API for Collecting Dynamic Status v2

Function

Collects dynamic status.

Format

Operation type: GET

URL: https://{ip}:{port}/v2/status

The {ip} field prioritizes reading the environment variable MIES_CONTAINER_MANAGEMENT_IP. If the environment variable does not exist, managementIpAddress in the configuration file is used. If managementIpAddress does not exist in the configuration file, ipAddress in the configuration file is used.
The {port} field prioritizes reading the managementPort parameter in the configuration file. If the configuration file does not contain the managementPort parameter, the port parameter in the configuration file is used.

Request Parameters

None

Usage Example

Request example:

GET https://{ip}:{port}/v2/status

Response example:

{
    "service": {
        "roleStatus": "RoleUnknown",
        "currentRole": "prefill"
    },
    "resource": {
        "totalAvailNpuSlotsNum": 100, 
        "totalAvailNpuBlockNum": 400, 
        "maxAvailNpuBlockNum": 100,
    },
    "linkStatus": { 
         "peers": [ 
             { 
                 "target": 20000,
                 "link": "ok" 
            }, 
             { 
                 "target": 20001,
                 "link": "ok" 
             } 
         ] 
     } 
 }

Output Description

Field	Mandatory/Optional	Type	Description
service	Mandatory	Structure	Service attributes.
service.roleStatus	Mandatory	String	Role status of the current service. Possible values: RoleUnknown RoleSwitching RoleReady In the prefill-decode hybrid deployment scenario, the value is always RoleReady.
service.currentRole	Mandatory	String	Role of the current instance. Possible values: prefill decode none In the prefill-decode hybrid deployment scenario, the value is always none. In the prefill-decode disaggregation scenario, if roleStatus is RoleReady, the value is prefill or decode. If roleStatus is not RoleReady, the value is none.
resource	Mandatory	Structure	Remaining resources.
resource.totalAvailNpuSlotsNum	Mandatory	Integer	Number of available slots in all DP groups.
resource.totalAvailNpuBlockNum	Mandatory	Integer	Number of available blocks in all DP groups.
resource.maxAvailNpuBlockNum	Mandatory	Integer	Number of available blocks in the DP group with the highest count.
peers	Optional	String array	The array member is the IP address, indicating the prefill instance that has been successfully connected. This field does not need to be returned in the prefill-decode hybrid deployment scenario or for prefill instances in the prefill-decode disaggregation scenario. This field needs to be returned for decode instances in the prefill-decode disaggregation scenario. Each time a decode instance successfully connects to a prefill instance, a member is added. If the target prefill instance remains unchanged and the pull kv operation on a prefill instance fails and cannot be recovered, the prefill instance will be deleted.
linkStatus.peers	Mandatory	Structure array	Connection status of the peer node of the current prefill or decode instance.
linkStatus.peers[].target	Mandatory	Integer	Instance ID.
linkStatus.peers[].status	Mandatory	String	Link status of the peer instance. The status of a peer node is set to ok only when the link of all DP groups under the peer node is successfully connected.

Parent topic: Prefill-Decode Disaggregation APIs