Search Docs...
⌘ K
OverviewDeploymentManagementOperationReferenceGlossary
    AFS 1.1.1
  • Managing file storage>
  • Managing file systems

Creating a file system

You can select File storage cluster in the sidebar on the left side of the Arcfra Operation Center (AOC) File storage page, click the name of the target cluster in the file storage cluster list, click + Create file system on the Overview page of the file storage cluster, and configure the following information for the file system in the pop-up dialog box.

Basic information

Parameter Description
Name The name of the file system.
Description (optional) The description added to the file system.
Storage capacity The logical capacity allocated to the file system from the file storage cluster.
Redundancy policy Supports multiple replicas and erasure coding technology.
  • Multiple replicas: One piece of data is saved in multiple copies on different hosts to improve data reliability and security.
    • 2 replicas
    • 3 replicas
  • Erasure coding: Through a specific algorithm, K original data blocks are processed to generate M parity blocks to achieve data recovery and fault tolerance.
Provisioning Supports thin provisioning, which only allocates real storage space when data is written, to improve the effective utilization of storage space.
Protocol The protocol selected for the file system:
  • NFS
  • HDFS

If you selected the NFS protocol, click Next to configure NFS access information for the file system. If not selected, click Create to complete the configuration of the file system.

NFS access information

NFS access Parameter Description Notes
NFS authentication UNIX system authentication The client provides local UID and GID for authentication, and the server trusts the UID and GID provided by the client. In this mode, the file system provides different file access permissions for different users by default, with high data security.
None All users on the client will be mapped to anonymous users on the server.

In this mode, all users have the same permissions and it is generally used for read-only scenarios.

When this mode is selected, regardless of which NFS Squash method is chosen, all users on the client will be mapped to anonymous users on the server.

When this mode is selected, sec=none needs to be configured when mounting the file system on the client.

NFS Squash Root squash When the client accesses as root user, it is mapped to an anonymous user on the server. In this mode, the root user's root access permissions are turned off, and data security is high.
No root squash When the client accesses as root user, it is mapped to the root user on the server. In this mode, the root user has root access permissions, including read, write, and delete permissions for all data.Please use with caution.
All squash All client users are mapped to anonymous users on the server. In this mode, all users have the same permissions and it is generally used for read-only scenarios.
Anonymous user UID The UID of the anonymous user on the server, the default value is -2, which can be modified. UID is displayed as a non-negative integer on the client, for example, -2 will be displayed as 4294967294 on the client.
Anonymous user group GID The GID of the anonymous user on the server, the default value is -2, which can be modified. GID is displayed as a non-negative integer on the client, for example, -2 will be displayed as 4294967294 on the client.
Default access right Read-write The client can mount the file system and read and write data. -
Read only The client can mount the file system and read data, but cannot write. -
No access The client cannot mount the file system and cannot read or write. -
Special rules (optional) Check as needed, you can add exception rules different from the default access right through the IP address of the client or the CIDR of the specified subnet. To set multiple exception rules, separate them with ,.

After filling in the configuration information, click Create.

Note:

The file system will be automatically taken online after creation.