使用指导

通过巡检相关API实现功能调用

智能巡检接口

示例：

curl -X 'POST' "http://127.0.0.1:8080/v1/api/app/real-time-inspection?inspection_type=real_time_check&start_time=1689210000000&end_time=1689296400000&instance=127.0.0.1:5432" -H 'accept: application/json' -H 'Content-Type: application/json' -d '{"system_resource": ["os_mem_usage"], "instance_status": [], "database_resource": [], "database_performance": [], "diagnosis_optimization": []}' -H "Authorization: bearer xxx"

如果使用HTTPS协议，则查询示例：

curl -X 'POST' "https://127.0.0.1:8080/v1/api/app/real-time-inspection?inspection_type=real_time_check&start_time=1689210000000&end_time=1689296400000&instance=127.0.0.1:5432" -H 'accept: application/json' -H 'Content-Type: application/json' -d '{"system_resource": ["os_mem_usage"], "instance_status": [], "database_resource": [], "database_performance": [], "diagnosis_optimization": []}' -H "Authorization: bearer xxx" --cacert xx.crt --key xx.key --cert xx.crt

如果使用自定义阈值，查询示例：

curl -X 'POST' "https://127.0.0.1:8080/v1/api/app/real-time-inspection?inspection_type=real_time_check&start_time=1689210000000&end_time=1689296400000&instance=127.0.0.1:5432" -H 'accept: application/json' -H 'Content-Type: application/json' -d '{"system_resource": [{"os_mem_usage": {"increase": false, "threshold": [], "forecast": [1440, 0.0, 0.8]}}], "instance_status": [], "database_resource": [], "database_performance": [], "diagnosis_optimization": []}' -H "Authorization: bearer xxx" --cacert xx.crt --key xx.key --cert xx.crt

返回结构示例如下：

{
 "data": {
  "conclusion": {
   "full_score": 0.06,
   "health_score": 0.06,
   "health_status": "bad",
   "top3": []
  },
  "database_performance": {},
  "database_resource": {},
  "diagnosis_optimization": {},
		"instance_status": {},
		"system_resource": {
			"os_mem_usage": {
				"127.0.0.1": {
					"data": [0.31643905373281667],
     "statistic": {
      "avg": 0.311,
      "max": 0.3166,
      "min": 0.3057,
      "the_95th": 0.3153
     },
     "timestamps": [1694674713000],
     "warnings": {
      "increase_warning": true
     }
    }
   }
  }
 },
 "success": true
}

展示巡检任务接口示例：

curl -X 'GET' "http://127.0.0.1:8080/v1/api/app/real-time-inspection/list?instance=127.0.0.1:5432" -H 'accept: application/json' -H "Authorization: bearer xxx"

如果使用HTTPS协议，则查询示例：

curl -X 'GET' "https://127.0.0.1:8080/v1/api/app/real-time-inspection/list?instance=127.0.0.1:5432" -H 'accept: application/json' -H "Authorization: bearer xxx" --cacert xx.crt --key xx.key --cert xx.crt

返回结构如下：

{"data":{"header":["instance","start","end","id","state","cost_time","inspection_type"],"rows":[["127.0.0.1:5432",1689210000000,1689296400000,5,"success",0.033701,"real_time_check"]]},"success":true}

获取指定巡检任务的巡检结果接口示例：

curl -X 'GET' "http://127.0.0.1:8080/v1/api/summary/real-time-inspection?spec_id=5&instance=127.0.0.1:5432" -H 'accept: application/json' -H "Authorization: bearer xxx"

如果使用HTTPS协议，则查询示例：

curl -X 'GET' "https://127.0.0.1:8080/v1/api/summary/real-time-inspection?spec_id=5&instance=127.0.0.1:5432" -H 'accept: application/json' -H "Authorization: bearer xxx" --cacert xx.crt --key xx.key --cert xx.crt

返回结构与智能巡检接口返回结构一致。

删除指定的巡检任务接口示例：

curl -X 'DELETE' "http://127.0.0.1:8080/v1/api/app/real-time-inspection? spec_id=5&instance=127.0.0.1:5432" -H 'accept: application/json' -H "Authorization: bearer xxx"

如果使用HTTPS协议，则删除示例：

curl -X 'DELETE' "https://127.0.0.1:8080/v1/api/app/real-time-inspection? spec_id=5&instance=127.0.0.1:5432" -H 'accept: application/json' -H "Authorization: bearer xxx" --cacert xx.crt --key xx.key --cert xx.crt

返回结构如下：

{"data":{"success":true},"success":true}

自定义阈值参数

自定义阈值传参方式，各告警类型对应的key如表1所示：

表 1 告警类型以及对应的key

告警类型	键
持续上升	increase
超过阈值	threshold
预测超过阈值	forecast
文件类型有误	ftype

启用默认告警配置（505.1.0之前的版本）

{
"system_resource": [
"os_mem_usage"
],
"instance_status": [],
"database_resource": [],
"database_performance": [],
"diagnosis_optimization": []
}

启用默认告警配置（505.1.0及之后版本）

{
"system_resource": [
{
"os_mem_usage": true
}
],
"instance_status": [],
"database_resource": [],
"database_performance": [],
"diagnosis_optimization": []
}

不启用告警

{
"system_resource": [
{
"os_mem_usage": false
}
],
"instance_status": [],
"database_resource": [],
"database_performance": [],
"diagnosis_optimization": []
}

启用自定义告警。如下所示，os_mem_usage启用持续上升告警、自定义设置阈值告警和预测阈值告警；os_disk_usage不启用持续上升告警、阈值告警、启用预测阈值告警。

{
"system_resource": [
{
"os_mem_usage": {
"increase": true,
"threshold": [0.0, 0.8],
"forecast": [1440, 0.0, 0.8]
},
"os_disk_usage": {
"forecast": [1440, 0.0, 0.8]
}
}
],
"instance_status": [],
"database_resource": [],
"database_performance": [],
"diagnosis_optimization": []
}

部分巡检项不支持自定义配置阈值：组件异常、日志异常检查、数据库Top Query、长事务、oldestXmin长时间未推进、Core dump、GUC参数。
各个巡检项支持的阈值类型见概述中自定义阈值表格，前端传入不支持的告警类型会报错。

对于部分巡检项，存在子巡检项，如表2所示：

表 2 巡检项及子巡检项

巡检项	子巡检项
os_cpu_usage	cpu_user
os_cpu_usage	cpu_iowait
user_login_out	login
user_login_out	logout
db_latency	p95
db_latency	p80
db_transaction	commit
db_transaction	rollback
db_exec_statement	select
	update
	insert
	delete
db_tps	tps
db_tps	qps
dynamic_memory	dynamic_used_memory
dynamic_memory	dynamic_used_shrctx

当子巡检项的告警配置一致时，可以省略重复值，具体方式如下：

重复设置os_cpu_usage阈值自定义

{
"system_resource": [
{
"os_cpu_usage": {
"cpu_user": {
"increase": false,
"threshold": [],
"forecast": [1440, 0.0, 0.8]
},
"cpu_iowait": {
"increase": false,
"threshold": [],
"forecast": [1440, 0.0, 0.8]
}
}
}
],
"instance_status": [],
"database_resource": [],
"database_performance": [],
"diagnosis_optimization": []
}

简化设置os_cpu_usage阈值自定义

{
"system_resource": [
{
"os_cpu_usage": {
"increase": false,
"threshold": [],
"forecast": [1440, 0.0, 0.8]
}
}
],
"instance_status": [],
"database_resource": [],
"database_performance": [],
"diagnosis_optimization": []
}

巡检项结果返回值示例

数据库实例
以1主DN、2备DN为例，共3个节点：
主DN：127.0.0.1:19996
备DN：127.0.0.2:19996, 127.0.0.3:19996
返回结构类型
- ①：以节点为key：{“127.0.0.1”: xxx, “127.0.0.2”: xxx, “127.0.0.3”: xxx}
- ②：以DN为key：{“127.0.0.1:19996”: xxx, “127.0.0.2:19996”: xxx, “127.0.0.3:19996”: xxx}
- ③：以DB为key：{“db1”: xxx, “db2”: xxx, “db3”: xxx}
- ④：返回list: [{xxx}]

巡检项对应的返回结构如表3所示。

表 3 巡检项返回结构

巡检项	返回结构	备注
os_cpu_usage	①	两层结构，子巡检项：cpu_user、cpu_iowait。
os_disk_ioutils	①	-
os_disk_usage	①	-
os_mem_usage	①	-
network_packet_loss	①	两层结构，展示节点到节点的网络状况。
component_error	①	-
data_directory	②	-
log_directory	②	-
db_size	③	-
buffer_hit_rate	③	-
user_login_out	②	两层结构，子巡检项：login、logout。
active_session_rate	②	-
log_error_check	①	-
thread_pool	②	-
db_latency	②	两层结构，子巡检项：p80、p95。
db_transaction	③	两层结构，子巡检项：commit、rollback。
db_tmp_file	③	-
db_exec_statement	②	两层结构，子巡检项：select、update、insert、delete。
db_deadlock	③	-
db_tps	②	两层结构，子巡检项：qps、tps。
db_top_query	④	-
long_transaction	④	-
xmin_stuck	②	-
xlog_accumulate	①	-
core_dump	①	-
dynamic_memory	②	两层结构，子巡检项：dynamic_used_memory、dynamic_used_shrctx。
process_memory	②	-
other_memory	②	-
guc_params	②	-

说明： 巡检结果会存储到DBMind元数据库中，DBMind会定期清除老数据以避免磁盘膨胀。