Obtaining Service Details

Function

This API is used to obtain the details about a model service based on the service ID.

URI

GET /v1/{project_id}/services/{service_id}

**Table 1** Path Parameters
Parameter	Mandatory	Type	Description
project_id	Yes	String	Project ID. For details, see Obtaining a Project ID and Name.
service_id	Yes	String	Service ID

Request Parameters

**Table 2** Request header parameters
Parameter	Mandatory	Type	Description
X-Auth-Token	Yes	String	User token. It can be obtained by calling the IAM API that is used to obtain a user token. The value of X-Subject-Token in the response header is the user token.

Response Parameters

Status code: 200

**Table 3** Response body parameters
Parameter	Type	Description
service_id	String	Service ID
service_name	String	Service name
description	String	Service description
tenant	String	Tenant to which a service belongs
project	String	Project to which a service belongs
owner	String	User to which a service belongs
publish_at	Number	Latest service release time, in milliseconds calculated from 1970.1.1 0:0:0 UTC.
infer_type	String	Inference mode. Options: real-time batch
workspace_id	String	Workspace ID
cluster_id	String	ID of the dedicated resource pool used by the real-time or batch service. This parameter is available only when a dedicated resource pool is used.
vpc_id	String	ID of the VPC to which the real-time service instance belongs. This parameter is available when the network configuration is customized.
subnet_network_id	String	ID of the subnet to which the real-time service instance belongs. This parameter is available when the network configuration is customized.
security_group_id	String	Security group to which the real-time service instance belongs. This parameter is available when the network configuration is customized.
status	String	Service status. The options are as follows: running: The service is running properly. deploying: The service is being deployed, including image creation and resource scheduling deployment. concerning: An alarm has been generated, indicating that some backend instances malfunction. failed: Deploying the service failed. For details about the failure cause, see the event and log tab pages. stopped: The service has been stopped. finished: Service running is completed. This status is available only for batch services. stopping: The VM is being stopped. deleting: The VM is being deleted. pending: to be started. This state is available only when the system is online. waiting: The resource is queuing. This state is available only for online services.
progress	Integer	Deployment progress. This parameter is available when the status is deploying.
error_msg	String	Error message. When status is failed, an error message carrying the failure cause is returned.
config	Array of QueryServiceConfig objects	Service configuration (If a service is shared, only model_id, model_name, and model_version are returned.)
access_address	String	Access address of an inference request. This parameter is available when infer_type is set to real-time.
bind_access_address	String	Request address of a custom domain name. This parameter is available after a domain name is bound.
invocation_times	Number	Total number of service calls
failed_times	Number	Number of failed service calls
is_shared	Boolean	Whether a service is subscribed
shared_count	Number	Number of subscribed services
schedule	Array of Schedule objects	Service scheduling configuration. If this parameter is not configured, no value will be returned.
update_time	Number	Time when the configuration used by the current service is updated, in milliseconds calculated from 1970.1.1 0:0:0 UTC.
debug_url	String	Online debugging address of a real-time service. This parameter is available only when the model supports online debugging and there is only one instance.
due_time	Number	Time when an online service automatically stops, in milliseconds calculated from 1970.1.1 0:0:0 UTC. If automatic stop is not configured, this parameter is not returned.
operation_time	Number	Operation time of a request
transition_at	Number	Time when the service status changes
is_free	Boolean	Whether a free-of-charge flavor is used
additional_properties	Map<String,String>	Additional service attribute
pool_name	String	Specifies the ID of the new dedicated resource pool. By default, this parameter is left blank, indicating that the dedicated resource pool is not used. This parameter corresponds to the ID of the new resource pool.

**Table 4** QueryServiceConfig
Parameter	Type	Description
model_version	String	Model version
finished_time	Number	Task completion time, in milliseconds calculated from 1970.1.1 0:0:0 UTC. This parameter is not returned before the task is complete.
custom_spec	CustomSpec object	Custom resource specifications
envs	Map<String,String>	Environment variable key-value pair required for running a model
specification	String	Resource flavor, which can be modelarts.vm.cpu.2u, modelarts.vm.gpu.p4, or modelarts.vm.ai1.a310/custom
weight	Integer	Traffic weight allocated to a model
source_type	String	Model source. This parameter is returned when a model is created using ExeML. The value is auto.
model_id	String	Model ID
src_path	String	OBS path to the input data of a batch job
req_uri	String	Inference path of a batch job
mapping_type	String	Mapping type of the input data, which can be file or csv
start_time	Number	Task start time, in milliseconds calculated from 1970.1.1 0:0:0 UTC. This parameter is not returned before the task starts.
cluster_id	String	ID of a dedicated resource pool used by a service instance. This parameter is returned only when a dedicated resource pool is configured.
nodes	Array of Nodes objects	Node information
mapping_rule	Object	Mapping between input parameters and CSV data. This parameter is mandatory only when mapping_type is set to csv.
model_name	String	Model name
src_type	String	Data source type. This parameter is returned only when ManifestFile is used.
dest_path	String	OBS path to the output data of a batch job
instance_count	Integer	Number of instances deployed for a model
status	String	Service status. Options: running: The service is running properly. deploying: The service is being deployed, including image creation and resource scheduling deployment. concerning: An alarm has been generated, indicating that some backend instances malfunction. failed: Deploying the service failed. For details about the failure cause, see the event and log tab pages. stopped: The service has been stopped. finished: Service running is completed. This status is available only for batch services.
scaling	Boolean	Whether auto scaling is enabled
support_debug	Boolean	Whether a model supports online debugging
additional_properties	Map<String,String>	Additional model deployment attribute
pool_name	String	Specifies the ID of the new dedicated resource pool. By default, this parameter is left blank, indicating that the dedicated resource pool is not used. This parameter corresponds to the ID of the new resource pool.

**Table 5** CustomSpec
Parameter	Type	Description
gpu_p4	Float	Number of GPUs, which can be a decimal. The value cannot be smaller than 0, with the third decimal place is rounded off. This parameter is optional and is not used by default.
memory	Integer	Memory in MB, which must be an integer
cpu	Float	Number of CPU cores, which can be a decimal. The value cannot be smaller than 0.01, with the third decimal place is rounded off.
ascend_a310	Integer	Number of Ascend chips. This parameter is optional and is not used by default. Either this parameter or gpu_p4 is configured.

**Table 6** Nodes
Parameter	Type	Description
memory	Integer	Memory size, in MB
os_version	String	OS version of a node
cpu	Integer	Number of CPU cores
created_at	String	Creation time, in the format of YYYY-MM-DDThh:mm:ss (UTC)
description	String	Description
message	String	Cause if instance_status is failed or notReady
predict_url	String	Inference URL of a node
enable_gpu	Boolean	Whether to enable GPUs
gpu_num	Integer	Number of GPUs
host_ips	Array of strings	Host IP address of a node
updated_at	String	Update time, in the format of YYYY-MM-DDThh:mm:ss (UTC)
node_label	String	Node label
os_type	String	OS type of a node
name	String	Name of an edge node
os_name	String	OS name of a node
arch	String	Node architecture
id	String	Edge node ID
instance_status	String	Status of a model instance on a node, which can be running, stopped, notReady, or failed
state	String	Host status, which can be RUNNING, FAIL, or UNCONNECTED
deployment_num	Integer	Number of application instances deployed on a node
host_name	String	Host name of a node

**Table 7** Schedule
Parameter	Type	Description
duration	Integer	Value mapping a time unit. For example, if the task stops after two hours, set time_unit to HOURS and duration to 2.
time_unit	String	Scheduling time unit. Possible values are DAYS, HOURS, and MINUTES.
type	String	Scheduling type. Only the value stop is supported.

Example Requests

GET https://{endpoint}/v1/{project_id}/services/{service_id}

Example Responses

Status code: 200

Service Details

{
  "service_id" : "f76f20ba-78f5-44e8-893a-37c8c600c02f",
  "service_name" : "service-demo",
  "tenant" : "xxxxx",
  "project" : "xxxxx",
  "owner" : "xxxxx",
  "publish_at" : 1585809231902,
  "update_time" : 1585809358259,
  "infer_type" : "real-time",
  "status" : "running",
  "progress" : 100,
  "access_address" : "https://xxxxx.apigw.xxxxx.com/v1/infers/088458d9-5755-4110-97d8-1d21065ea10b/f76f20ba-78f5-44e8-893a-37c8c600c02f",
  "cluster_id" : "088458d9-5755-4110-97d8-1d21065ea10b",
  "workspace_id" : "0",
  "additional_properties" : { },
  "is_shared" : false,
  "invocation_times" : 0,
  "failed_times" : 0,
  "shared_count" : 0,
  "operation_time" : 1586249085447,
  "config" : [ {
    "model_id" : "044ebf3d-8bf4-48df-bf40-bad0e664c1e2",
    "model_name" : "jar-model",
    "model_version" : "1.0.1",
    "specification" : "custom",
    "custom_spec" : { },
    "status" : "notReady",
    "weight" : 100,
    "instance_count" : 1,
    "scaling" : false,
    "envs" : { },
    "additional_properties" : { },
    "support_debug" : false
  } ],
  "transition_at" : 1585809231902,
  "is_free" : false
}