Help Center> Scalable File Service> Service Overview> Limitations and Constraints
Updated on 2024-01-04 GMT+08:00

Limitations and Constraints

SFS Capacity-Oriented

Table 1 SFS Capacity-Oriented restrictions

Item

Description

Access method

SFS Capacity-Oriented file systems can only be accessed over the intranet and used on the cloud.

Supported protocols

  • Only NFSv3 is supported (NFSv4 is not supported), and CIFS is supported (SMB 2.0, 2.1, and 3.0 are supported, but SMB 1.0 is not supported).
  • Export options with NFSv3 include rw, no_root_squash, no_all_squash, and sync. Export options with CIFS include rw and sync.
  • Encrypted CIFS file systems do not support copychunk.
  • CIFS file systems cannot be mounted to Linux servers.
  • A file system can use either the NFS or CIFS protocol. It cannot use both protocols.
  • SMB file systems provide permission control only on the file systems, but not on files or directories.

Max. number of clients per file system

10,000

Max. capacity per file system

4 PB

Max. size of a single file

240 TB

Multi-VPC access

Supported. A maximum of 20 VPCs can be added for a file system, and a maximum of 400 ACL rules are allowed for all added VPCs.

Max. enterprise projects supported

20. If there are more than 20 enterprise projects, SFS Capacity-Oriented file systems may fail to be created. In this case, SFS Turbo file systems are recommended.

Replication

Not supported

Cross-region mounting

Not supported

Cross-account mounting

Not supported

SFS 3.0 Capacity-Oriented

Table 2 SFS 3.0 Capacity-Oriented restrictions

Item

Description

Access method

SFS 3.0 Capacity-Oriented file systems can only be accessed over the intranet.

Supported protocols

Only NFSv3 is supported (NFSv4 is not supported).

Max. number of clients per file system

10,000

File system encryption

Not supported

Max. number of files or subdirectories in a single directory

30 million

File system name

Must be globally unique. It cannot be the same as the name of any existing SFS 3.0 Capacity-Oriented file system, including one created by the current user or any other user. And it cannot be changed after the file system is created.

File system deletion

An SFS 3.0 Capacity-Oriented file system with the same name as a deleted SFS 3.0 file system can only be created 30 minutes after that file system has been deleted.

Client OS

  • Cannot be mounted to 32-bit Linux servers.
  • Cannot be mounted to Windows servers.

Changing root directory permissions

Not supported

Restrictions in the CCE and CCI scenarios

  • When SFS 3.0 Capacity-Oriented is used as the storage backend of CCE or CCI, you need to empty the in-use file systems before deleting any PVCs or PVs. If you directly delete the PVCs or PVs, the file systems may fail to be deleted. Check whether the file systems are deleted on the SFS 3.0 Capacity-Oriented console.
  • Deleting PVCs or PVs takes some time. The billing ends until the corresponding SFS 3.0 Capacity-Oriented file systems are deleted.

Lifecycle management

A maximum of 20 lifecycle rules can be configured for a file system.

SFS Turbo

Table 3 SFS Turbo restrictions

Item

General

Access method

VPN, Direct Connect, and Cloud Connect

Max. bandwidth

2 GB/s

Max. IOPS

100,000

Min. latency

1 to 2 ms

Max. capacity per file system

320 TB

Supported protocol

NFSv3

Max. number of clients per file system

500

Max. number of authorized VPCs per file system

20

Max. size of a single file

16 TB

Max. number of files or subdirectories per file system

1 billion

Max. number of files or subdirectories in a single directory

20 million

NOTE:

If you need to execute the ls, du, cp, chmod, or chown command on a directory, you are advised to place no more than 500,000 files or subdirectories in that directory. Otherwise, requests may take long times as the NFS protocol sends a large number of requests to traverse directory files and requests are queueing up.

Max. directory depth (unit: layer)

100

Max. path length (unit: byte)

1,024

Max. soft link length (unit: byte)

1,024

Max. hard link length (unit: byte)

255

Max. number of file systems

32 by default. You can submit a service ticket to increase the quota.

File system backup

Supported

Backup data restoring to the original file system

Not supported

NOTE:

SFS Turbo uses single-AZ deployment. SFS Turbo file systems will fail if their AZ fails.

File locking with Flock

Not supported

Cross-region mounting via domain name

Not supported

NOTE:
  • When you mount a file system using its domain name, cross-region mounting is not supported. To mount a file system across regions, use the file system IP address for mounting.
  • The file system domain name randomly resolves to a corresponding domain name server IP address, so some domain name servers may have lots of mounts while others have only a few. It is recommended that you use the domain name servers to the maximum extent as uneven mounts may affect the file system performance.

Cache acceleration

Not supported

File system tagging

  • A maximum of 10 tags can be added to a file system.
  • Tag keys of a file system must be unique.