Identity-based policy examples for Amazon S3
This section shows several example AWS Identity and Access Management (IAM) identity-based policies for controlling access to Amazon S3. For example bucket policies (resource-based policies), see Bucket policies for Amazon S3. For information about IAM policy language, see Policies and permissions in Amazon S3.
The following example policies will work if you use them programmatically. However, to use them with the Amazon S3 console, you must grant additional permissions that are required by the console. For information about using policies such as these with the Amazon S3 console, see Controlling access to a bucket with user policies.
Topics
- Allowing an IAM user access to one of your buckets
- Allowing each IAM user access to a folder in a bucket
- Allowing a group to have a shared folder in Amazon S3
- Allowing all your users to read objects in a portion of a bucket
- Allowing a partner to drop files into a specific portion of a bucket
- Restricting access to Amazon S3 buckets within a specific AWS account
- Restricting access to Amazon S3 buckets within your organizational unit
- Restricting access to Amazon S3 buckets within your organization
- Granting permission to retrieve the PublicAccessBlock configuration for an AWS account
- Restricting bucket creation to one Region
Allowing an IAM user access to one of your buckets
In this example, you want to grant an IAM user in your AWS account access to one of
your buckets, amzn-s3-demo-bucket1
, and allow the user to add, update, and delete objects.
In addition to granting the s3:PutObject
, s3:GetObject
, and
s3:DeleteObject
permissions to the user, the policy also grants the
s3:ListAllMyBuckets
, s3:GetBucketLocation
, and
s3:ListBucket
permissions. These are the additional permissions required by
the console. Also, the s3:PutObjectAcl
and the s3:GetObjectAcl
actions are required to be able to copy, cut, and paste objects in the console. For an
example walkthrough that grants permissions to users and tests them using the console, see
Controlling access to a bucket with user policies.
{ "Version":"2012-10-17", "Statement":[ { "Effect":"Allow", "Action": "s3:ListAllMyBuckets", "Resource":"*" }, { "Effect":"Allow", "Action":["s3:ListBucket","s3:GetBucketLocation"], "Resource":"arn:aws:s3:::
amzn-s3-demo-bucket1
" }, { "Effect":"Allow", "Action":[ "s3:PutObject", "s3:PutObjectAcl", "s3:GetObject", "s3:GetObjectAcl", "s3:DeleteObject" ], "Resource":"arn:aws:s3:::amzn-s3-demo-bucket1
/*" } ] }
Allowing each IAM user access to a folder in a bucket
In this example, you want two IAM users, Mary and Carlos, to have access to your
bucket, amzn-s3-demo-bucket1
, so that they can add, update, and delete objects. However,
you want to restrict each user's access to a single prefix (folder) in the bucket. You
might create folders with names that match their
usernames.
amzn-s3-demo-bucket1
Mary
/Carlos
/
To grant each user access only to their folder, you can write a policy for each user and
attach it individually. For example, you can attach the following policy to the user Mary
to allow her specific Amazon S3 permissions on the
folder.amzn-s3-demo-bucket1
/Mary
{ "Version":"2012-10-17", "Statement":[ { "Effect":"Allow", "Action":[ "s3:PutObject", "s3:GetObject", "s3:GetObjectVersion", "s3:DeleteObject", "s3:DeleteObjectVersion" ], "Resource":"arn:aws:s3:::
amzn-s3-demo-bucket1
/Mary
/*" } ] }
You can then attach a similar policy to the user Carlos, specifying the folder
in the Carlos
Resource
value.
Instead of attaching policies to individual users, you can write a single policy that
uses a policy variable and then attach the policy to a group. First, you must create a
group and add both Mary and Carlos to the group. The following example policy allows a set
of Amazon S3 permissions in the
folder. When
the policy is evaluated, the policy variable amzn-s3-demo-bucket1
/${aws:username}${aws:username}
is replaced by
the requester's
username.
For example, if Mary sends a request to put an object, the operation is allowed only if
Mary is uploading the object to the
folder.amzn-s3-demo-bucket1
/Mary
{ "Version":"2012-10-17", "Statement":[ { "Effect":"Allow", "Action":[ "s3:PutObject", "s3:GetObject", "s3:GetObjectVersion", "s3:DeleteObject", "s3:DeleteObjectVersion" ], "Resource":"arn:aws:s3:::
amzn-s3-demo-bucket1
/${aws:username}/*" } ] }
Note
When using policy variables, you must explicitly specify version
2012-10-17
in the policy. The default version of the IAM policy
language, 2008-10-17, does not support policy variables.
If you want to test the preceding policy on the Amazon S3 console, the console requires additional permissions, as shown in the following policy. For information about how the console uses these permissions, see Controlling access to a bucket with user policies.
{ "Version":"2012-10-17", "Statement": [ { "Sid": "AllowGroupToSeeBucketListInTheConsole", "Action": [ "s3:ListAllMyBuckets", "s3:GetBucketLocation" ], "Effect": "Allow", "Resource": "arn:aws:s3:::*" }, { "Sid": "AllowRootLevelListingOfTheBucket", "Action": "s3:ListBucket", "Effect": "Allow", "Resource": "arn:aws:s3:::
amzn-s3-demo-bucket1
", "Condition":{ "StringEquals":{ "s3:prefix":[""], "s3:delimiter":["/"] } } }, { "Sid": "AllowListBucketOfASpecificUserPrefix", "Action": "s3:ListBucket", "Effect": "Allow", "Resource": "arn:aws:s3:::amzn-s3-demo-bucket1
", "Condition":{ "StringLike":{"s3:prefix":["${aws:username}/*"] } } }, { "Sid": "AllowUserSpecificActionsOnlyInTheSpecificUserPrefix", "Effect":"Allow", "Action":[ "s3:PutObject", "s3:GetObject", "s3:GetObjectVersion", "s3:DeleteObject", "s3:DeleteObjectVersion" ], "Resource":"arn:aws:s3:::amzn-s3-demo-bucket1
/${aws:username}/*" } ] }
Note
In the 2012-10-17 version of the policy, policy variables start with $
.
This change in syntax can potentially create a conflict if your object key (object name)
includes a $
.
To avoid this conflict, specify the $
character by using
${$}
. For example, to include the object key my$file
in a
policy, specify it as my${$}file
.
Although IAM user names are friendly, human-readable identifiers, they aren't required to be globally unique. For example, if the user Carlos leaves the organization and another Carlos joins, then the new Carlos could access the old Carlos's information.
Instead of using
usernames, you
could create folders based on
IAM user
IDs. Each
IAM user
ID is unique. In this case, you must modify the preceding policy to use the
${aws:userid}
policy variable. For more information about user identifiers,
see IAM Identifiers in the
IAM User Guide.
{ "Version":"2012-10-17", "Statement":[ { "Effect":"Allow", "Action":[ "s3:PutObject", "s3:GetObject", "s3:GetObjectVersion", "s3:DeleteObject", "s3:DeleteObjectVersion" ], "Resource":"arn:aws:s3:::
amzn-s3-demo-bucket1
/home/${aws:userid}/*" } ] }
Allowing non-IAM users (mobile app users) access to folders in a bucket
Suppose that you want to develop a mobile app, a game that stores users' data in an S3 bucket. For each app user, you want to create a folder in your bucket. You also want to limit each user's access to their own folder. But you can't create folders before someone downloads your app and starts playing the game, because you don't have their user ID.
In this case, you can require users to sign in to your app by using public identity providers such as Login with Amazon, Facebook, or Google. After users have signed in to your app through one of these providers, they have a user ID that you can use to create user-specific folders at runtime.
You can then use web identity federation in AWS Security Token Service to integrate information from the identity provider with your app and to get temporary security credentials for each user. You can then create IAM policies that allow the app to access your bucket and perform such operations as creating user-specific folders and uploading data. For more information about web identity federation, see About web identity Federation in the IAM User Guide.
Allowing a group to have a shared folder in Amazon S3
Attaching the following policy to the group grants everybody in the group access to the
following folder in Amazon S3:
. Group
members are allowed to access only the specific Amazon S3 permissions shown in the policy and
only for objects in the specified folder. amzn-s3-demo-bucket1
/share/marketing
{ "Version":"2012-10-17", "Statement":[ { "Effect":"Allow", "Action":[ "s3:PutObject", "s3:GetObject", "s3:GetObjectVersion", "s3:DeleteObject", "s3:DeleteObjectVersion" ], "Resource":"arn:aws:s3:::
amzn-s3-demo-bucket1
/share/marketing/*" } ] }
Allowing all your users to read objects in a portion of a bucket
In this example, you create a group named
, which contains all the IAM users
that are owned by the AWS account. You then attach a policy that gives the group access
to AllUsers
GetObject
and GetObjectVersion
, but only for objects in the
folder. amzn-s3-demo-bucket1
/readonly
{ "Version":"2012-10-17", "Statement":[ { "Effect":"Allow", "Action":[ "s3:GetObject", "s3:GetObjectVersion" ], "Resource":"arn:aws:s3:::
amzn-s3-demo-bucket1
/readonly/*" } ] }
Allowing a partner to drop files into a specific portion of a bucket
In this example, you create a group called
that represents a partner company.
You create an
IAM user
for the specific person or application at the partner company that needs access, and then
you put the user in the group. AnyCompany
You then attach a policy that gives the group PutObject
access to the
following folder in a bucket:
amzn-s3-demo-bucket1
/uploads/anycompany
You want to prevent the
group from
doing anything else with the bucket, so you add a statement that explicitly denies
permission to any Amazon S3 actions except AnyCompany
PutObject
on any Amazon S3 resource in the
AWS account.
{ "Version":"2012-10-17", "Statement":[ { "Effect":"Allow", "Action":"s3:PutObject", "Resource":"arn:aws:s3:::
amzn-s3-demo-bucket1
/uploads/anycompany
/*" }, { "Effect":"Deny", "Action":"s3:*", "NotResource":"arn:aws:s3:::amzn-s3-demo-bucket1
/uploads/anycompany
/*" } ] }
Restricting access to Amazon S3 buckets within a specific AWS account
If you want to ensure that your Amazon S3 principals are accessing only the resources that
are inside of a trusted AWS account, you can restrict access. For example, this identity-based
IAM policy uses a Deny
effect to block access to Amazon S3 actions,
unless the Amazon S3 resource that's being accessed is in account
. To prevent an IAM principal
in an AWS account from accessing Amazon S3 objects outside of the account, attach the
following IAM policy:222222222222
{ "Version": "2012-10-17", "Statement": [ { "Sid": "DenyS3AccessOutsideMyBoundary", "Effect": "Deny", "Action": [ "s3:*" ], "Resource": "*", "Condition": { "StringNotEquals": { "aws:ResourceAccount": [ "
222222222222
" ] } } } ] }
Note
This policy doesn't replace your existing IAM access controls, because it doesn't grant any access. Instead, this policy acts as an additional guardrail for your other IAM permissions, regardless of the permissions granted through other IAM policies.
Make sure to replace account ID
in
the policy with your own AWS account. To apply a policy to multiple accounts while still
maintaining this restriction, replace the account ID with the
222222222222
aws:PrincipalAccount
condition key. This condition requires that the
principal and the resource must be in the same account.
Restricting access to Amazon S3 buckets within your organizational unit
If you have an organizational unit
(OU) set up in AWS Organizations, you might want to restrict Amazon S3
bucket access to a specific part of your organization. In this example, we'll use the
aws:ResourceOrgPaths
key to restrict Amazon S3 bucket access to an OU in your
organization. For this example, the OU ID is
. Make sure to replace
this value in your own policy with your own OU IDs.ou-acroot-exampleou
{ "Version": "2012-10-17", "Statement": [ { "Sid": "AllowS3AccessOutsideMyBoundary", "Effect": "Allow", "Action": [ "s3:*" ], "Resource": "*", "Condition": { "ForAllValues:StringNotLike": { "aws:ResourceOrgPaths": [ "
o-acorg/r-acroot/ou-acroot-exampleou/
" ] } } } ] }
Note
This policy doesn't grant any access. Instead, this policy acts as a backstop for your other IAM permissions, preventing your principals from accessing Amazon S3 objects outside of an OU-defined boundary.
The policy denies access to Amazon S3 actions unless the Amazon S3 object that's being accessed is
in the
OU in your organization.
The IAM policy
condition requires ou-acroot-exampleou
aws:ResourceOrgPaths
, a multivalued condition
key, to contain any of the listed OU paths. The policy uses the
ForAllValues:StringNotLike
operator to compare the values of
aws:ResourceOrgPaths
to the listed OUs without case-sensitive
matching.
Restricting access to Amazon S3 buckets within your organization
To restrict access to Amazon S3 objects within your organization, attach an IAM policy to the root of the organization, applying it to all accounts in your organization. To require your IAM principals to follow this rule, use a service-control policy (SCP). If you choose to use an SCP, make sure to thoroughly test the SCP before attaching the policy to the root of the organization.
In the following example policy, access is denied to Amazon S3 actions unless the Amazon S3 object that's being accessed is in the same organization as the IAM principal that is accessing it:
{ "Version": "2012-10-17", "Statement": [ { "Sid": "DenyS3AccessOutsideMyBoundary", "Effect": "Deny", "Action": [ "s3:*" ], "Resource": "arn:aws:s3:::*/*", "Condition": { "StringNotEquals": { "aws:ResourceOrgID": "${aws:PrincipalOrgID}" } } } ] }
Note
This policy doesn't grant any access. Instead, this policy acts as a backstop for your other IAM permissions, preventing your principals from accessing any Amazon S3 objects outside of your organization. This policy also applies to Amazon S3 resources that are created after the policy is put into effect.
The IAM policy
condition in this example requires aws:ResourceOrgID
and
aws:PrincipalOrgID
to be equal to each other. With this requirement, the
principal making the request and the resource being accessed must be in the same
organization.
Granting permission to retrieve the PublicAccessBlock configuration for an AWS account
The following example identity-based policy grants the
s3:GetAccountPublicAccessBlock
permission to a user. For these
permissions, you set the Resource
value to "*"
. For
information about resource ARNs, see Policy
resources for Amazon S3.
{ "Version":"2012-10-17", "Statement":[ { "Sid":"statement1", "Effect":"Allow", "Action":[ "s3:GetAccountPublicAccessBlock" ], "Resource":[ "*" ] } ] }
Restricting bucket creation to one Region
Suppose that an AWS account administrator wants to grant its user (Dave)
permission to create a bucket in the South America (São Paulo) Region only. The
account administrator can attach the following user policy granting the
s3:CreateBucket
permission with a condition as shown. The
key-value pair in the Condition
block specifies the
s3:LocationConstraint
key and the sa-east-1
Region as its value.
Note
In this example, the bucket owner is granting permission to one of its users, so either a bucket policy or a user policy can be used. This example shows a user policy.
For a list of Amazon S3 Regions, see Regions and Endpoints in the AWS General Reference.
{ "Version":"2012-10-17", "Statement":[ { "Sid":"statement1", "Effect":"Allow", "Action": "s3:CreateBucket", "Resource": "arn:aws:s3:::*", "Condition": { "StringLike": { "s3:LocationConstraint": "sa-east-1" } } } ] }
Add explicit deny
The preceding policy restricts the user from creating a bucket in any other Region
except sa-east-1
. However, some other policy might grant this user
permission to create buckets in another Region. For example, if the user belongs to a
group, the group might have a policy attached to it that allows all users in the group
permission to create buckets in another Region. To ensure that the user doesn't get
permission to create buckets in any other Region, you can add an explicit deny statement
in the above policy.
The Deny
statement uses the StringNotLike
condition. That
is, a create bucket request is denied if the location constraint is not
sa-east-1
. The explicit deny doesn't allow the user to create a bucket
in any other Region, no matter what other permission the user gets. The
following
policy includes an explicit deny statement.
{ "Version":"2012-10-17", "Statement":[ { "Sid":"statement1", "Effect":"Allow", "Action": "s3:CreateBucket", "Resource": "arn:aws:s3:::*", "Condition": { "StringLike": { "s3:LocationConstraint": "sa-east-1" } } }, { "Sid":"statement2", "Effect":"Deny", "Action": "s3:CreateBucket", "Resource": "arn:aws:s3:::*", "Condition": { "StringNotLike": { "s3:LocationConstraint": "sa-east-1" } } } ] }
Test the policy with the AWS CLI
You can test the policy using the following create-bucket
AWS CLI command.
This example uses the bucketconfig.txt
file to specify the location
constraint. Note the
Windows
file path. You need to update the bucket name and path as appropriate. You must provide
user credentials using the --profile
parameter. For more information about
setting up and using the AWS CLI, see Developing with Amazon S3 using the AWS CLI.
aws s3api create-bucket --bucket
examplebucket
--profile AccountADave --create-bucket-configuration file://c:/Users/someUser/bucketconfig.txt
The bucketconfig.txt
file specifies the configuration
as follows.
{"LocationConstraint": "sa-east-1"}