Updated on 2025-08-29 GMT+08:00

Adding a Backend Server

Function

This API is used to add a backend server.

Constraints

When you add backend servers, note the following:

  • Two backend servers in the same backend server group must have different IP addresses and ports.

  • If no subnets are specified during cloud server creation, IP as backend servers can be added. In this case, address must be set to an IPv4 address, the protocol of the backend server group must be TCP, HTTP, or HTTPS, and IP as a Backend must have been enabled for the load balancer.

  • If a subnet is specified during cloud server creation, the subnet must be in the same VPC where the load balancer resides.

  • If the backend server group supports IPv4/IPv6 dual stack, address can be an IPv4 address or an IPv6 address. If the backend server group supports only IPv4, address can only be an IPv4 address.

  • If type of the backend server is set to instance, address must be a private IP address that is not used by any load balancer.

  • If the backend server group protocol is IP, protocol_port of the backend server must be 0 and IP as backend servers cannot be added.

Calling Method

For details, see Calling APIs.

URI

POST /v3/{project_id}/elb/pools/{pool_id}/members

Table 1 Path Parameters

Parameter

Mandatory

Type

Description

pool_id

Yes

String

Definition: Specifies the backend server group ID.

project_id

Yes

String

Definition: Specifies the project ID. For details about how to obtain a project ID, see Obtaining a Project ID.

Constraints: N/A

Range: The value can contain a maximum of 32 characters, including digits and lowercase letters.

Default value: N/A

Request Parameters

Table 2 Request header parameters

Parameter

Mandatory

Type

Description

X-Auth-Token

Yes

String

Definition: Specifies the token used for IAM authentication.

Constraints: N/A

Range: N/A

Default value: N/A

Table 3 Request body parameters

Parameter

Mandatory

Type

Description

member

Yes

CreateMemberOption object

Definition: Specifies the backend server.

Table 4 CreateMemberOption

Parameter

Mandatory

Type

Description

address

Yes

String

Definition: Specifies the IP address of the backend server.

Constraints:

  • If subnet_cidr_id is left blank, IP as backend servers will be added. In this case, the IP address must be an IPv4 address.

  • If subnet_cidr_id is not left blank, ECSs will be added as backend servers. It must be in the subnet specified by subnet_cidr_id.

admin_state_up

No

Boolean

Definition: Specifies the administrative status of the backend server.

Constraints: Although this parameter can be used in the APIs for creating and updating backend servers, its actual value depends on whether ECSs exist. If ECSs exist, the value is true. Otherwise, the value is false.

Range: true or false

name

No

String

Definition: Specifies the backend server name. Note: The name is not an ECS name. If this parameter is not specified, an empty value will be returned.

project_id

No

String

Definition: Specifies the project ID of the backend server.

protocol_port

No

Integer

Definition: Specifies the port used by the backend server to receive requests.

Constraints:

  • This parameter can be left blank because it does not take effect if any_port_enable is set to true for a backend server group.

  • protocol_port must be set to 0 for gateway load balancers with IP backend server groups associated.

subnet_cidr_id

No

String

Definition: Specifies the ID of the IPv4 or IPv6 subnet where the backend server resides. neutron_subnet_id defines IPv4 subnets, and neutron_network_id defines IPv6 subnets.

You can query parameter neutron_subnet_id in the response by calling the API (GET https://{VPC_Endpoint}/v1/{project_id}/subnets) to get the IPv4 subnet ID.

You can query parameter neutron_network_id in the response by calling the API (GET https://{VPC_Endpoint}/v1/{project_id}/subnets) to get the IPv6 subnet ID.

Constraints:

  • The IPv4 or IPv6 subnet must be in the same VPC as the subnet of the load balancer.

  • If ip_target_enable is set to true, this parameter can be left blank, indicating that an IP as backend server will be added. In this case, IP as backend servers must use private IPv4 addresses, and the protocol of the backend server group must be TCP, UDP, TLS, HTTP, HTTPS, QUIC, or GRPC.

  • If ip_target_enable is set to false, this parameter must be specified.

  • This parameter must be specified for gateway load balancers with IP backend server groups. The subnet of the backend server must be in the same VPC as the load balancer.

weight

No

Integer

Definition: Specifies the weight of the backend server. Requests are routed based on the load balancing algorithm configured for the backend server group and weights of backend servers.

The larger the weight is, the higher proportion of requests the backend server receives. If the weight is set to 0, the backend server will not accept new requests.

Constraints: If lb_algorithm is set to SOURCE_IP or QUIC_CID, this parameter will not take effect.

Range: 0 to 100. The default value is 1.

availability_zone

No

String

Definition: Specifies the AZ of the backend server.

Constraints:

  • This parameter can be only set for IP as backend servers. If AZ affinity is enabled for the backend server group, this parameter must be set to a valid value for IP as backend servers.

Range: AZs available for ECSs in the current region.

Response Parameters

Status code: 201

Table 5 Response body parameters

Parameter

Type

Description

request_id

String

Definition: Specifies the request ID.

Range: The value is automatically generated, and can contain characters including digits, lowercase letters, and hyphens (-).

member

Member object

Definition: Specifies the backend server.

Table 6 Member

Parameter

Type

Description

id

String

Definition: Specifies the backend server ID.

NOTE:
The value of this parameter is not the ID of the server but an ID automatically generated for the backend server that is associated with the load balancer.

availability_zone

String

Definition: Specifies the AZ where the backend server is running.

name

String

Definition: Specifies the backend server name.

Note: The name is not an ECS name.

project_id

String

Definition: Specifies the project ID of the backend server.

admin_state_up

Boolean

Definition: Specifies the administrative status of the backend server.

Constraints: Although this parameter can be used in the APIs for creating and updating backend servers, its actual value depends on whether ECSs exist. If ECSs exist, the value is true. Otherwise, the value is false.

Range: true or false

subnet_cidr_id

String

Definition: Specifies the ID of the IPv4 or IPv6 subnet where the backend server resides. neutron_subnet_id defines IPv4 subnets, and neutron_network_id defines IPv6 subnets.

You can query parameter neutron_subnet_id in the response by calling the API (GET https://{VPC_Endpoint}/v1/{project_id}/subnets) to get the IPv4 subnet ID.

You can query parameter neutron_network_id in the response by calling the API (GET https://{VPC_Endpoint}/v1/{project_id}/subnets) to get the IPv6 subnet ID.

Constraints:

  • The IPv4 or IPv6 subnet must be in the same VPC as the subnet of the load balancer.

  • If ip_target_enable is set to true, this parameter can be left blank, indicating that an IP as backend server will be added. In this case, IP as backend servers must use private IPv4 addresses, and the protocol of the backend server group must be TCP, UDP, TLS, HTTP, HTTPS, QUIC, or GRPC.

  • This parameter must be specified for gateway load balancers with IP backend server groups. The subnet of the backend server must be in the same VPC as the load balancer.

protocol_port

Integer

Definition: Specifies the port used by the backend server to receive requests.

Constraints: protocol_port must be set to 0 for gateway load balancers with IP backend server groups associated.

NOTE:
This parameter can be left blank because it does not take effect if any_port_enable is set to true for a backend server group.

weight

Integer

Definition: Specifies the weight of the backend server. Requests are routed based on the load balancing algorithm configured for the backend server group and weights of backend servers.

The larger the weight is, the higher proportion of requests the backend server receives. If the weight is set to 0, the backend server will not accept new requests.

Constraints: If lb_algorithm is set to SOURCE_IP or QUIC_CID, this parameter will not take effect.

Range: 0 to 100

Default value: 1

address

String

Definition: Specifies the IP address of the backend server.

Constraints:

  • If subnet_cidr_id is left blank, IP as backend servers will be added. In this case, the IP address must be an IPv4 address.

  • If subnet_cidr_id is not left blank, ECSs will be added as backend servers. It must be in the subnet specified by subnet_cidr_id.

ip_version

String

Definition: Specifies the IP address version supported by the backend server. The version depends on the value of address returned by the system.

Range: v4 or v6

operating_status

String

Definition: Specifies the operating status of the backend server.

Range:

  • NO_MONITOR: No health check is configured for the backend server group to which the backend server belongs.

  • INITIAL: The backend server is initializing. Health check is configured for the backend server group, but no data has been reported yet.

  • ONLINE: The backend server is running normally.

  • OFFLINE: The backend server is detected unhealthy, or the cloud server used as the backend server is stopped or does not exist.

  • UNKNOWN: The health check has not started yet, or no listener is associated with the backend server group, or the member IP address is not assigned to a cloud service instance (such as ECS, network interface, or supplementary network interface).

Multiple statuses can be used for query in the format of operating_status=xxx&operating_status=xxx.

status

Array of MemberStatus objects

Definition: Specifies the health status of the backend server.

Constraints:

  • If listener_id under status is specified, the status defined by operating_status is used.

  • If listener_id under status is not specified, the status defined by operating_status under member is used.

reason

MemberHealthCheckFailedReason object

Specifies why health check fails.

created_at

String

Definition: Specifies the creation time. The value must be a UTC time in the format of yyyy-MM-dd'T'HH:mm:ss'Z'.

updated_at

String

Definition: Specifies the update time. The value must be a UTC time in the format of yyyy-MM-dd'T'HH:mm:ss'Z'.

member_type

String

Definition: Specifies the backend server type.

Range:

  • ip: IP as backend servers

  • instance: ECSs used as backend servers

instance_id

String

Definition: Specifies the ID of the instance associated with the backend server. If this parameter is left blank, the backend server is not a real device. It may be an IP address.

Table 7 MemberStatus

Parameter

Type

Description

listener_id

String

Definition: Specifies the listener ID.

operating_status

String

Definition: Specifies the operating status of the backend server.

Range:

  • ONLINE: The backend server is running normally.

  • NO_MONITOR: No health check is configured for the backend server group to which the backend server belongs.

  • OFFLINE: The ECS used as the backend server is stopped or does not exist.

reason

MemberHealthCheckFailedReason object

Specifies why health check fails.

created_at

String

Specifies the time when the backend server group was created. The format is yyyy-MM-dd'T'HH:mm:ss'Z' (UTC time).

updated_at

String

Specifies the time when the backend server group was updated. The format is yyyy-MM-dd'T'HH:mm:ss'Z' (UTC time).

Table 8 MemberHealthCheckFailedReason

Parameter

Type

Description

reason_code

String

Definition: Specifies the code of the health check failures.

Range:

  • CONNECT_TIMEOUT: The connection with the backend server times out during a health check.

  • CONNECT_REFUSED: The load balancer rejects connections with the backend server during a health check.

  • CONNECT_FAILED: The load balancer fails to establish connections with the backend server during a health check.

  • CONNECT_INTERRUPT: The load balancer is disconnected from the backend server during a health check.

  • SSL_HANDSHAKE_ERROR: The SSL handshakes with the backend server fail during a health check.

  • RECV_RESPONSE_FAILED: The load balancer fails to receive responses from the backend server during a health check.

  • RECV_RESPONSE_TIMEOUT: The load balancer does not receive responses from the backend server within the timeout duration during a health check.

  • SEND_REQUEST_FAILED: The load balancer fails to send a health check request to the backend server during a health check.

  • SEND_REQUEST_TIMEOUT: The load balancer fails to send a health check request to the backend server within the timeout duration.

  • RESPONSE_FORMAT_ERROR: The load balancer receives invalid responses from the backend server during a health check.

  • RESPONSE_MISMATCH: The response code received from the backend server is different from the preset code.

expected_response

String

Definition: Specifies the expected HTTP status code. This parameter will take effect only when type is set to HTTP, HTTPS, or GRPC. The status code cannot be null if reason_code is RESPONSE_MISMATCH.

Range:

  • A specific status code. If type is set to GRPC, the status code ranges from 0 to 99. If type is set to other values, the status code ranges from 200 to 599. For example, the status code can be 0 or 200.

  • A list of status codes that are separated with commas (,), for example, 200,202 or 0,1. A maximum of five status codes are supported.

  • A status code range. Different ranges are separated with commas (,), for example, 200-204,300-399 or 0-5,10-12,20-30. A maximum of five ranges are supported.

healthcheck_response

String

Definition: Specifies the returned HTTP status code in the response. This parameter will take effect only when type is set to HTTP, HTTPS, or GRPC. The status code cannot be null if reason_code is RESPONSE_MISMATCH.

Range:

  • A specific status code. If type is set to GRPC, the status code ranges from 0 to 99. If type is set to other values, the status code ranges from 200 to 599. For example, the status code can be 0 or 200.

Example Requests

  • Example 1: Adding a backend server

    POST https://{ELB_Endpoint}/v3/99a3fff0d03c428eac3678da6a7d0f24/elb/pools/36ce7086-a496-4666-9064-5ba0e6840c75/members
    
    {
      "member" : {
        "subnet_cidr_id" : "c09f620e-3492-4429-ac15-445d5dd9ca74",
        "protocol_port" : 89,
        "name" : "My member",
        "address" : "120.10.10.16"
      }
    }
  • Example 2: Adding an IP address as a backend server

    POST https://{ELB_Endpoint}/v3/99a3fff0d03c428eac3678da6a7d0f24/elb/pools/36ce7086-a496-4666-9064-5ba0e6840c75/members
    
    {
      "member" : {
        "protocol_port" : 89,
        "name" : "My member",
        "address" : "120.10.10.16"
      }
    }
  • Example 3: Adding a backend server to an IP backend server group

    POST https://{ELB_Endpoint}/v3/99a3fff0d03c428eac3678da6a7d0f24/elb/pools/36ce7086-a496-4666-9064-5ba0e6840c75/members
    
    {
      "member" : {
        "protocol_port" : 0,
        "name" : "My IP pool member",
        "address" : "120.10.10.16"
      }
    }

Example Responses

Status code: 201

Successful request.

{
  "member" : {
    "name" : "My member",
    "weight" : 1,
    "admin_state_up" : false,
    "subnet_cidr_id" : "c09f620e-3492-4429-ac15-445d5dd9ca74",
    "project_id" : "99a3fff0d03c428eac3678da6a7d0f24",
    "address" : "120.10.10.16",
    "protocol_port" : 89,
    "id" : "1923923e-fe8a-484f-bdbc-e11559b1f48f",
    "operating_status" : "NO_MONITOR",
    "status" : [ {
      "listener_id" : "427eee03-b569-4d6c-b1f1-712032f7ec2d",
      "operating_status" : "NO_MONITOR"
    } ],
    "ip_version" : "v4"
  },
  "request_id" : "f354090d-41db-41e0-89c6-7a943ec50792"
}

Status Codes

Status Code

Description

201

Successful request.

Error Codes

See Error Codes.