Simple Log Service:GetMLServiceResults

This API endpoint is available only in the China (Shanghai) and Singapore regions. The following basic permissions are required:

{
    "Version": "1",
    "Statement": [
        {
            "Action": [
                "log:Get*"
            ],
            "Resource": [
                "acs:log:*:*:mlservice/sls_builtin_*"
            ],
            "Effect": "Allow"
        }
    ]
}

The request parameters vary depending on the service. The details are as follows:

Named Entity Recognition for logs (sls_builtin_service_log_struct)

Details of body.parameter

{
  "is_struct": "true",
  "use_gpu": "true/false",
  "max_fields": "1"
}

• If is_struct is true, the log is in JSON format. Currently, you can only set this parameter to true.

• use_gpu specifies whether to use GPU resources.

• max_fields specifies the maximum number of fields to analyze when the log is in JSON format. By default, one field is analyzed.

Details of body.input

[
  {
    "{column_name}": "{column_value}"
  }
]

The input is an array of JSON objects, where:

• column_name is the name of a field in the JSON-formatted log.

• column_value is the value of the field in the JSON-formatted log.

Time series anomaly detection (sls_builtin_service_metric_anomaly)

Details of body.parameter

{
  "isOrdered": "true/false",
  "keys": "[]",
  "timeColumnName": "",
  "analysisColumnNames": "[]",
  "interval": "-1"
}

• isOrdered specifies whether the input data is ordered. Set this parameter to true if the input time series is sorted in ascending order by time. Otherwise, set it to false.

• interval specifies the interval of the time series in seconds. For example, if the series has one data point per minute, set this parameter to 60. If you are unsure of the interval, set this parameter to -1.

• timeColumnName specifies the name of the time dimension in the input series.

• analysisColumnNames specifies the names of the numerical feature dimensions to be analyzed in the input series. This is an array that must be serialized into a string.

• keys specifies the names of all columns in the input data. This is an array that must be serialized into a string.

Details of body.input

[
  {
    "{column_name}": "{column_value}"
  }
]

The input is an array, where:

• column_name is the name of a field in the series.

• column_value is the value of the field. If {column_name} is the same as {timeColumnName}, the value is a UNIX timestamp in seconds.

High-latency trace detection (sls_builtin_service_trace_rca)

Details of body.parameter

{
  "project": "",
  "logstore": "",
  "endpoint": "",
  "role_arn": ""
}

• project specifies the name of the project in SLS where the trace data is stored.

• logstore specifies the name of the Logstore in SLS where the trace data is stored.

• endpoint specifies the endpoint of the region where the project is located. You must use an Internet endpoint. Internal endpoints in the same region cannot be accessed.

• role_arn specifies the ARN of the role that is granted permissions to the resource. You must use the ARN of an ETLRole or AuditRole.

Details of body.input

[
  {
    "service": "",
    "name": "",
    "from_ts_sec": "",
    "to_ts_sec": "",
    "slo_ts_micro": "",
    "batch_id": "",
    "trace_ids": "[]"
  }
]

The input is an array with a length of 1. The parameters are described as follows:

• service specifies the service name for one or more TraceIDs to be diagnosed. You can retrieve this from the Logstore.

• name specifies the service name for one or more TraceIDs to be diagnosed. You can retrieve this from the Logstore.

• from_ts_sec specifies the start timestamp for the analysis, in seconds.

• to_ts_sec specifies the end timestamp for the analysis, in seconds.

• batch_id specifies the name for the current batch of TraceIDs to be analyzed.

• trace_ids specifies the list of trace_ids for the current request. This must be a JSON-serialized string.

The return value varies depending on the service. The details are as follows:

Named Entity Recognition for logs (sls_builtin_service_log_struct)

Details of data

[
  {
    "{column_name}": "{column_value}"
  }
]

Each JSON object in the output corresponds to a JSON object in the input log, where:

• column_name is the name of an analyzed field in the JSON-formatted log. A maximum of max_fields fields are analyzed.

• column_value is the value of the analyzed field. The NER result in the field value is enclosed in <ml_ner_${ner_type}></ml_ner_${ner_type}> tags, where ${ner_type} is the specific NER type.

Time series anomaly detection (sls_builtin_service_metric_anomaly)

Details of data

[
  {
    "start": "",
    "end": "",
    "label": ""
  }
]

• start specifies the start time of the detected anomaly interval.

• end specifies the end time of the detected anomaly interval.

• label specifies the type of anomaly detected in the current interval.

  • SPIKE_UP_TYPE

  • SPIKE_DOWN_TYPE

  • TREND_UP_TYPE

  • TREND_DOWN_TYPE

  • MEANSHIFT_UPWARD_TYPE

  • MEANSHIFT_DOWNWARD_TYPE

High-latency trace detection (sls_builtin_service_trace_rca)

Details of data

[
  {
    "traceID": "",
    "service": "",
    "name": "",
    "rootCauses": "[{}]"
  }
]

• traceID specifies the ID of the trace that was analyzed.

• service specifies the service name of the root node of the trace.

• name specifies the name of the root node of the trace.

• rootCauses specifies the list of spans that caused the high latency in the trace. This is a serialized string. When deserialized, the structure of this field is as follows:

[
  {
    "spanID": "",
    "service": "",
    "name": "",
    "host": "xxxx",
    "predicateDuration": 10
  }
]