Overview
Scalable File Service (SFS) provides scalable, high-performance (NAS) file storage. With SFS, you can enjoy shared file access spanning multiple Elastic Cloud Servers (ECSs), Bare Metal Servers (BMSs), and containers created on Cloud Container Engine (CCE). See Figure 1.
+Overview
Scalable File Service (SFS) provides scalable, high-performance NAS file storage. With SFS, you can enjoy shared file access spanning multiple Elastic Cloud Servers (ECSs), Bare Metal Servers (BMSs), and containers created on Cloud Container Engine (CCE), as shown in Figure 1.
-
- File sharing
Servers in multiple availability zones (AZs) of a same region can access the same file system concurrently and share files.
+Compared with traditional shared file storage, SFS has the following advantages:- File sharing
Cloud servers in multiple availability zones (AZs) of the same region can access the same file system concurrently and share files.
- Elastic scaling
Storage can be scaled up or down on demand to dynamically adapt to service changes without interrupting applications. You can complete resizing with a few clicks.
- - Superior performance
SFS enables file system performance to increase as capacity grows, and it delivers a high data durability to support rapid service growth.
-The backend storage system supports HDD and SSD storage media. It adopts a distributed architecture and uses full redundant design for modules, eliminating single-node faults.
- - Seamless integration
SFS supports Network File System (NFS). With this standard protocol, a broad range of mainstream applications can read and write data in the file system.
- - Easy operation
In an intuitive graphical user interface (GUI), you can create and manage file systems with ease.
+ - Superior performance
File system performance increases as capacity grows. SFS can support rapid service growth while ensuring a high data durability.
+The background storage system supports HDD and SSD storage media. It adopts a distributed architecture and uses full redundant design for modules, eliminating single-node faults.
+ - Seamless integration
SFS supports Network File System (NFS), through which a broad range of applications can read data from and write data to the file system.
+ - Easy operation
On an intuitive graphical user interface (GUI), you can create and manage file systems with ease.
- File sharing
Accessing SFS
You can access SFS on the management console or through APIs by sending HTTPS requests.
+Accessing SFS
You can access SFS on the console or through APIs by sending HTTPS requests.
- APIs
Use APIs if you need to integrate SFS into a third-party system for secondary development. For detailed operations, see Scalable File Service API Reference.
- - Management console
Use the console if you prefer a web-based UI to perform operations.
+ - Console
Use the console if you prefer a web-based UI to perform operations.
Create a File System
-You can create a file system and mount it to multiple servers. Then the servers can share this file system. You can create three types of file systems: SFS Capacity-Oriented, SFS Turbo, and General Purpose File System.
-Prerequisites
- Before creating an SFS Turbo, SFS Capacity-Oriented, or a general purpose file system, ensure that a VPC is available.
If no VPC is available, create one by referring to section "Creating a VPC" in the Virtual Private Cloud User Guide.
- - Before creating an SFS Turbo SFS Capacity-Oriented or a general purpose file system, ensure that ECSs are available and are in the created VPC.
If no ECS is available, create an ECS by referring to "Creating an ECS" in the Elastic Cloud Server User Guide.
+diff --git a/docs/sfs/umn/en-us_topic_0034428728.html b/docs/sfs/umn/en-us_topic_0034428728.html index 7c6ed974b..5cdea3709 100644 --- a/docs/sfs/umn/en-us_topic_0034428728.html +++ b/docs/sfs/umn/en-us_topic_0034428728.html @@ -1,10 +1,10 @@You can create a file system and mount it to multiple cloud servers. Then, the servers can share this file system. You can create three types of file systems: SFS Capacity-Oriented, SFS Turbo, and General Purpose File System.
+Prerequisites
- Before creating an SFS Turbo, SFS Capacity-Oriented, or a general purpose file system, ensure that a VPC is available.
If no VPC is available, create one by referring to section "Creating a VPC with a Subnet" in the Virtual Private Cloud User Guide.
+ - Before creating an SFS Turbo SFS Capacity-Oriented or a general purpose file system, ensure that ECSs are available and are in the created VPC.
If no ECS is available, create ECSs by referring to "Creating an ECS" in the Elastic Cloud Server User Guide.
Creating an SFS Capacity-Oriented File System
- Log in to the management console using a cloud account.
- Log in to the management console and select a region and a project.
- Choose Storage > Scalable File Service.
Creating a General Purpose File System
- Log in to the management console using a cloud account.
- Log in to the management console and select a region and a project.
- Choose Storage > Scalable File Service.
- In the navigation pane, choose General Purpose File System. In the upper right corner of the page, click Create File System.
- Set the file system parameters, as shown in Figure 2. Table 2 describes the parameters. +
- In the navigation pane on the left, choose General Purpose File System. In the upper right corner of the page, click Create File System.
- Set the file system parameters, as shown in Figure 2. Table 2 describes the parameters.
-
- Click Create Now.
- Confirm the file system information and click Submit.
- Complete the creation and go back to the file system list.
If the status of the created file system is Available, the file system is created successfully. If the status is Creation failed, contact the administrator.
+ - Click Create Now.
- Confirm the file system information and click Submit.
- Complete the creation and go back to the file system list.
If the status of the created file system is Available, the file system is created successfully. If the status is Creation failed, contact the administrator.
Mounting an NFS File System to ECSs (Linux)
-After creating a file system, you need to mount the file system to cloud servers so that they can share the file system.
+After creating a file system, you need to mount it to cloud servers so that they can share the file system.
In this section, ECSs are used as example servers. Operations on BMSs and containers (CCE) are the same as those on ECSs.
General purpose file systems cannot be mounted to 32-bit Linux servers.
-Prerequisites
- You have checked the OS type of each ECS. Different OSs require different commands to install the NFS client.
- You have created a file system and have obtained the mount point of the file system.
- At least one ECS that is in the same VPC as the file system is available.
- The IP address of the DNS server for resolving the domain names of the file systems has been configured on the ECS. SFS Turbo file systems do not require domain name resolution.
Prerequisites
- You have checked the OS type of each ECS. Different OSs require different commands to install the NFS client.
- You have created a file system and have obtained the mount point of the file system.
- At least one ECS that is in the same VPC as the file system is available.
- The IP address of the DNS server for resolving the file system domain name has been configured on the ECS. SFS Turbo file systems do not require domain name resolution.
Constraints
@@ -43,14 +43,14 @@ nfs-client -
This constraint only applies to local paths (mount points) and does not affect other files or directories.
- Run the following command to check whether the domain name in the file system mount point can be resolved. SFS Turbo file systems do not require domain name resolution. Skip this step and directly mount the file system.
nslookup File system domain name
+- Run the following command to check whether the domain name in the file system mount point can be resolved. SFS Turbo file systems do not require domain name resolution. Skip this step and directly mount the file system.
nslookup File system domain name
- A file system domain name is just a part of the mount point, for example, sfs-nas1.example.com. You can obtain a file system domain name from the mount point of a file system. In this step, you are not supposed to enter the entire mount point but only the domain name.
- If the nslookup command cannot be used, install the bind-utils software package by running the yum install bind-utils command.
- If the domain name can be resolved, go to 5.
- If the domain name cannot be resolved, configure the DNS server IP address and then mount the file system. For details, see Configuring DNS.
- Run the following command to create a local path for mounting the file system:
mkdir Local path
- If there is any resource, such as a disk, already mounted on the local path, create a new path. (NFS clients do not refuse repeated mounts. If there are repeated mounts, information of the last successful mount is displayed.)
- Run the following command to mount the file system to the ECS in the same VPC as the file system. You can now mount the file system to Linux ECSs using NFSv3 only.
Table 1 describes the variables.
+- Mount the file system to the ECSs in the same VPC as the file system. You can now mount the file system to Linux ECSs using NFSv3 only.
Table 1 describes the variables.
To mount an SFS Capacity-Oriented file system, run the following command: mount -t nfs -o vers=3,timeo=600,noresvport,nolock Mount point Local path
To mount an SFS Turbo file system, run the following command: mount -t nfs -o vers=3,timeo=600,noresvport,nolock,tcp Mount point Local path
To mount a general purpose file system, run the following command: mount -t nfs -o vers=3,timeo=600,noresvport,nolock,proto=tcp Mount point Local path
@@ -94,7 +94,7 @@ nfs-clientThe format for an SFS Capacity-Oriented file system is File system domain name:/Path, for example, example.com:/share-xxx.
The format for a general purpose file system is File system domain name:/File system name, for example, example.com:/xxx.
-For an SFS Turbo Standard, Standard-Enhanced, Performance, or Performance-Enhanced file system, the format is File system IP address:/, for example, 192.168.0.0:/.
+The format for an SFS Turbo Standard, Standard-Enhanced, Performance, or Performance-Enhanced file system is File system IP address:/, for example, 192.168.0.0:/.
For an SFS Turbo 1,000 MB/s/TiB, 500 MB/s/TiB, 250 MB/s/TiB, 125 MB/s/TiB, 40 MB/s/TiB, or 20 MB/s/TiB file system, the format is File system domain name:/, for example, xxx.sfsturbo.internal:/.
NOTE:diff --git a/docs/sfs/umn/en-us_topic_0105224109.html b/docs/sfs/umn/en-us_topic_0105224109.html index 490639f1e..6d2053166 100644 --- a/docs/sfs/umn/en-us_topic_0105224109.html +++ b/docs/sfs/umn/en-us_topic_0105224109.html @@ -1,10 +1,10 @@- Variable x is a digit or letter.
- If the mount point is too long to display completely, you can adjust the column width.
- Hover the mouse over the mount point to display the complete mount command.
Mounting an NFS File System to ECSs (Windows)
-After creating a file system, you need to mount the file system to cloud servers so that they can share the file system.
+After creating a file system, you need to mount it to cloud servers so that they can share the file system.
This section uses Windows Server 2012 as the example OS to describe how to mount an NFS file system. For other versions, perform the steps based on the actual situation.
In this section, ECSs are used as example servers. Operations on BMSs and containers (CCE) are the same as those on ECSs.
-Prerequisites
- You have created a file system and have obtained the mount point of the file system.
- At least one ECS that is in the same VPC as the file system is available.
- The IP address of the DNS server for resolving the file system domain name has been configured on the ECS. For details, see Configuring DNS. SFS Turbo file systems do not require domain name resolution.
Prerequisites
- You have created a file system and have obtained the mount point of the file system.
- At least one ECS that is in the same VPC as the file system is available.
- The IP address of the DNS server for resolving the file system domain name has been configured on the ECSs. For details, see Configuring DNS. SFS Turbo file systems do not require domain name resolution.
Constraints
SFS Turbo file systems cannot be mounted to Windows ECSs.
General purpose file systems cannot be mounted to Windows servers.
@@ -19,7 +19,7 @@- Right-click Client for NFS, choose Properties, change the transport protocol to TCP, and select Use hard mounts, as shown in Figure 6 and Figure 7.
-- Check that the IP address of the DNS server for resolving the domain names of the file systems has been configured on the ECS before mounting the file system. For details, see Configuring DNS. SFS Turbo file systems do not require domain name resolution.
- Run the following command in the Command Prompt of the Windows Server 2012 (X is the drive letter of the free disk). Select the ECS that is in the same VPC as the file system to mount the file system.
For SFS Capacity-Oriented file systems: mount -o nolock Mount point X:
+- Check that the IP address of the DNS server for resolving the file system domain name has been configured on the ECS before mounting the file system. For details, see Configuring DNS. SFS Turbo file systems do not require domain name resolution.
- Run the following command in the Command Prompt of the Windows Server 2012 (X is the drive letter of the free disk). Select the ECS that is in the same VPC as the file system to mount the file system.
For SFS Capacity-Oriented file systems: mount -o nolock Mount point X:
- Free drive letter of the disk: A drive letter that is not in use, such as drive letter E or X.
You can move the cursor to the mount point and click
diff --git a/docs/sfs/umn/sfs_01_0004.html b/docs/sfs/umn/sfs_01_0004.html index 7e7a2d5a3..df6f41371 100644 --- a/docs/sfs/umn/sfs_01_0004.html +++ b/docs/sfs/umn/sfs_01_0004.html @@ -2,17 +2,17 @@
next to the mount point to copy it. For details, see Figure 8. If the information shown in Figure 9 is displayed, the mount is successful.Application Scenarios
-SFS Capacity-Oriented and General Purpose File System
Expandable to petabytes, SFS Capacity-Oriented provides fully hosted shared file storage. It features high availability and durability, and seamlessly handles data-intensive and bandwidth-intensive applications. It is suitable for multiple scenarios, including high-performance computing (HPC), media processing, file sharing, as well as content management and web services.
-- Media processing
Services of TV stations and new media are more likely to be deployed on cloud platforms than before. Such services include streaming media, archiving, editing, transcoding, content distribution, and video on demand (VoD). In such scenarios, a large number of workstations are involved in the whole program production process. Different operating systems may be used by different workstations, requiring file systems to share materials. In addition, HD/4K videos have become a major trend in the broadcasting and TV industry. Taking video editing as an example, to improve audiences' audiovisual experience, HD editing is being transformed to 30- to 40-layer editing. A single editing client may require a file system with a bandwidth up to hundreds of MB per second. Usually, producing a single TV program needs several editing clients to process a lot of video materials concurrently. To meet such requirement, SFS provides customers with stable, bandwidth-intensive, and latency-sensitive performance.
+- Media processing
Services of TV stations and new media are more likely to be deployed on cloud platforms than before. Such services include streaming media, archiving, editing, transcoding, content distribution, and video on demand (VOD). In such scenarios, a large number of workstations are involved in the whole program production process. Different operating systems may be used by different workstations, requiring file systems to share materials. In addition, HD/4K videos have become a major trend in the broadcasting and TV industry. Taking video editing as an example, to improve audiences' audiovisual experience, HD editing is being transformed to 30- to 40-layer editing. A single editing client may require a file system with a bandwidth up to hundreds of MB per second. Usually, producing a single TV program needs several editing clients to process a lot of video materials concurrently. To meet such requirement, SFS provides customers with stable, bandwidth-intensive, and latency-sensitive performance.
- Content management and web service
SFS can be used in various content management systems to store and provide information for websites, home directories, online releases, and archiving.
- - Big data and analytic applications
SFS delivers an aggregate bandwidth of up to 10 Gbit/s, capable of handling ultra-large data files such as satellite images.
+ - Big data and analytic applications
SFS delivers an aggregate bandwidth of up to 10 GB/s, capable of handling ultra-large data files such as satellite images.
- Media processing
SFS Turbo
Expandable to 32 TB, SFS Turbo provides fully hosted shared file storage. It features high availability and durability to support massive small files and applications requiring low latency and high IOPS. SFS Turbo is perfect to scenarios such as high-performance websites, log storage, compression and decompression, DevOps, enterprise offices, and container applications.
+SFS Turbo
Expandable to 32 TB, SFS Turbo provides fully hosted shared file storage. It features high availability and durability to support massive small files and applications requiring low latency and high IOPS. You can use SFS Turbo for high-performance websites, log storage, compression and decompression, DevOps, enterprise OA, and containerized applications.
- High-performance websites
For I/O-intensive website services, SFS Turbo can provide shared website source code directories for multiple web servers, enabling low-latency and high-IOPS concurrent shared access.
- Log storage
SFS Turbo can provide shared log output directories for multiple service nodes, facilitating log collection and management of distributed applications.
+- Log storage
SFS Turbo can provide multiple service nodes with a shared log output directory, easily collecting and managing distributed applications' logs.
- DevOps
The development directory can be shared with multiple VMs or containers, which simplifies the configuration process and improves R&D experience.
+- DevOps
The development directory can be shared with multiple VMs or containers, simplifying the configuration process and improving R&D experience.
- Enterprise OA
Office documents of enterprises or organizations can be saved in an SFS Turbo file system for high-performance shared access.
File System Encryption
SFS provides you with the encryption function. You can encrypt data on newly created file systems if needed.
Keys for encrypting file systems are provided by Key Management Service (KMS), which is secure and convenient. You do not need to establish and maintain key management infrastructure. If you want to use your own key material, use the key import function on the KMS console to create a custom key whose key material is empty and import the key material to the custom key. For details, see section "Importing Key Materials" in Key Management Service User Guide.
-To use the file system encryption function, you need to authorize SFS Capacity-Oriented to access KMS when creating an SFS Capacity-Oriented file system. SFS Turbo file systems do not need authorization.
-Encryption Key
Keys provided by KMS for encrypting SFS Capacity-Oriented file systems include a default key and custom keys.
+To use file system encryption, you need to authorize SFS to access KMS when creating an SFS Capacity-Oriented or general purpose file system. SFS Turbo file systems do not need authorization.
+Encryption Key
Keys used for encrypting SFS Capacity-Oriented and general purpose file systems are provided by KMS. They include a default key and custom keys.
- Default key: SFS automatically creates a default key and names it sfs/default.
The default key cannot be disabled and does not support scheduled deletion.
- - Custom keys: Existing or newly created custom keys. For details, see Creating a Custom Key in the Key Management Service User Guide.
If the custom key used by the encrypted file system is disabled or scheduled for deletion, the file system can only be used within a certain period of time (30s by default). Exercise caution in this case.
+ - Custom keys: Existing or newly created custom keys. For details, see Creating a Key in the Key Management Service User Guide.
If the custom key used by the encrypted file system is disabled or scheduled for deletion, the file system can only be used within a certain period of time (30s by default). Exercise caution in this case.
SFS Turbo file systems do not have default keys. You can use your existing key or create a key. For details, see section "Creating a Key" in the Key Management Service User Guide.
-General Purpose File System does not support file system encryption.
Who Has the Rights to Encrypt File Systems?
- The security administrator who has the "Security Administrator" permission can grant the KMS access rights for encryption.
- A common user who does not have the "Security Administrator" permission needs to contact the system administrator to obtain the "Security Administrator" permission.
As long as the KMS access rights have been granted to SFS Capacity-Oriented, all common users in the same region can directly use the encryption function.
diff --git a/docs/sfs/umn/sfs_01_0007.html b/docs/sfs/umn/sfs_01_0007.html index fd6b7adc2..aabf93e94 100644 --- a/docs/sfs/umn/sfs_01_0007.html +++ b/docs/sfs/umn/sfs_01_0007.html @@ -41,7 +41,7 @@The encryption feature relies on KMS, which improves the data security of your file systems.
+File system encryption depends on KMS. You can use the keys provided by KMS to encrypt file systems to improve data security.
Key Management Service (KMS)
General
- SFS supports the NFSv3 protocol only. The default export options are rw, no_root_squash, no_all_squash, and sync.
- Currently, SFS does not support replication.
- Currently, SFS does not support cross-region access.
- SFS Capacity-Oriented is not suitable for file storage scenarios requiring low latency and high IOPS, such as database services, website building, and code storage.
-SFS Capacity-Oriented
- SFS Capacity-Oriented can be accessed only on the intranet and used only on the cloud.
- Only NFSv3 is supported (NFSv4 is not supported).
- A maximum of 10,000 compute nodes can be mounted to and access a single file system at the same time.
- The maximum capacity of a single file system is 2 PB, and that of a single file is 240 TB.
- Multi-VPC access is supported. You can add a maximum of 20 VPCs for one file system and create a maximum of 400 ACL rules for all added VPCs.
- NFSv3 is supported (NFSv4 is not supported).
- A maximum of 10,000 compute nodes can be mounted to and access a single file system at the same time.
- The maximum capacity of a single file system is 2 PB, and that of a single file is 240 TB.
- Multi-VPC access is supported. You can add a maximum of 20 VPCs for one file system and create a maximum of 400 ACL rules for all added VPCs.
General Purpose File System
- Only the NFSv3 protocol is supported (NFSv4 is not supported).
- General purpose file systems do not support file system encryption.
- General purpose file systems can only be accessed over the intranet.
- A single directory can contain a maximum of 30 million files.
- General purpose file systems cannot be mounted to 32-bit Linux servers.
- The name of a general purpose file system must be globally unique. It cannot be the same as the name of any existing general purpose file system or one created by any other user. And it cannot be changed after the file system is created.
- If a general purpose file system is deleted, you can create a new file system with the same name after at least 12 hours.
- General purpose file systems cannot be mounted to Windows servers.
- General purpose file systems' root directory permissions cannot be changed.
- When general purpose file systems are used as the storage backend of CCE or CCI, you need to empty the file systems before you can delete 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 General Purpose File System console.
Deleting PVCs or PVs takes some time. The billing ends until the corresponding general purpose file systems are deleted.
+General Purpose File System
- Only the NFSv3 protocol is supported (NFSv4 is not supported).
- General purpose file systems can only be accessed over the intranet.
- A single directory can contain a maximum of 30 million files.
- General purpose file systems cannot be mounted to 32-bit Linux servers.
- The name of a general purpose file system must be globally unique. It cannot be the same as the name of any existing general purpose file system or one created by any other user. And it cannot be changed after the file system is created.
- If a general purpose file system is deleted, you can create a new file system with the same name after at least 12 hours.
- General purpose file systems cannot be mounted to Windows servers.
- General purpose file systems' root directory permissions cannot be changed.
- When general purpose file systems are used as the storage backend of CCE or CCI, you need to empty the file systems before you can delete 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 General Purpose File System console.
Deleting PVCs or PVs takes some time. The billing ends until the corresponding general purpose file systems are deleted.
SFS Turbo
- Only the NFSv3 protocol is supported (NFSv4 is not supported).
- A maximum of 500 compute nodes can be mounted to and access a single file system at the same time.
- The maximum capacity of a single file system is 32 TB, and the maximum size of a single file allowed is 320 TB.
- Maximum number of files supported by a single file system = Capacity/16. For example, the maximum number of files supported by a 500 GB file system is 32,768,000 (500 GB/16 KB = 500 x 1024 x 1024/16).
- By default, a single directory can contain a maximum of 20 million files.
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 queuing up.
diff --git a/docs/sfs/umn/sfs_01_0025.html b/docs/sfs/umn/sfs_01_0025.html index c73661ade..f5f7e08c9 100644 --- a/docs/sfs/umn/sfs_01_0025.html +++ b/docs/sfs/umn/sfs_01_0025.html @@ -2,7 +2,7 @@Mounting a File System Automatically
File system mount information may be lost after a server is restarted. You can configure auto mount on the server to avoid losing the mount information.
-Restrictions
Because service startup sequences in different OSs vary, some servers running CentOS may not support the following auto mount plan. In this case, manually mount the file system.
+Constraints
Because service startup sequences in different OSs vary, some servers running CentOS may not support the following auto mount plan. In this case, manually mount the file system.
Procedure (Linux)
- Log in to the management console using a cloud account.
- Log in to the management console and select a region and a project.
- Choose Compute > Elastic Cloud Server to go to the ECS console.
- Log in to the ECS as user root.
- Run the vi /etc/fstab command to edit the /etc/fstab file.At the end of the file, add the file system information, for example:
Mount point /local_path nfs vers=3,timeo=600,nolock 0 0
@@ -32,7 +32,7 @@
- Log in to the management console using a cloud account.
/local_path
A directory created on the ECS used to mount the file system. Set it to the local path in the mount command in Mounting an NFS File System to ECSs (Linux).
+A directory on the ECS used to mount the file system. Set it to the local path in the mount command in Mounting an NFS File System to ECSs (Linux).
nfs
diff --git a/docs/sfs/umn/sfs_01_0032.html b/docs/sfs/umn/sfs_01_0032.html index cde2c01d0..5af5c0ea5 100644 --- a/docs/sfs/umn/sfs_01_0032.html +++ b/docs/sfs/umn/sfs_01_0032.html @@ -10,7 +10,7 @@Constraints
- Both system-defined policies and custom policies are supported in SFS Turbo file systems.
Process Flow
-- Create a user group and assign permissions to it.
Create a user group on the IAM console, and attach the SFS Turbo ReadOnlyAccess policy to the group.
+- Create a user group and assign permissions to it.
Create a user group on the IAM console and attach the SFS Turbo ReadOnlyAccess policy to the group.
- Create a user and add it to a user group.
Create a user on the IAM console and add the user to the group created in 1.
- Log in and verify permissions.
Log in to the SFS console using the created user, and verify that the user only has read permissions for SFS.
- Choose Scalable File Service. Click Create File System on the SFS console. If a message appears indicating that you have insufficient permissions to perform the operation, the SFS Turbo ReadOnlyAccess policy has already taken effect.
- Choose any other service. If a message appears indicating that you have insufficient permissions to access the service, the SFS Turbo ReadOnlyAccess policy has already taken effect.
Creating a Custom Policy
Custom policies can be created to supplement the system-defined policies of SFS. For the actions supported for custom policies, see section "Permissions Policies and Supported Actions" in the Scalable File Service API Reference.
-You can create custom policies in either of the following two ways:
+You can create custom policies in either of the following ways:
- Visual editor: Select cloud services, actions, resources, and request conditions. This does not require knowledge of policy syntax.
- JSON: Edit JSON policies from scratch or based on an existing policy.
This section provides examples of common custom SFS policies.
Example Custom Policies
- Example 1: Allowing users to create file systems
{ @@ -17,7 +17,7 @@ ] } - Example 2: Denying file system deletion
A policy with only "Deny" permissions must be used in conjunction with other policies to take effect. If the permissions assigned to a user contain both "Allow" and "Deny", the "Deny" permissions take precedence over the "Allow" permissions.
-The following method can be used if you need to assign permissions of the SFS Turbo FullAccess policy to a user but also forbid the user from deleting file systems. Create a custom policy for denying file system deletion, and attach both policies to the group to which the user belongs. Then, the user can perform all operations on SFS except deleting file systems. The following is an example of a deny policy:
+Assume that you want to grant the permissions of the SFS Turbo FullAccess policy to a user but want to prevent them from deleting file systems. You can create a custom policy for denying file system deletion, and attach this policy together with the SFS Turbo FullAccess policy to the user. As an explicit deny in any policy overrides any allows, the user can perform all operations on file systems except deleting them. The following is an example of a deny policy:
{ "Version": "1.1", "Statement": [ diff --git a/docs/sfs/umn/sfs_01_0036.html b/docs/sfs/umn/sfs_01_0036.html index 47d0d73c2..53f839bed 100644 --- a/docs/sfs/umn/sfs_01_0036.html +++ b/docs/sfs/umn/sfs_01_0036.html @@ -62,14 +62,14 @@
- Create a user group and assign permissions to it.
User Root Permission
Whether to allow the root permission of the client. The default value is no_root_squash.
+Whether to allow the client to access as root. The default value is no_root_squash.
- root_squash: Clients cannot access as the root user. When a client accesses as the root user, the user is mapped to the nobody user.
- no_root_squash: Clients are allowed to access as the root user who has full control and access permissions of the root directories.
Priority
The value must be an integer ranging from 0 to 100. 0 indicates the highest priority, and 100 indicates the lowest priority. In the same VPC, the permission of the IP address or address segment with the highest priority is preferentially used. If some IP addresses or address segments are of the same priority, the permission of the most recently added or modified one is used.
-For example, if the IP address for mounting is 10.1.1.32 and both 10.1.1.32 (read/write) with priority 100 and 10.1.1.0/24 (read-only) with priority 50 meet the requirements, the permission of 10.1.1.0/24 (read-only) with priority 50 is used. That is, if there is no other authorized priority, the permission of all IP addresses in the 10.1.1.0/24 segment, including 10.1.1.32, is read-only.
+For example, if the client IP address is 10.1.1.32 and both authorized addresses 10.1.1.32 (with read/write permission and priority 100) and 10.1.1.0/24 (with read-only permission and priority 50) meet the requirements, the permission of 10.1.1.0/24 (read-only) is used because it has a higher priority. If no higher-priority rule applies, all IP addresses in 10.1.1.0/24, including 10.1.1.32, receive read-only permission.
Authorizations
The value can be Read/Write or Read-only. The default value is Read/Write.
+The value can be Read/Write or Read-only. The default value is Read/Write.
The format is as follows:nameserver 100.125.4.25
- Press Esc, input :wq, and press Enter to save the changes and exit the vi editor.
- Run the following command to check whether the IP address is successfully added:
cat /etc/resolv.conf
-- Run the following command to check whether an IP address can be resolved from the file system domain name:
nslookup File system domain name
+- Run the following command to check whether an IP address can be resolved from the file system domain name:
nslookup File system domain name
Obtain the file system domain name from the file system mount point.
- (Optional) In a network environment that uses the DHCP server, edit the /etc/resolv.conf file to prevent the file from being automatically modified upon an ECS restart, and prevent the DNS server IP address added in 2 from being reset.
- Run the following command to lock the file:chattr +i /etc/resolv.conf -
Run the chattr -i /etc/resolv.conf command to unlock the file if needed.
@@ -21,7 +21,7 @@Procedure (Windows)
- Go to the ECS console and log in to the ECS running Windows Server 2012.
- Click This PC in the lower left corner.
- On the page that is displayed, right-click Network and choose Properties from the drop-down list. The Network and Sharing Center page is displayed, as shown in Figure 3. Click Local Area Connection.
+-
Procedure (Windows)
- Go to the ECS console and log in to the ECS running Windows Server 2012.
- Click This PC in the lower left corner.
- On the page that is displayed, right-click Network and choose Properties from the drop-down list. The Network and Sharing Center page is displayed, as shown in Figure 3. Click Local Area Connection.
- In the Activity area, select Properties. See Figure 4.
- In the Local Area Connection Properties dialog box that is displayed, select Internet Protocol Version 4 (TCP/IPv4) and click Properties. See Figure 5.
- In the dialog box that is displayed, select Use the following DNS server addresses: and configure DNS, as shown in Figure 6. The DNS server IP address is 100.125.4.25. After completing the configuration, click OK. diff --git a/docs/sfs/umn/sfs_01_0039.html b/docs/sfs/umn/sfs_01_0039.html index 68f782725..a17aede6e 100644 --- a/docs/sfs/umn/sfs_01_0039.html +++ b/docs/sfs/umn/sfs_01_0039.html @@ -66,7 +66,7 @@
- In the displayed dialog box, confirm the information and click OK.
- In the file system list, check the capacity information after resizing.
- On the displayed page, confirm the information and click OK.
- In the file system list, check the capacity information after resizing.
diff --git a/docs/sfs/umn/sfs_01_0042.html b/docs/sfs/umn/sfs_01_0042.html index 56df8806e..8c6d31a56 100644 --- a/docs/sfs/umn/sfs_01_0042.html +++ b/docs/sfs/umn/sfs_01_0042.html @@ -1,9 +1,8 @@Encryption
-Creating an Encrypted File System
To use the file system encryption function, you need to authorize SFS Capacity-Oriented to access KMS when creating an SFS Capacity-Oriented file system. If you have the Security Administrator rights, grant SFS the permissions to access KMS directly. Otherwise, you need to contact the system administrator to obtain the "Security Administrator" rights first. For details, see File System Encryption.
-SFS Turbo file systems do not require authorization.
-You can create a file system that is encrypted or not, but you cannot change the encryption attribute of an existing file system.
+-Creating an Encrypted File System
- To use SFS Capacity-Oriented file system encryption, you need to authorize SFS to access KMS when creating an SFS Capacity-Oriented file system. If you have the Security Administrator permissions, grant SFS the KMS access permissions directly. Otherwise, you need to contact the system administrator to obtain the Security Administrator permissions first. For details, see File System Encryption.
- To use general purpose file system encryption, you can directly select server-side encryption when creating a general purpose file system. Ensure that you have the kms:cmk:get, kms:cmk:list, kms:cmk:decrypt, kms:cmk:create, kms:cmk:encrypt, kms:dek:create, kms:dek:decrypt, kms:dek:encrypt, iam:agencies:getAgency, and iam:agencies:createAgency permissions before creating encrypted file systems. For details about how to obtain required IAM action permissions, see section "Creating a Custom Policy" in the Identity and Access Management User Guide. If this is your first time using encryption, the system will prompt you the Create Agency window. You only need to click OK to have the required authorization automatically granted to you. The system will automatically create an agency named SFSAccessKMS to grant KMS access permissions to SFS. The delegated account and permissions of the agency are op_svc_sfs and KMS Administrator. After permissions are granted, SFS can obtain KMS keys to encrypt and decrypt file systems.
- You can directly use encryption when creating SFS Turbo file systems. No authorization is required.
You can create an SFS file system that is encrypted or not, but you cannot change the encryption attribute of an existing file system.
For details about how to create an encrypted file system, see Create a File System.
Unmounting an Encrypted File System
If the custom key used by the encrypted file system is disabled or scheduled for deletion, the file system can only be used within a certain period of time (30s by default). Exercise caution in this case.
diff --git a/docs/sfs/umn/sfs_01_0047.html b/docs/sfs/umn/sfs_01_0047.html index b14cf6875..ff506853a 100644 --- a/docs/sfs/umn/sfs_01_0047.html +++ b/docs/sfs/umn/sfs_01_0047.html @@ -166,7 +166,7 @@Viewing Monitoring Statistics
- Log in to the management console.
- Choose Management & Deployment > Cloud Eye > Cloud Service Monitoring > Scalable File Service. In the file system list, click View Metric in the Operation column of the desired file system.
- View the SFS file system monitoring data by metric or monitored duration.
Figure 1 shows the monitoring graphs. For more information about Cloud Eye, see the Cloud Eye User Guide.
+diff --git a/docs/sfs/umn/sfs_01_0050.html b/docs/sfs/umn/sfs_01_0050.html index 50a431ada..70a4755f1 100644 --- a/docs/sfs/umn/sfs_01_0050.html +++ b/docs/sfs/umn/sfs_01_0050.html @@ -15,49 +15,49 @@ -Viewing Monitoring Statistics
- Log in to the management console.
- Choose Management & Deployment > Cloud Eye > Cloud Service Monitoring > Scalable File Service. In the file system list, click View Metric in the Operation column of the desired file system.
- View the SFS file system monitoring data by metric or monitored duration.
Figure 1 shows the monitoring graphs. For more information about Cloud Eye, see the Cloud Eye User Guide.
Creating a shared file system
+Creating a file system
sfs
createShare
Modifying a shared file system
+Modifying a file system
sfs
updateShareInfo
Deleting a shared file system
+Deleting a file system
sfs
deleteShare
Adding a share access rule
+Adding a file system access rule
sfs
addShareACL
Deleting a share access rule
+Deleting a file system access rule
sfs
deleteShareACL
Expanding a shared file system
+Expanding the capacity of a file system
sfs
extendShare
Shrinking a shared file system
+Shrinking the capacity of a file system
sfs
- A File System Is Automatically Disconnected from the Server
-
- A Client Server Failed to Access a File System
+- A Client Server Failed to Access a General Purpose File System
- File System Is Abnormal
diff --git a/docs/sfs/umn/sfs_01_0057.html b/docs/sfs/umn/sfs_01_0057.html index 485eb400f..80f73dd3a 100644 --- a/docs/sfs/umn/sfs_01_0057.html +++ b/docs/sfs/umn/sfs_01_0057.html @@ -11,12 +11,12 @@
- Cause 2: The server and the file system are not in the same VPC. -
- Cause 3: The mount point in the mount command is incorrect.
- Log in to the management console and check whether the mount point is the same as the one in the mount command.
- If the mount point in the mount command is incorrectly entered, correct it and run the command again.
- Cause 4: The IP address used for accessing SFS is a virtual IP address.
Log in to the server and run the ping command and use the server IP address to access SFS. Check whether the service is reachable. See Figure 1.
+- Cause 3: The mount point in the mount command is incorrect.
- Log in to the management console and check whether the mount point is the same as the one in the mount command.
- If the mount point in the mount command is incorrectly entered, correct it and run the command again.
- Cause 4: The IP address used for accessing SFS is a virtual IP address.
Log in to the server and run the ping command and use the server IP address to access SFS. Check whether the service is reachable. See Figure 1.
- If yes, the network problem has been resolved. Check other possible causes.
- If no, the network is disconnected. Use the server's private IP address and the ping command to access SFS and check whether the service is reachable.
- Cause 5: The DNS used for accessing the file system is incorrect.
diff --git a/docs/sfs/umn/sfs_01_0058.html b/docs/sfs/umn/sfs_01_0058.html index 17cfa960c..b457be729 100644 --- a/docs/sfs/umn/sfs_01_0058.html +++ b/docs/sfs/umn/sfs_01_0058.html @@ -1,6 +1,6 @@ -Run the following command to check whether the DNS is correct:
-nslookup File system domain name
+nslookup File system domain name
Check whether the resolved IP address is in segment 100.
- If yes, the DNS configuration is correct. Check other possible causes.
- If no, the DNS configuration is incorrect. Reconfigure DNS by referring to Configuring DNS.
A Client Server Failed to Access a File System
+A Client Server Failed to Access a General Purpose File System
Symptom
Access from a client server to a file system was denied. All services on the server were abnormal.
Possible Causes
- Cause 1: The file system is abnormal.
- Cause 2: The file system fails to be mounted to the server after being forcibly unmounted.
Gene sequencing, image rendering, media processing, file sharing, content management, and web services
Big data analysis, static website hosting, online video on demand (VoD), gene sequencing, and intelligent video surveillance
+Big data analysis, static website hosting, online video on demand (VOD), gene sequencing, and intelligent video surveillance
Industrial design, energy exploration, critical clustered applications, enterprise application systems, and development and testing
- What Is the Maximum Size of a File That Can Be Stored in a File System?
- - What Access Protocols Are Supported by SFS?
+ - What Access Protocols Does SFS Support?
- - How Many File Systems Can Be Created by Each Account?
+ - How Many File Systems Can I Create with One Account?
- How Many Cloud Servers Can I Mount a File System To?
diff --git a/docs/sfs/umn/sfs_01_0072.html b/docs/sfs/umn/sfs_01_0072.html
index bfa72ba8a..92d89cd49 100644
--- a/docs/sfs/umn/sfs_01_0072.html
+++ b/docs/sfs/umn/sfs_01_0072.html
@@ -1,6 +1,6 @@
- - SFS Capacity-Oriented file systems can be created in batches. To create more than 10 SFS Capacity-Oriented file systems, click Increase quota on the page for creating a file system.
- Only one SFS Turbo file system can be created at a time. To create more than 10 SFS Turbo file systems, contact customer service to apply for a higher quota. For details, see How Do I Apply for a Higher Quota?
- Can the Capacity of a File System Be Expanded?
+ - Can I Expand the File System Capacity If I Start to Run Out of Space?
- Can I Migrate My File System Data to Another Region?
- - Can a File System Be Mounted to Multiple Accounts?
+ - Can I Mount a File System Across Accounts?
- Can a File System Be Accessed Across VPCs?
+ - Can I Access a File System Across VPCs?
- Does SFS Support Cross-Region Mounting?
- - Does the Security Group of a VPC Affect SFS?
+ - Does the Security Group of a VPC Affect the Use of SFS?
- What Can I Do If Data of My File System Is Not the Same When Accessed from Two Client Servers?
diff --git a/docs/sfs/umn/sfs_01_0080.html b/docs/sfs/umn/sfs_01_0080.html
index f405c0cf3..72bf7b002 100644
--- a/docs/sfs/umn/sfs_01_0080.html
+++ b/docs/sfs/umn/sfs_01_0080.html
@@ -1,6 +1,6 @@
- - Multi-VPC access can be configured for an SFS Capacity-Oriented or a general purpose file system so that servers in different VPCs can share the same file system, as long as the VPCs that the servers belong to are added as authorized VPCs of the file system or the server IP addresses are added as authorized IP addresses of the VPCs. For details, see Configuring Multi-VPC Access.
- An SFS Turbo file system allows two or more VPCs in the same region to interconnect with each other through VPC peering connection. In this case, different VPCs are in the same network, and servers in these VPCs can share the same file system. For details about VPC peering connection, see section "VPC Peering Connection" in Virtual Private Cloud User Guide.
- How Do I Access a File System from a Client Server?
- - How Do I Check Whether a File System on a Linux Server Is Available?
+ - How Do I Check Whether a File System Is Available on a Linux Server?
- What Resources Does SFS Occupy?
diff --git a/docs/sfs/umn/sfs_01_0093.html b/docs/sfs/umn/sfs_01_0093.html
index f7c9dc086..c97d03521 100644
--- a/docs/sfs/umn/sfs_01_0093.html
+++ b/docs/sfs/umn/sfs_01_0093.html
@@ -1,8 +1,8 @@
- - For SFS Capacity-Oriented or general purpose file systems:
- When a file system is created, the inbound rules of ports 111, 445, 2049, 2051, and 2052 are enabled in the security group entered by the user. The default source IP address is 0.0.0.0/0. You can change the IP address as required.
- If an encrypted SFS Capacity-Oriented file system is created, the KMS key entered by the user is used for encryption. Note that if the key is deleted, data in the file system cannot be used.
- For SFS Turbo file systems:
- When an SFS Turbo file system is created, two private IP addresses and one virtual IP address are created in the subnet entered by the user.
- When an SFS Turbo file system is created, the inbound rules of ports 111, 445, 2049, 2051, 2052, and 20048 are enabled in the security group selected by the user. The default source IP address is 0.0.0.0/0. You can change the IP address as required.
- If an encrypted SFS Turbo file system is created, the KMS key entered by the user is used for encryption. Note that if the key is deleted, data in the file system cannot be used.
- For SFS Capacity-Oriented or general purpose file systems:
- When a file system is created, the inbound rules of ports 111, 445, 2049, 2051, and 2052 are enabled in the security group you select. The default source IP address is 0.0.0.0/0. You can change the IP address as required.
- If an encrypted SFS Capacity-Oriented or general purpose file system is created, the KMS key you select is used for encryption. Note that if the key is deleted, data in the file system cannot be used.
- For SFS Turbo file systems:
- When an SFS Turbo file system is created, two private IP addresses and one virtual IP address are created in the subnet you select.
- When an SFS Turbo file system is created, the inbound rules of ports 111, 445, 2049, 2051, 2052, and 20048 are enabled in the security group you select. The default source IP address is 0.0.0.0/0. You can change the IP address as required.
- If an encrypted SFS Turbo file system is created, the KMS key you select is used for encryption. Note that if the key is deleted, data in the file system cannot be used.
When data is written to the folders of a file system, the running memory of the server is occupied, but the storage space of the server disk is not occupied. The file system uses independent space.
- A non-root user has been created on the ECS.
- A file system has been created and can be mounted to the ECS as root.
- The mount point of the file system has been obtained.
- A non-root user has been created on the ECS.
- A file system has been created and can be mounted to the ECS as root.
- The mount point of the file system has been obtained.
- Log in to the ECS as user root.
- Assign the permissions of user root to the non-root user.
- Run the chmod 777 /etc/sudoers command to change the sudoers file to be editable.
- Use the which command to view the mount and umount command paths.Figure 1 Viewing command paths
- Run the vi /etc/sudoers command to edit the sudoers file.
- Add a common user under the root account. In this example, user Mike is added.Figure 2 Adding a userdiff --git a/docs/sfs/umn/sfs_01_0104.html b/docs/sfs/umn/sfs_01_0104.html index 0074acee5..e435e67d4 100644 --- a/docs/sfs/umn/sfs_01_0104.html +++ b/docs/sfs/umn/sfs_01_0104.html @@ -8,7 +8,13 @@
What Access Protocols Are Supported by SFS?
+What Access Protocols Does SFS Support?
SFS Capacity-Oriented and SFS Turbo all support the standard NFSv3 protocol.
diff --git a/docs/sfs/umn/sfs_01_0073.html b/docs/sfs/umn/sfs_01_0073.html index 16b2df491..2e4ae2b20 100644 --- a/docs/sfs/umn/sfs_01_0073.html +++ b/docs/sfs/umn/sfs_01_0073.html @@ -1,6 +1,6 @@ -How Many File Systems Can Be Created by Each Account?
+How Many File Systems Can I Create with One Account?
You can create a maximum of 10 SFS Capacity-Oriented file systems and 10 SFS Turbo file systems with each account.
You can create a maximum of 100 general purpose file systems with one account.
-
-
Can the Capacity of a File System Be Expanded?
+Can I Expand the File System Capacity If I Start to Run Out of Space?
diff --git a/docs/sfs/umn/sfs_01_0078.html b/docs/sfs/umn/sfs_01_0078.html index ea82b3e9e..ea104e856 100644 --- a/docs/sfs/umn/sfs_01_0078.html +++ b/docs/sfs/umn/sfs_01_0078.html @@ -1,6 +1,6 @@ -Both SFS Capacity-Oriented and SFS Turbo file systems can be expanded by capacity resizing.
General purpose file systems have no capacity limit and do not support resizing.
Can a File System Be Mounted to Multiple Accounts?
+Can I Mount a File System Across Accounts?
SFS Capacity-Oriented file systems can be mounted across accounts.
General purpose file systems cannot be mounted across accounts.
SFS Turbo file systems can be mounted across accounts based on VPC peering. For details about VPC peering connection and usage instructions, see section "VPC Peering Connection" in Virtual Private Cloud User Guide.
diff --git a/docs/sfs/umn/sfs_01_0079.html b/docs/sfs/umn/sfs_01_0079.html index 7c61a3bc0..671dbeee8 100644 --- a/docs/sfs/umn/sfs_01_0079.html +++ b/docs/sfs/umn/sfs_01_0079.html @@ -4,11 +4,11 @@-
-
Can a File System Be Accessed Across VPCs?
+Can I Access a File System Across VPCs?
diff --git a/docs/sfs/umn/sfs_01_0081.html b/docs/sfs/umn/sfs_01_0081.html index f6c43e939..31c2b3867 100644 --- a/docs/sfs/umn/sfs_01_0081.html +++ b/docs/sfs/umn/sfs_01_0081.html @@ -1,7 +1,7 @@ -Yes.
Does the Security Group of a VPC Affect SFS?
-A security group is a collection of access control rules for ECSs that have the same security protection requirements and are mutually trusted in a VPC. After a security group is created, you can create different access rules for the security group to protect ECSs that are added to this security group. The default security group rule allows all outgoing data packets. ECSs in a security group can access each other without the need to add rules. The system creates a security group for each cloud account by default. Users can also create custom security groups by themselves.
+Does the Security Group of a VPC Affect the Use of SFS?
+A security group is a collection of access control rules for ECSs that have the same security protection requirements and are mutually trusted in a VPC. After a security group is created, you can create different access rules for the security group to protect the ECSs that are added to this security group. The default security group rule allows all outbound data packets. ECSs in a security group can communicate with each other without the need to add rules. The system creates a security group for each cloud account by default. You can also create custom security groups by yourself.
After an SFS Turbo file system is created, the system automatically enables the security group port required by the NFS protocol. This ensures that the SFS Turbo file system can be accessed by your servers and prevents file system mounting failures. The inbound ports required by the NFS protocol are ports 111, 2049, 2051, 2052, and 20048. If you need to change the enabled ports, go to the VPC console, choose Access Control > Security Groups, locate the target security group, and change the ports.
You are advised to use an independent security group for an SFS Turbo file system to isolate it from service nodes.
You need to add inbound and outbound rules for the security group of an SFS Capacity-Oriented file system. For details, see section "Adding a Security Group Rule" in the Virtual Private Cloud User Guide. In an SFS Capacity-Oriented file system, the inbound ports required by the NFS protocol are ports 111, 2049, 2051, and 2052. The inbound port required by the DNS server is port 53.
diff --git a/docs/sfs/umn/sfs_01_0082.html b/docs/sfs/umn/sfs_01_0082.html index d6f3515c3..8fad513ea 100644 --- a/docs/sfs/umn/sfs_01_0082.html +++ b/docs/sfs/umn/sfs_01_0082.html @@ -3,7 +3,7 @@What Can I Do If Data of My File System Is Not the Same When Accessed from Two Client Servers?
-Symptom
A file system was mounted to two servers. There was a delay in synchronizing files from one server to another. However, there was no delay when files were uploaded to a server.
Fault Diagnosis
Add noac, lookupcache=none to the mount command.
+diff --git a/docs/sfs/umn/sfs_01_0091.html b/docs/sfs/umn/sfs_01_0091.html index 2e50a69ec..8a4b2db4c 100644 --- a/docs/sfs/umn/sfs_01_0091.html +++ b/docs/sfs/umn/sfs_01_0091.html @@ -6,7 +6,7 @@Fault Diagnosis
Add noac,lookupcache=none to the mount command.
The noac option disables file attribute caching and forces write synchronization. By default, an NFS client's file attribute information is cached using the ac option to improve performance, and the client checks file attribute information periodically and updates it if there are any changes. Within the cache validity period, the client does not check whether file attribute information on the server is changed. By default, the value of this option is ac. Set it to noac.
The lookupcache option is related to directory entry caching, and the value can be all, none, pos, or positive. With lookupcache=none, the client neither trust the positive nor negative lookup results. In this way, lookup caching is disabled.
How Do I Check Whether a File System on a Linux Server Is Available?
+How Do I Check Whether a File System Is Available on a Linux Server?
Log in to the server as the root user. Run the following command to list all available file systems with the specified domain name or IP address:
-showmount -e File system domain name or IP address
+showmount -e File system domain name or IP address
diff --git a/docs/sfs/umn/sfs_01_0094.html b/docs/sfs/umn/sfs_01_0094.html index 5ade59c13..184053b34 100644 --- a/docs/sfs/umn/sfs_01_0094.html +++ b/docs/sfs/umn/sfs_01_0094.html @@ -2,8 +2,8 @@What Resources Does SFS Occupy?
diff --git a/docs/sfs/umn/sfs_01_0100.html b/docs/sfs/umn/sfs_01_0100.html index 8acbec4c2..f6869d04f 100644 --- a/docs/sfs/umn/sfs_01_0100.html +++ b/docs/sfs/umn/sfs_01_0100.html @@ -3,7 +3,7 @@To ensure that file systems can be used properly, the service occupies the following resources:
-Mounting a File System to a Linux ECS as a Non-root User
-Scenarios
By default, a Linux ECS allows only the root user to use the mount command to mount file systems, but you can grant the permissions of user root to other users. Such users can then use the mount command to mount the file systems. The following describes how to mount a file system to a Linux ECS as a non-root user. EulerOS is used in this example.
Prerequisites
Prerequisites
Procedure
2026-01-07
+2026-01-21
+Updated the following content:
+- Added support for server-side encryption for general purpose file systems and added related description.
- Added FAQs in section Managing SFS Turbo+OBS Storage Interworking.
2026-01-07
Updated the following content:
- Added descriptions about the 500 MB/s/TiB and 1,000 MB/s/TiB SFS Turbo file system types.
- Updated section Notes and Constraints.
- Updated section Create a File System.
- Updated section Managing SFS Turbo+OBS Storage Interworking.
2024-11-21
Updated the following content:
-- Updated the maximum capacity of a general purpose file system.
- Added What Are the Differences Between SFS, OBS, and EVS?
- Added Can I Migrate My File System Data to Another Region?
- Added Can a File System Be Mounted to Multiple Accounts?
- Added Does SFS Support Cross-Region Mounting?
- Added What Can I Do If Data of My File System Is Not the Same When Accessed from Two Client Servers?
- Updated the maximum capacity of a general purpose file system.
- Added What Are the Differences Between SFS, OBS, and EVS?
- Added Can I Migrate My File System Data to Another Region?
- Added Can I Mount a File System Across Accounts?
- Added Does SFS Support Cross-Region Mounting?
- Added What Can I Do If Data of My File System Is Not the Same When Accessed from Two Client Servers?
2024-10-23
Updated the following content:
-- Added descriptions about General Purpose File System.
- Deleted section "Supported OSs."
- Added descriptions about General Purpose File System.
- Deleted section "Supported Operating Systems."
2024-07-30
diff --git a/docs/sfs/umn/sfs_01_0125.html b/docs/sfs/umn/sfs_01_0125.html index a62b84d3e..c37179fb4 100644 --- a/docs/sfs/umn/sfs_01_0125.html +++ b/docs/sfs/umn/sfs_01_0125.html @@ -5,7 +5,7 @@-Possible Causes
The ECS security group configuration is incorrect. The port used to communicate with the file system is not enabled.
Fault Diagnosis
Check whether the port of the target server is enabled and correctly configure the port on the security group console.
+Fault Diagnosis
On the security group console, check the port configuration for the target ECS and ensure that required ports are correctly configured.
-Solution
- Log in to the ECS console.
- Log in to the management console.
- Click
in the upper left corner and select your desired region and project. - Under Compute, choose Elastic Cloud Server.
- In the navigation pane on the left, choose Elastic Cloud Server. On the page displayed, select the target server. Go to the server details page.
- Click the Security Groups tab and select the target security group. Click Manage Rule to go to the security group console.
- On the displayed page, click the Inbound Rules tab and click Add Rule. The Add Inbound Rule page is displayed. Add rules as follows:
After an SFS Turbo file system is created, the system automatically enables the security group port required by the NFS protocol. This ensures that the SFS Turbo file system can be accessed by your servers and prevents file system mounting failures. The inbound ports required by the NFS protocol are ports 111, 2049, 2051, 2052, and 20048. If you need to change the enabled ports, go to the VPC console, choose Access Control > Security Groups, locate the target security group, and change the ports.
diff --git a/docs/sfs/umn/sfs_01_0134.html b/docs/sfs/umn/sfs_01_0134.html index b1a9d55c4..34b741f44 100644 --- a/docs/sfs/umn/sfs_01_0134.html +++ b/docs/sfs/umn/sfs_01_0134.html @@ -5,7 +5,7 @@Before mounting a general purpose file system to a compute resource, you need to create a VPC endpoint in the region where the compute resource belongs.
VPC endpoints are not required for SFS Capacity-Oriented and SFS Turbo file systems.
Prerequisites
- Before creating a general purpose file system, ensure that a VPC is available.
If no VPC is available, create one by referring to section "Creating a VPC" in the Virtual Private Cloud User Guide.
+diff --git a/docs/sfs/umn/sfs_01_0334.html b/docs/sfs/umn/sfs_01_0334.html index 964b6eaf8..487ce32f2 100644 --- a/docs/sfs/umn/sfs_01_0334.html +++ b/docs/sfs/umn/sfs_01_0334.html @@ -3,7 +3,7 @@Prerequisites
- Before creating a general purpose file system, ensure that a VPC is available.
If no VPC is available, create one by referring to section "Creating a VPC with a Subnet" in the Virtual Private Cloud User Guide.
- Before creating a general purpose file system, ensure that ECSs are available and in the created VPC.
If no ECS is available, create ECSs by referring to "Creating an ECS" in the Elastic Cloud Server User Guide.
Managing SFS Turbo+OBS Storage Interworking
-Overview
In scenarios like AI training and inference, high-performance data preprocessing, EDA, rendering, and simulation, you can use SFS Turbo file systems to speed access to your data in OBS buckets. After binding a directory in your file system with an OBS bucket, you can synchronize data between the file system and bucket through import and export tasks. You can enjoy the following benefits from SFS Turbo file caching: Before starting upper-layer training tasks, you can preload data in your OBS bucket to an SFS Turbo file system to speed up data access. Intermediate data and result data generated from upper-layer tasks is written to SFS Turbo file systems at a high speed. Downstream services can read and process the intermediate data, and you can asynchronously export the result data to OBS buckets for long-term low-cost storage. In addition, SFS Turbo allows you to configure a cache data eviction duration to delete data that has not been accessed for a long time to free up the cache space.
Constraints
- SFS Turbo file system types that support storage interworking: 1,000 MB/s/TiB, 500 MB/s/TiB 250 MB/s/TiB, 125 MB/s/TiB, 40 MB/s/TiB, and 20 MB/s/TiB
- After you have configured storage interworking between OBS and a directory in SFS Turbo, operations like creating hard links, configuring directory quota limits, and renaming are no longer supported.
- You can configure a maximum of 16 interworking directories for a single SFS Turbo file system.
- Adding OBS buckets as storage backends depends on the OBS service, so you must have the OBS Administrator permissions.
- Files and directories with the same name cannot coexist in directories of the same level.
- The maximum supported path length is 1,023 characters.
- For import tasks, the length of a file or subdirectory name cannot exceed 255 bytes.
- If an OBS bucket has WORM enabled, you can only import data from OBS to SFS Turbo, but cannot export data from SFS Turbo to OBS.
- OBS buckets configured with server-side encryption and OBS parallel file systems cannot be added as storage backends.
- Windows systems do not allow special characters (/\:*?"<>|) in file or folder names. You are advised not to use these characters when naming files or folders. Otherwise, files (objects) or directories downloaded to Windows from SFS Turbo file systems or OBS buckets will be renamed.
Constraints
- SFS Turbo file system types that support storage interworking: 1,000 MB/s/TiB, 500 MB/s/TiB, 250 MB/s/TiB, 125 MB/s/TiB, 40 MB/s/TiB, and 20 MB/s/TiB
- After you have configured storage interworking between OBS and a directory in SFS Turbo, operations like creating hard links, configuring directory quota limits, and renaming are no longer supported.
- You can configure a maximum of 16 interworking directories for a single SFS Turbo file system.
- Adding OBS buckets as storage backends depends on the OBS service, so you must have the OBS Administrator permissions.
- Files and directories with the same name cannot coexist in directories of the same level.
- The maximum supported path length is 1,023 characters.
- For import tasks, the length of a file or subdirectory name cannot exceed 255 bytes.
- If an OBS bucket has WORM enabled, you can only import data from OBS to SFS Turbo, but cannot export data from SFS Turbo to OBS.
- OBS buckets configured with server-side encryption and OBS parallel file systems cannot be added as storage backends.
- Windows systems do not allow special characters (/\:*?"<>|) in file or folder names. You are advised not to use these characters when naming files or folders. Otherwise, files (objects) or directories downloaded to Windows from SFS Turbo file systems or OBS buckets will be renamed.
@@ -111,7 +111,7 @@Adding an OBS Bucket
- Access the SFS Turbo file system list page.
- In the file system list, click the name of the desired file system to go to its details page.
- On the Storage Backends tab, click Add OBS Bucket.Figure 1 Add OBS Bucket@@ -91,7 +91,7 @@
SFS Turbo supports two metadata import methods: quick import and additional metadata import. After the metadata import is complete, you can view the imported directories and files in the interworking directory.
- Quick import: Use quick import if data in the bucket has not been exported from SFS Turbo before. A quick import only imports the object metadata (name, size, and last modification time). After the import is complete, SFS Turbo will, by default, generate the additional metadata (uid, gid, directory permissions, and file permissions). If you want to specify the permissions of imported directories and files, follow the instructions in section "Creating a Data Import or Export Task" in the Scalable File Service API Reference. Such an operation is only valid for the current task. Quick import is faster, so you are advised to use quick import.
- Additional metadata import: Use additional metadata import if data in the bucket has been exported from SFS Turbo before. With additional metadata import, both the object metadata (name, size, and last modification time) and the additional metadata (uid, gid, and mode) will be imported. If there is no additional metadata, the permissions you specified will be used for imported directories and files.
- Find the added OBS bucket and click Import Metadata in the Operation column.Figure 4 Import Metadata-
- Set Object Prefix to the prefix of objects in the OBS bucket. It can be a specific object name. To import metadata of all the objects in the OBS bucket, leave the prefix field empty.
- Select Import Additional Metadata to import additional metadata. If this option is not selected, the system will perform a quick import.
- Click OK.
- Set Object Prefix to the prefix of objects in the OBS bucket. It can be a specific object name. To import metadata of all the objects in the OBS bucket, leave the object prefix field empty.
- Select Import Additional Metadata to import additional metadata. If this option is not selected, the system will perform a quick import.
- Click OK.
- After you import data from OBS to SFS Turbo, if new data is written to the bucket or existing data is modified, you need to import the data to SFS Turbo again.
- The length of a file or subdirectory name cannot exceed 255 bytes.
Data export allows you to export to the OBS bucket the files newly created in the interworking directory or the objects previously imported and then modified in the interworking directory. You can specify a prefix for data export. Then, only directories and files that match the specified prefix will be exported to the bucket.
- Find the added OBS bucket and click More > Export in the Operation column.Figure 6 Export
- Set File Prefix to the path of directories or files (excluding the interworking directory name) or that of a specific file. To export all files in the interworking directory to the bucket, leave the file prefix field empty.
- Click OK.
diff --git a/docs/sfs/umn/sfs_01_0348.html b/docs/sfs/umn/sfs_01_0348.html index bdcb8b3fe..55ae2e50a 100644 --- a/docs/sfs/umn/sfs_01_0348.html +++ b/docs/sfs/umn/sfs_01_0348.html @@ -1,7 +1,7 @@ - Before data is exported, SFS Turbo starts asynchronous tasks to scan the files in the target directories. If there is any file that has been updated in the last 10 seconds, this file will not be exported.
- For a given file, if no changes were made since the last time it was exported to OBS, it will not be exported in the next export task even if the previously exported file has been deleted from the OBS bucket.
- After files are exported to OBS, any SFS Turbo metadata with a name starting with x-obs-meta-sfsturbo-st- will be included in the objects' custom metadata.
- The maximum file path that supports export is 1,023 characters.
- The maximum file size supported in an SFS Turbo file system is 320 TB, and the maximum file size that can be exported is 48.8 TB.
- When large files are exported, temporary files generated during the export will be stored in the x-obs-upload-sfsturbo-temp-part directory in the bucket. After the export is complete, SFS Turbo will automatically delete this directory as well as the temporary files in it.
- When a file is exported from SFS Turbo to OBS:
If it was previously imported to and then modified in SFS Turbo, it will overwrite its peer object in the bucket if it is newer. Otherwise, it will not overwrite its peer object in the bucket. During an overwritten process, the file's peer object in the bucket is deleted first, and then the updated file is written to the bucket. This may remove a peer object that is still in use. When auto export is enabled, if a file in the SFS Turbo file system is frequently modified, it is recommended that you avoid accessing its peer object in the OBS bucket.
+@@ -140,6 +140,7 @@ - Before data is exported, SFS Turbo starts asynchronous tasks to scan the files in the target directories. If there is any file that has been updated in the last 10 seconds, this file will not be exported.
- For a given file, if no changes were made since the last time it was exported to OBS, it will not be exported in the next export task even if the previously exported file has been deleted from the OBS bucket.
- After files are exported to OBS, any SFS Turbo metadata with a name starting with x-obs-meta-sfsturbo-st- will be included in the objects' custom metadata.
- The maximum file path that supports export is 1,023 characters.
- The maximum file size supported in an SFS Turbo file system is 320 TB, and the maximum file size that can be exported is 48.8 TB.
- When large files are exported, temporary files generated during the export will be stored in the x-obs-upload-sfsturbo-temp-part directory in the bucket. After the export is complete, SFS Turbo will automatically delete this directory as well as the temporary files in it.
- When a file is exported from SFS Turbo to OBS:
If it was previously imported to and then modified in SFS Turbo, it will overwrite its peer object in the bucket if it is newer. Otherwise, it will not overwrite its peer object in the bucket. During an overwrite process, the file's peer object in the bucket is deleted first, and then the updated file is written to the bucket. This may remove a peer object that is still in use. When auto export is enabled, if a file in the SFS Turbo file system is frequently modified, it is recommended that you avoid accessing its peer object in the OBS bucket.
If you upload an object to OBS when an object with the same name is being exported, the object you uploaded may be overwritten.
- If the OBS bucket is enabled with WORM, data cannot be exported.
- In what scenarios will data import fail?
When the SFS Turbo file system contains only the file metadata (only metadata is imported or data eviction happens) and the object in the OBS bucket has been deleted, importing data or accessing the file will fail.
- Are the import or export tasks synchronous or asynchronous?
Tasks are asynchronous. After a task is submitted, you can query the task status based on the task ID.
- If I delete the files in the SFS Turbo interworking directory, will the objects in the OBS bucket be deleted as well?
If auto synchronization is disabled, the answer is no. If auto synchronization is enabled, the answer is yes.
+ - Why isn't my file exported after I edit it with vim, even though auto synchronization is set to export changed data?
In the SFS Turbo file system, when you edit a file (data or metadata) with vim, the editor doesn't just modify the file. It actually deletes and recreates it. Auto synchronization treats this as a new file rather than a changed file, so the changes cannot be exported with the "changed data" policy. To fix this, configure auto synchronization to export new data. That way, files edited with vim will be properly exported.
Mounting a File System Times Out
-Symptom
When a file system was mounted to a cloud server using the mount command, message timed out was returned.
+Symptom
When a file system was mounted to a cloud server using the mount command, message timed out was returned.
@@ -9,14 +9,14 @@Possible Causes
- Cause 1: The network status is not stable.
- Cause 2: The network connection is abnormal.
- Cause 3: The DNS configuration of the server is incorrect. As a result, the domain name of the file system cannot be resolved, and the mount fails. This issue will not occur on SFS Turbo file systems.
- Cause 4: The server that mounts the file system runs Ubuntu18 or later.
Solution
- Cause 1 and Cause 2: The network status is not stable or the network connection is abnormal.
Re-mount the file system after the network issue is addressed.
- If the mount is successful, no further action is required.
- If the problem persists, see the solution for cause 3.
- Cause 3: The DNS configuration of the server is incorrect. As a result, the domain name of the file system cannot be resolved, and the mount fails.
- Check the DNS configuration of the tenant and run the cat /etc/resolv.conf command.
- If the DNS has not been configured, configure it. For details, see Configuring DNS.
- If the DNS has been configured, run the following command to check whether the DNS is correct: +
- Cause 3: The DNS configuration of the server is incorrect. As a result, the domain name of the file system cannot be resolved, and the mount fails.
- Check the DNS configuration of the tenant and run the cat /etc/resolv.conf command.
- If the DNS has not been configured, configure it. For details, see Configuring DNS.
- If the DNS has been configured, run the following command to check whether the DNS is correct:
nslookup File system domain name
If the resolved IP address is in network segment 100, the DNS configuration is correct. If the IP address is in another network segment, the DNS configuration is incorrect. In this case, go to 2.
- Modify the /etc/resolv.conf configuration file, configure the correct tenant DNS, and run vi /etc/resolv.conf to edit the /etc/resolv.conf file. Add the DNS server IP address above the existing nameserver information. The DNS server IP address is 100.125.4.25.Figure 1 Configuring DNS
The format is as follows:nameserver 100.125.4.25
- If the configuration succeeds, go to 3.
- If the configuration fails, run the lsattr /etc/resolv.conf command. If the information shown in Figure 2 is displayed, the file is locked.
-
Run the chattr -i/etc/resolv.conf command to unlock the file. Then, re-configure the DNS and go to 3.
+Run the chattr -i /etc/resolv.conf command to unlock the file. Then, re-configure the DNS and go to 3.
- Press Esc, input :wq, and press Enter to save the changes and exit the vi editor.
- By default, the server inherits the DNS configuration of the VPC every time the server restarts. Changing only the server DNS configuration does not resolve the issue completely. You need to modify configurations in the VPC. Set a correct tenant DNS for the subnet of the VPC to which the ECS belongs. See Figure 3 and Figure 4.
- Check the DNS configuration of the tenant and run the cat /etc/resolv.conf command.
- Check the DNS configuration of the tenant and run the cat /etc/resolv.conf command.
- Before creating a general purpose file system, ensure that a VPC is available.
- DevOps
- Log storage
- Before creating an SFS Turbo, SFS Capacity-Oriented, or a general purpose file system, ensure that a VPC is available.
