aws-mfa makes it easy to manage your AWS SDK Security Credentials when Multi-Factor Authentication (MFA) is enforced on your AWS account. It automates the process of obtaining temporary credentials from the AWS Security Token Service and updating your AWS Credentials file (located at ~/.aws/credentials). Traditional methods of managing MFA-based credentials requires users to write their own bespoke scripts/wrappers to fetch temporary credentials from STS and often times manually update their AWS credentials file.
The concept behind aws-mfa is that there are 2 types of credentials:
long-term - Your typcial AWS access keys, consisting of an AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY
short-term - A temporary set of credentials that are generated by AWS STS using your long-term credentials in combination with your MFA device serial number (either a hardware device serial number or virtual device ARN) and one time token code. Your short term credentials are the credentials that are actively utilized by the AWS SDK in use.
If you haven't yet enabled multi-factor authentication for AWS API access, check out the AWS article on doing so.
Option 1
$ pip install aws-mfa
Option 2
1. Clone this repo
2. $ python setup.py install
In a typical AWS credentials file (located at ~/.aws/credentials), credentials are stored in sections, denoted by a pair of brackets: []. The [default] section stores your default credentials. You can store multiple sets of credentials using different profile names. If no profile is specified, the [default] section is always used.
By default long term credential sections are identified by the convention [<profile_name>-long-term] and short term credentials are identified by the typical convention: [<profile_name>]. The following illustrates how you would configure you credentials file using aws-mfa with your default credentials:
[default-long-term]
aws_access_key_id = YOUR_LONGTERM_KEY_ID
aws_secret_access_key = YOUR_LONGTERM_ACCESS_KEY
After running aws-mfa, your credentials file would read:
[default-long-term]
aws_access_key_id = YOUR_LONGTERM_KEY_ID
aws_secret_access_key = YOUR_LONGTERM_ACCESS_KEY
[default]
aws_access_key_id = <POPULATED_BY_AWS-MFA>
aws_secret_access_key = <POPULATED_BY_AWS-MFA>
aws_security_token = <POPULATED_BY_AWS-MFA>
Similarly, if you utilize a credentials profile named development, your credentials file would look like:
[development-long-term]
aws_access_key_id = YOUR_LONGTERM_KEY_ID
aws_secret_access_key = YOUR_LONGTERM_ACCESS_KEY
After running aws-mfa, your credentials file would read:
[development-long-term]
aws_access_key_id = YOUR_LONGTERM_KEY_ID
aws_secret_access_key = YOUR_LONGTERM_ACCESS_KEY
[development]
aws_access_key_id = <POPULATED_BY_AWS-MFA>
aws_secret_access_key = <POPULATED_BY_AWS-MFA>
aws_security_token = <POPULATED_BY_AWS-MFA>
The default naming convention for the credential section can be overriden by using the --long-term-suffix and
--short-term-suffix command line arguments. For example, in a multi account scenario you can have one AWS account
that manages the IAM users for your organization and have other AWS accounts for development, staging and production
environments.
After running aws-mfa once for each environment with a different value for --short-term-suffix, your credentials
file would read:
[myorganization-long-term]
aws_access_key_id = YOUR_LONGTERM_KEY_ID
aws_secret_access_key = YOUR_LONGTERM_ACCESS_KEY
[myorganization-development]
aws_access_key_id = <POPULATED_BY_AWS-MFA>
aws_secret_access_key = <POPULATED_BY_AWS-MFA>
aws_security_token = <POPULATED_BY_AWS-MFA>
[myorganization-staging]
aws_access_key_id = <POPULATED_BY_AWS-MFA>
aws_secret_access_key = <POPULATED_BY_AWS-MFA>
aws_security_token = <POPULATED_BY_AWS-MFA>
[myorganization-production]
aws_access_key_id = <POPULATED_BY_AWS-MFA>
aws_secret_access_key = <POPULATED_BY_AWS-MFA>
aws_security_token = <POPULATED_BY_AWS-MFA>
This allows you to access multiple environments without the need to run aws-mfa each time you want to switch
environments.
If you don't like the a long term suffix, you can omit it by passing the value none for the --long-term-suffix
command line argument. After running aws-mfa once for each environment with a different value for
--short-term-suffix, your credentials file would read:
[myorganization]
aws_access_key_id = YOUR_LONGTERM_KEY_ID
aws_secret_access_key = YOUR_LONGTERM_ACCESS_KEY
[myorganization-development]
aws_access_key_id = <POPULATED_BY_AWS-MFA>
aws_secret_access_key = <POPULATED_BY_AWS-MFA>
aws_security_token = <POPULATED_BY_AWS-MFA>
[myorganization-staging]
aws_access_key_id = <POPULATED_BY_AWS-MFA>
aws_secret_access_key = <POPULATED_BY_AWS-MFA>
aws_security_token = <POPULATED_BY_AWS-MFA>
[myorganization-production]
aws_access_key_id = <POPULATED_BY_AWS-MFA>
aws_secret_access_key = <POPULATED_BY_AWS-MFA>
aws_security_token = <POPULATED_BY_AWS-MFA>
--device arn:aws:iam::123456788990:mfa/dudeman
The MFA Device ARN. This value can also be provided
via the environment variable 'MFA_DEVICE' or the
~/.aws/credentials variable 'aws_mfa_device'.
--duration DURATION The duration, in seconds, that the temporary
credentials should remain valid. Minimum value: 900
(15 minutes). Maximum: 129600 (36 hours). Defaults to
43200 (12 hours), or 3600 (one hour) when using
'--assume-role'. This value can also be provided via
the environment variable 'MFA_STS_DURATION'.
--profile PROFILE If using profiles, specify the name here. The default
profile name is 'default'. The value can also be
provided via the environment variable 'AWS_PROFILE'.
--long-term-suffix LONG_TERM_SUFFIX
To identify the long term credential section by
[<profile_name>-LONG_TERM_SUFFIX]. Use 'none' to
identify the long term credential section by
[<profile_name>]. Omit to identify the long term
credential section by [<profile_name>-long-term].
--short-term-suffix SHORT_TERM_SUFFIX
To identify the short term credential section by
[<profile_name>-SHORT_TERM_SUFFIX]. Omit or use 'none'
to identify the short term credential section by
[<profile_name>].
--assume-role arn:aws:iam::123456788990:role/RoleName
The ARN of the AWS IAM Role you would like to assume,
if specified. This value can also be provided via the
environment variable 'MFA_ASSUME_ROLE'
--role-session-name ROLE_SESSION_NAME
Friendly session name required when using --assume-
role. By default, this is your local username.
Argument precedence: Command line arguments take precedence over environment variables.
Run aws-mfa before running any of your scripts that use any AWS SDK.
Using command line arguments:
$> aws-mfa --duration 1800 --device arn:aws:iam::123456788990:mfa/dudeman
INFO - Using profile: default
INFO - Your credentials have expired, renewing.
Enter AWS MFA code for device [arn:aws:iam::123456788990:mfa/dudeman] (renewing for 1800 seconds):123456
INFO - Success! Your credentials will expire in 1800 seconds at: 2015-12-21 23:07:09+00:00
Using environment variables:
export MFA_DEVICE=arn:aws:iam::123456788990:mfa/dudeman
$> aws-mfa --duration 1800
INFO - Using profile: default
INFO - Your credentials have expired, renewing.
Enter AWS MFA code for device [arn:aws:iam::123456788990:mfa/dudeman] (renewing for 1800 seconds):123456
INFO - Success! Your credentials will expire in 1800 seconds at: 2015-12-21 23:07:09+00:00
export MFA_DEVICE=arn:aws:iam::123456788990:mfa/dudeman
export MFA_STS_DURATION=1800
$> aws-mfa
INFO - Using profile: default
INFO - Your credentials have expired, renewing.
Enter AWS MFA code for device [arn:aws:iam::123456788990:mfa/dudeman] (renewing for 1800 seconds):123456
INFO - Success! Your credentials will expire in 1800 seconds at: 2015-12-21 23:07:09+00:00
Output of running aws-mfa while credentials are still valid:
$> aws-mfa
INFO - Using profile: default
INFO - Your credentials are still valid for 1541.791134 seconds they will expire at 2015-12-21 23:07:09
Using a profile: (profiles allow you to reference different sets of credentials, perhaps for different users or different regions)
$> aws-mfa --duration 1800 --device arn:aws:iam::123456788990:mfa/dudeman --profile development
INFO - Using profile: development
Enter AWS MFA code for device [arn:aws:iam::123456788990:mfa/dudeman] (renewing for 1800 seconds):666666
INFO - Success! Your credentials will expire in 1800 seconds at: 2015-12-21 23:09:04+00:00
Using a profile that is set via the environment variable AWS_PROFILE:
$> export AWS_PROFILE=development
$> aws-mfa --duration 1800 --device arn:aws:iam::123456788990:mfa/dudeman
INFO - Using profile: development
Enter AWS MFA code for device [arn:aws:iam::123456788990:mfa/dudeman] (renewing for 1800 seconds):666666
INFO - Success! Your credentials will expire in 1800 seconds at: 2015-12-21 23:09:04+00:00
Assuming a role:
$> aws-mfa --duration 1800 --device arn:aws:iam::123456788990:mfa/dudeman --assume-role arn:aws:iam::123456788990:role/some-role --role-session-name some-role-session
INFO - Validating credentials for profile: default with assumed role arn:aws:iam::123456788990:role/some-role
INFO - Obtaining credentials for a new role or profile.
Enter AWS MFA code for device [arn:aws:iam::123456788990:mfa/dudeman] (renewing for 1800 seconds):123456
INFO - Success! Your credentials will expire in 1800 seconds at: 2016-10-24 18:58:17+00:00
Assuming a role: Assume a role specified in your long-term configuration
[default-long-term]
aws_access_key_id = YOUR_LONGTERM_KEY_ID
aws_secret_access_key = YOUR_LONGTERM_ACCESS_KEY
assume_role = arn:aws:iam::123456788990:role/some-role
$> aws-mfa --duration 1800 --device arn:aws:iam::123456788990:mfa/dudeman --role-session-name some-role-session
Assuming a role using a profile:
$> aws-mfa --duration 1800 --device arn:aws:iam::123456788990:mfa/dudeman --profile development --assume-role arn:aws:iam::123456788990:role/some-role --role-session-name some-role-session
INFO - Validating credentials for profile: development with assumed role arn:aws:iam::123456788990:role/some-role
INFO - Obtaining credentials for a new role or profile.
Enter AWS MFA code for device [arn:aws:iam::123456788990:mfa/dudeman] (renewing for 1800 seconds):123456
INFO - Success! Your credentials will expire in 1800 seconds at: 2016-10-24 18:58:17+00:00
Assuming a role in multiple accounts and be able to work with both accounts simultaneously (i.e. production an staging):
$> aws-mfa —profile myorganization --assume-role arn:aws:iam::222222222222:role/Administrator --short-term-suffix production --long-term-suffix none --role-session-name production
INFO - Validating credentials for profile: myorganization-production with assumed role arn:aws:iam::222222222222:role/Administrator
INFO - Your credentials have expired, renewing.
Enter AWS MFA code for device [arn:aws:iam::111111111111:mfa/me] (renewing for 3600 seconds):123456
INFO - Success! Your credentials will expire in 3600 seconds at: 2017-07-10 07:16:43+00:00
$> aws-mfa —profile myorganization --assume-role arn:aws:iam::333333333333:role/Administrator --short-term-suffix staging --long-term-suffix none --role-session-name staging
INFO - Validating credentials for profile: myorganization-staging with assumed role arn:aws:iam::333333333333:role/Administrator
INFO - Your credentials have expired, renewing.
Enter AWS MFA code for device [arn:aws:iam::111111111111:mfa/me] (renewing for 3600 seconds):123456
INFO - Success! Your credentials will expire in 3600 seconds at: 2017-07-10 07:16:44+00:00
$> aws s3 list-objects —bucket my-production-bucket —profile myorganization-production
$> aws s3 list-objects —bucket my-staging-bucket —profile myorganization-staging
FAQs
Manage AWS MFA Security Credentials
We found that aws-mfa demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?
Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.
Security News
In this segment of the Risky Business podcast, Feross Aboukhadijeh and Patrick Gray discuss the challenges of tracking malware discovered in open source softare.
Research
Security News
A threat actor's playbook for exploiting the npm ecosystem was exposed on the dark web, detailing how to build a blockchain-powered botnet.
Security News
NVD’s backlog surpasses 20,000 CVEs as analysis slows and NIST announces new system updates to address ongoing delays.