

  • use the AWS profile acme-bastion
  • if the user has permissions to assume the role read on account 123456789012 ...
  • ... temporary credentials are requested for that account
  • extract the relevant properties from the returned JSON file
  • sets the environment variables
  • and also prints the export commands to stdout for the user to copy/paste
  • sets the environment variables if any are defined in the configuration for that profile
  • loads the SSH key, if one is defined for the selected profile
  • start a new shell

In the new shell, the permissions linked to the role that is being assumed is available for the aws CLI.

Exiting the shell will unset the credential environment variables.

Pre-requisites

  • The script uses jq to parse the JSON configuration file ~/.assumerole
  • The AWS CLI, see here on how to install this

Usage

$ assumerole [profile [mfatoken]]
$ assumerole 
Select from these available accounts:
cust1-prod-read cust1-staging-power
Account:   cust1-prod-read
MFA token: 331956
export AWS_ACCESS_KEY_ID=ASIAXXXXXXXXXXXXXXXX
export AWS_SECRET_ACCESS_KEY=7O801XXXXXXXXXXXXXXXXXahoLwdz8KLtRCc1Bvh
export AWS_SESSION_TOKEN=FQoDYXdXXXXXX...XXXX//

During its execution, assumerole starts a new shell (${SHELL}). The environment variables that build the credentials are set before starting the shell.

The user has to enter 2 values (either on the commandline or interactively):

  • The account: this is a account string defined in the configuration file ~/.assumerole
  • MFA token: the current value of the MFA token used as multi factor device

Commandline completion

Bash commandline completion was introduced in version v0.1.1 and can be used to complete account names.

bash commandline completion

For command completion to work, add these lines to your ~/.bash_profile file:

# Bash completion for assumerole
_assumerole_accounts=$(assumerole accountlist)
complete -W "${_assumerole_accounts}" 'assumerole'

You need to star a new shell (or source the ~/.bash_profile file) before command completion will be avialable.

zsh commandline completion

Create the file ~/.oh-my-zsh/completions/_assumerole with this content:

#compdef assumerole

function _assumerole {
  _describe -t 'assumerole' $(/usr/local/bin/assumerole accountlist)
}

Add these lines to your ~/.zshrc file:

### Commandline completion
autoload -U compinit
compinit

source /usr/local/bin/aws_zsh_completer.sh

Environment Variables

AWS_STS_DURATION_SECONDS

Set this envvar to the number of seconds you wish the temporary credentials to be valid.

Mind though, the assumed role should be configured to allow at least the requested session duration. If the role's session duration is less than what you specify here, the assumerole operation will fail.

To increase the role's maximume CLI/API session duration, use the AWS Console or this CLI command:

aws iam update-role -–role-name name-of-the-role -–max-session-duration 14400

${ASSUMEROLE_COMMAND}

When this environment variable is set and the assumerole command is executed, it will not start a new shell, but run the command in the environmnet variable instead, returning to the current shell after the command finishes.

This can be used to loop over different accounts and have the same command run against those accounts.

Custom profile environment variables

When required, custom environment variables can be defined in the configuration file. When that profile is activated, the environment variables are set before the new shell is spawned.

To define custom environment variables, add a list environment to the profile, for example:

{
  "assume_roles": {
    "acme-sandbox-read": {
       ...
       "environment": [
        {
          "name": "MYVAR1",
          "value": "MYVALUE1"
        },
        {
          "name": "MYVAR2",
          "value": "MYVALUE2"
        }
      ]
    },
    ...
}

The configuration file ~/.assumerole

This is an example configuration file:

{
  "assume_roles": {
    "logical_name_for_stsAssumeRole": {
      "aws_profile": "aws_profile_to_use",
      "aws_account": "account_number_of_account_where_to_assume_the_role",
      "aws_mfa_arn": "arn_of_user_mfa_devoce",
      "aws_role": "name_of_role_to_assume",
      "sshkey": "/full/path/to/private/key/for/this/profile"
    },
    "mycompany-prod-read": {
      "aws_profile": "mycompany-bastion",
      "aws_account": "123456789012",
      "aws_mfa_arn": "arn:aws:iam::210987654321:mfa/myuser",
      "aws_role": "read",
      "environment": [
        {
          "name": "MYVAR1",
          "value": "MYVALUE1"
        },
        {
          "name": "MYVAR2",
          "value": "MYVALUE2"
        }
      ]
    },
    ...
  }
}

logical_name_for_stsAssumeRole

This is a description for the role that will be assumed. It typically contains the name of the account where the role is going to be assumed, and the name of the role to be assumed. This is a string that can be chosen freely.

aws_profile

The assumerole script will set the environment variable AWS_PROFILE to this value. That means that the AWS CLI configuration file ~/.aws/credentials should contain a named profile that matches this string.

aws-account

The numeric account ID of the AWS account where a role is to be assumed.

aws_role

The name of the role to assume on the remote account.

aws_mfa_arn

The ARN of the MFA device of the user on the account that will assume the role. This can be a bastion account, where only user are defined, and groups that allow sts::AssumeRole permissions for a selection of accounts.

environment

A list of dicts with name and value keys, used to set custom environment in the spawned shell.

sshkey

Private key to load when activating the profile.

Put the active credentials in your shell prompt

bash

### Prompt
PROMPT_COMMAND="[[ -n \${AWS_ACCOUNT} ]] && export AWS_PROMPT=\" AWS:\${AWS_ACCOUNT}\" || export AWS_PROMPT=''"
PS1="\s-\v\$AWS_PROMPT $ "

zsh

POWERLEVEL9K

### POWERLEVEL9K config
POWERLEVEL9K_LEFT_PROMPT_ELEMENTS=( dir vcs  custom_aws_credentials)
POWERLEVEL9K_CUSTOM_AWS_CREDENTIALS="[[ -n \${AWS_ACCOUNT} ]] && echo AWS \${AWS_ACCOUNT}"

aws-sts-assumerole's People

Contributors

rik2803 avatar ddewaele avatar chazchandler avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.