Working with users, groups, and permissions at the Network File System (NFS) level
After creating a file system, by default only the root user (UID 0) has read, write, and execute permissions. For other users to modify the file system, the root user must explicitly grant them access. You can use access points to automate the creation of directories that a nonroot user can write from. For more information, see Working with Amazon EFS access points.
Amazon EFS file system objects have a Unix-style mode associated with them. This mode value defines the permissions for performing actions on that object. Users familiar with Unix-style systems can easily understand how Amazon EFS behaves with respect to these permissions.
Additionally, on Unix-style systems, users and groups are mapped to numeric identifiers, which Amazon EFS uses to represent file ownership. For Amazon EFS, file system objects (that is, files, directories, and so on) are owned by a single owner and a single group. Amazon EFS uses the mapped numeric IDs to check permissions when a user attempts to access a file system object.
Note
The NFS protocol supports a maximum of 16 group IDs (GIDs) per user and any additional GIDs are truncated from NFS client requests. For more information, see Access denied to allowed files on NFS file system.
Following, you can find examples of permissions and a discussion about NFS permissions considerations for Amazon EFS.
Topics
Example Amazon EFS file system use cases and permissions
After you create an Amazon EFS file system and mount targets for the file system in your VPC,
you can mount the remote file system locally on your Amazon EC2 instance. The mount
command can mount any directory in the file system. However, when you first create the file
system, there is only one root directory at /. The root user and root group
own the mounted directory.
The following mount command mounts the root directory of an Amazon EFS file
system, identified by the file system DNS name, on the /efs-mount-point local
directory.
sudo mount -t nfs -o nfsvers=4.1,rsize=1048576,wsize=1048576,hard,timeo=600,retrans=2,noresvportfile-system-id.efs.aws-region.amazonaws.com:/ efs-mount-point
The initial permissions mode allows:
-
read-write-executepermissions to the owner root -
read-executepermissions to the group root -
read-executepermissions to others
Only the root user can modify this directory. The root user can also grant other users permissions to write to this directory, for example:
-
Create writable per-user subdirectories. For step-by-step instructions, see Walkthrough: Create Writable Per-User Subdirectories and Configure Automatic Remounting on Reboot.
-
Allow users to write to the Amazon EFS file system root. A user with root privileges can grant other users access to the file system.
-
To change the Amazon EFS file system ownership to a non-root user and group, use the following:
$ sudo chownuser:group/EFSroot -
To change permissions of the file system to something more permissive, use the following:
$ sudo chmod 777 /EFSrootThis command grants read-write-execute privileges to all users on all EC2 instances that have the file system mounted.
-
User and group ID permissions for files and directories within a file system
Files and directories in an Amazon EFS file system support standard Unix-style read, write, and execute permissions based on the user ID and group IDs. When an NFS client mounts an EFS file system without using an access point, the user ID and group ID provided by the client is trusted. You can use EFS access points to override user ID and group IDs used by the NFS client. When users attempt to access files and directories, Amazon EFS checks their user IDs and group IDs to verify that each user has permission to access the objects. Amazon EFS also uses these IDs to indicate the owner and group owner for new files and directories that the user creates. Amazon EFS doesn't examine user or group names—it only uses the numeric identifiers.
Note
When you create a user on an EC2 instance, you can assign any numeric user ID (UID)
and group ID (GID) to the user. The numeric user IDs are set in the
/etc/passwd file on Linux systems. The numeric group IDs are in the
/etc/group file. These files define the mappings between names and IDs.
Outside of the EC2 instance, Amazon EFS doesn't perform any authentication of these IDs,
including the root ID of 0.
If a user accesses an Amazon EFS file system from two different EC2 instances, depending on whether the UID for the user is the same or different on those instances you see different behavior, as follows:
-
If the user IDs are the same on both EC2 instances, Amazon EFS considers them to indicate the same user, regardless of the EC2 instance used. The user experience when accessing the file system is the same from both EC2 instances.
-
If the user IDs aren't the same on both EC2 instances, Amazon EFS considers the users to be different users. The user experience isn't the same when accessing the Amazon EFS file system from the two different EC2 instances.
-
If two different users on different EC2 instances share an ID, Amazon EFS considers them to be the same user.
You might consider managing user ID mappings across EC2 instances consistently. Users
can check their numeric ID using the id command.
$ id uid=502(joe) gid=502(joe) groups=502(joe)
Turn Off the ID Mapper
The NFS utilities in the operating system include a daemon called an ID Mapper that
manages mapping between user names and IDs. In Amazon Linux, the daemon is called
rpc.idmapd and on Ubuntu is called idmapd. It translates user
and group IDs into names, and vice versa. However, Amazon EFS deals only with numeric IDs. We
recommend that you turn this process off on your EC2 instances. On Amazon Linux, the ID
mapper is usually disabled, and if it is don't enable it. To turn off the ID mapper, use
the commands shown following.
$ service rpcidmapd status $ sudo service rpcidmapd stop
No root squashing
By default, root squashing is disabled on EFS file systems. Amazon EFS behaves like a
Linux NFS server with no_root_squash. If a user or
group ID is 0, Amazon EFS treats that user as the root user, and bypasses
permissions checks (allowing access and modification to all file system objects).
Root squashing can be enabled on a client connection when the AWS Identity and Access Management (AWS IAM)
identity or resource policy does not allow access to the ClientRootAccess action.
When root squashing is enabled, the root user is converted to a user with limited permissions on the NFS server.
For more information, see Using IAM to control file system data access and Walkthrough: Enable root squashing using IAM authorization for NFS clients.
Permissions caching
Amazon EFS caches file permissions for a small time period. As a result, there might be a brief window where a user whose access was revoked recently can still access that object.
Changing file system object ownership
Amazon EFS enforces the POSIX chown_restricted attribute. This means only the
root user can change the owner of a file system object. The root or the owner user can
change the owner group of a file system object. However, unless the user is root, the group
can only be changed to one that the owner user is a member of.
EFS access points
An access point applies an operating system user, group, and file system path to any file system request made using the access point. The access point's operating system user and group override any identity information provided by the NFS client. The file system path is exposed to the client as the access point's root directory. This approach ensures that each application always uses the correct operating system identity and the correct directory when accessing shared file-based datasets. Applications using the access point can only access data in its own directory and below. For more information about access points, see Working with Amazon EFS access points.
