首批通过分布式安全可靠测评,为关键业务系统打造
监控指标列表
更新时间:2026-04-14 11:55:54
本节介绍 OCP 有哪些监控指标及各指标的含义和计算方法。
| 指标分组 | 指标名称 | 指标解释 | 计算表达式 |
|---|---|---|---|
| QPS | sql_all_count | 每秒处理 SQL 语句数 | sum(rate(ob_sysstat{stat_id="40000",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40002",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40004",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40006",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40008",@LABELS}[@INTERVAL])) by (@GBLABELS) R_FILL_0_ADD sum(rate(ob_sysstat{stat_id="40018",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | sql_select_count | 每秒处理 Select 语句数 | sum(rate(ob_sysstat{stat_id="40000",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | sql_insert_count | 每秒处理 Insert 语句数 | sum(rate(ob_sysstat{stat_id="40002",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | sql_update_count | 每秒处理 Update 语句数 | sum(rate(ob_sysstat{stat_id="40006",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | sql_replace_count | 每秒处理 Replace 语句数 | sum(rate(ob_sysstat{stat_id="40004",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | sql_delete_count | 每秒处理 DDL、DCL、DTL 等其他语句数 | sum(rate(ob_sysstat{stat_id="40008",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | sql_other_count | 每秒处理 Delete 语句数 | sum(rate(ob_sysstat{stat_id="40018",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| SQL 响应时间 | sql_all_rt | 服务端每条 SQL 语句平均处理耗时 | (sum(rate(ob_sysstat{stat_id="40001",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40003",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40005",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40007",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40009",@LABELS}[@INTERVAL])) by (@GBLABELS) R_FILL_0_ADD sum(rate(ob_sysstat{stat_id="40019",@LABELS}[@INTERVAL])) by (@GBLABELS)) / (sum(rate(ob_sysstat{stat_id="40000",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40002",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40004",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40006",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40008",@LABELS}[@INTERVAL])) by (@GBLABELS) R_FILL_0_ADD sum(rate(ob_sysstat{stat_id="40018",@LABELS}[@INTERVAL])) by (@GBLABELS)) |
| SQL 响应时间 | sql_select_rt | 服务端每条 Select 语句平均处理耗时 | sum(rate(ob_sysstat{stat_id="40001",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="40000",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| SQL 响应时间 | sql_insert_rt | 服务端每条 Insert 语句平均处理耗时 | sum(rate(ob_sysstat{stat_id="40003",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="40002",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| SQL 响应时间 | sql_update_rt | 服务端每条 Update 语句平均处理耗时 | sum(rate(ob_sysstat{stat_id="40007",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="40006",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| SQL 响应时间 | sql_replace_rt | 服务端每条 Replace 语句平均处理耗时 | sum(rate(ob_sysstat{stat_id="40005",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="40004",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| SQL 响应时间 | sql_delete_rt | 服务端每条 Delete 语句平均处理耗时 | sum(rate(ob_sysstat{stat_id="40009",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="40008",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| SQL 响应时间 | sql_other_rt | 服务端处理每条 DDL、DCL、DTL 等其他语句的平均耗时 | sum(rate(ob_sysstat{stat_id="40019",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="40000",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| TPS | tps | 集群每秒处理事务数 | sum(rate(ob_sysstat{stat_id="30005",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| TPS | transaction_commit_count | 集群每秒提交事务数 | sum(rate(ob_sysstat{stat_id="30007",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| TPS | transaction_rollback_count | 集群每秒回滚事务数 | sum(rate(ob_sysstat{stat_id="30009",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| TPS | transaction_timeout_count | 集群每秒超时事务数 | sum(rate(ob_sysstat{stat_id="30011",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务响应时间 | tps_rt | 集群处理事务的平均耗时 | sum(rate(ob_sysstat{stat_id="30006",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="30005",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务响应时间 | transaction_commit_rt | 集群提交事务的平均耗时 | sum(rate(ob_sysstat{stat_id="30008",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="30007",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务响应时间 | transaction_rollback_rt | 集群回滚事务的平均耗时 | sum(rate(ob_sysstat{stat_id="30010",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="30009",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 请求等待队列耗时 | request_queue_time | SQL 请求在等待队列中等待耗时 | sum(rate(ob_sysstat{stat_id="20002",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="20001",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 会话数 | active_session | 集群当前活跃会话数 | sum(ob_active_session_num{@LABELS}) by (@GBLABELS) |
| 会话数 | all_session | 集群当前会话数 | sum(ob_all_session_num{@LABELS}) by (@GBLABELS) |
| 事务数量 | transaction_single_partition_count | 集群每秒处理普通事务次数 | sum(rate(ob_sysstat{stat_id="30012",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务数量 | transaction_multi_partition_count | 集群每秒处理分布式事务次数 | sum(rate(ob_sysstat{stat_id="30013",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务日志数 | trans_commit_log_count | 集群每秒提交的事务日志数 | sum(rate(ob_sysstat{stat_id="30002",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务日志量 | clog_trans_log_total_size | 集群每秒提交的事务日志大小 | sum(rate(ob_sysstat{stat_id="80057",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务日志耗时 | clog_write_disk_rt | 集群每次事务日志写盘的平均耗时 | sum(rate(ob_sysstat{stat_id="80041",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="80040",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务日志耗时 | trans_commit_log_sync_rt | 集群每次事务日志网络同步平均耗时 | sum(rate(ob_sysstat{stat_id="30000",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="30001",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 物理 IO 次数 | io_read_count | 集群每秒读 SSStore 次数 | sum(rate(ob_sysstat{stat_id="60000",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 物理 IO 次数 | io_write_count | 集群每秒写 SSStore 次数 | sum(rate(ob_sysstat{stat_id="60003",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 物理 IO 次数 | clog_io_read_count | 集群每秒读 clog 次数 | sum(rate(ob_sysstat{stat_id="80007",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 物理 IO 次数 | clog_io_write_count | 集群每秒写 clog 次数 | sum(rate(ob_sysstat{stat_id="80001",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 物理 IO 耗时 | io_read_rt | 集群每次读取 SSStore 的平均耗时 | sum(rate(ob_sysstat{stat_id="60001",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="60000",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 物理 IO 耗时 | io_write_rt | 集群每次写入 SSStore 的平均耗时 | sum(rate(ob_sysstat{stat_id="60004",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="60003",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 物理 IO 耗时 | clog_io_read_rt | 集群每次读取 clog 的平均耗时 | sum(rate(ob_sysstat{stat_id="80009",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="80007",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 物理 IO 耗时 | clog_io_write_rt | 集群每次写入 clog 的平均耗时 | sum(rate(ob_sysstat{stat_id="80003",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="80001",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 物理 IO 吞吐量 | io_read_size | 集群每秒从 SSStore 读取的数据量 | sum(rate(ob_sysstat{stat_id="60002",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 物理 IO 吞吐量 | io_write_size | 集群每秒向 SSStore 写入的数据量 | sum(rate(ob_sysstat{stat_id="60005",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 物理 IO 吞吐量 | clog_io_read_size | 集群每秒从 clog 读取的数据量 | sum(rate(ob_sysstat{stat_id="80008",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 物理 IO 吞吐量 | clog_io_write_size | 集群每秒向 clog 写入的数据量 | sum(rate(ob_sysstat{stat_id="80002",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 进程内存 | observer_rss_memory | 集群 OBServer 节点中进程所占用内存的大小 | sum(process_memory_rss_bytes{name="observer",@LABELS}) by (@GBLABELS) |
| 进程 CPU 使用率 | observer_process_cpu_percent | 集群 OBServer 节点中进程 CPU 的使用率 | avg(process_cpu_percent{name="observer",@LABELS}) by (@GBLABELS) |
| 进程文件句柄 | observer_fd_count | 集群 OBServer 节点的文件句柄数量 | avg(process_fd_count{name="observer",@LABELS}) by (@GBLABELS) |
| 进程文件句柄 | observer_fd_usage | 集群 OBServer 节点文件句柄使用率 | round(100 * avg(process_fd_count{name="observer",@LABELS}) by (@GBLABELS) / avg(observer_ulimit_max_fd_count{@LABELS}) by (@GBLABELS)) |
| 进程线程数 | observer_thread_count | 集群 OBServer 节点的进程线程数 | avg(process_thread_count{name="observer",@LABELS}) by (@GBLABELS) |
| 500 租户内存占用 | tenant500_mem_hold | 集群 500 租户内存占用大小(不包括 OB_KVSTORE_CACHE_MB) | sum(ob_tenant500_memory_hold_bytes{@LABELS}) by (@GBLABELS) / 1073741824 |
| 500 租户内存占用 | tenant500_mem_hold_percent | 集群 500 租户内存使用(不包括 OB_KVSTORE_CACHE_MB) / OBServer 内存资源的大小 | 100 * sum(ob_tenant500_memory_hold_bytes{@LABELS}) by (@GBLABELS) / sum(ob_server_resource_memory_bytes{@LABELS}) by (@GBLABELS) |
| 错误日志 | observer_log_error_count | 集群 OBServer 节点 ERROR 日志个数 | sum(rate(ob_log_level_count{level="ERROR",logSourceType="observer",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 错误日志 | election_log_error_count | 集群 OBServer 节点 election ERROR 日志个数 | sum(rate(ob_log_level_count{level="ERROR",logSourceType="election",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 错误日志 | rootservice_log_error_count | 集群 OBServer 节点 rootservice ERROR 日志个数 | sum(rate(ob_log_level_count{level="ERROR",logSourceType="rootservice",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | sql_all_count | 租户每秒处理 SQL 语句总次数 | sum(rate(ob_sysstat{stat_id="40000",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40002",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40004",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40006",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40008",@LABELS}[@INTERVAL])) by (@GBLABELS) R_FILL_0_ADD sum(rate(ob_sysstat{stat_id="40018",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | sql_select_count | 租户每秒处理 SELECT 语句次数 | sum(rate(ob_sysstat{stat_id="40000",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | sql_insert_count | 租户每秒处理 INSERT 语句次数 | sum(rate(ob_sysstat{stat_id="40002",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | sql_update_count | 租户每秒处理 UPDATE 语句次数 | sum(rate(sysstat_value{metric_group="sysstat",stat_id="40006",@LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) |
| QPS | sql_replace_count | 租户每秒处理 REPLACE 语句次数 | sum(rate(ob_sysstat{stat_id="40004",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | sql_delete_count | 每秒处理 DELETE 语句次数 | sum(rate(ob_sysstat{stat_id="40008",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | sql_other_count | 租户每秒处理 DDL、DCL、DTL 等其他语句次数 | sum(rate(ob_sysstat{stat_id="40018",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 响应时间 | sql_all_rt | 租户服务端处理每条 SQL 语句的平均耗时 | (sum(rate(ob_sysstat{stat_id="40001",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40003",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40005",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40007",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40009",@LABELS}[@INTERVAL])) by (@GBLABELS) R_FILL_0_ADD sum(rate(ob_sysstat{stat_id="40019",@LABELS}[@INTERVAL])) by (@GBLABELS)) / (sum(rate(ob_sysstat{stat_id="40000",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40002",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40004",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40006",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="40008",@LABELS}[@INTERVAL])) by (@GBLABELS) R_FILL_0_ADD sum(rate(ob_sysstat{stat_id="40018",@LABELS}[@INTERVAL])) by (@GBLABELS)) |
| 响应时间 | sql_select_rt | 租户服务端处理每条 select 语句的平均耗时 | sum(rate(ob_sysstat{stat_id="40001",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="40000",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 响应时间 | sql_insert_rt | 租户服务端处理每条 insert 语句的平均耗时 | sum(rate(ob_sysstat{stat_id="40003",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="40002",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 响应时间 | sql_update_rt | 租户服务端处理每条 update 语句的平均耗时 | sum(rate(ob_sysstat{stat_id="40007",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="40006",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 响应时间 | sql_replace_rt | 租户服务端处理每条 replace 语句的平均耗时 | sum(rate(ob_sysstat{stat_id="40005",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="40004",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 响应时间 | sql_delete_rt | 租户服务端处理每条 Delete 语句的平均耗时 | sum(rate(ob_sysstat{stat_id="40009",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="40008",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 响应时间 | sql_other_rt | 租户服务端处理每条 DDL、DCL、DTL 等其他语句的平均耗时 | sum(rate(ob_sysstat{stat_id="40019",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="40000",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| SQL 执行计划类别 | sql_local_count | 每秒处理本地执行数 | sum(rate(ob_sysstat{stat_id="40010",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| SQL 执行计划类别 | sql_remote_count | 每秒处理远程执行计划数 | sum(rate(ob_sysstat{stat_id="40011",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| SQL 执行计划类别 | sql_distributed_count | 每秒处理分布式执行计划数 | sum(rate(ob_sysstat{stat_id="40012",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| SQL 执行时间 | sql_local_time | 本地 SQL 的执行耗时 | sum(rate(ob_sysstat{stat_id="40116",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="40010",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| SQL 执行时间 | sql_remote_time | 远程 SQL 的执行耗时 | sum(rate(ob_sysstat{stat_id="40117",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="40011",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| SQL 执行时间 | sql_distributed_time | 分布式 SQL 的执行耗时 | sum(rate(ob_sysstat{stat_id="40118",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="40012",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 会话数 | active_session | 租户当前活跃会话数 | sum(ob_active_session_num{@LABELS}) by (@GBLABELS) |
| 会话数 | all_session | 租户当前会话总数 | sum(ob_all_session_num{@LABELS}) by (@GBLABELS) |
| 等待事件 | wait_event_count | 租户每秒发生的等待事件次数 | sum(rate(ob_waitevent_wait_total{@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件 | system_event_application_total_waits | 租户每秒发生的 application 等待事件次数 | sum(rate(ob_system_event_total_waits{event_group="CONFIGURATION",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件 | system_event_application_total_waits | 租户每秒发生的 configuration 事件次数 | sum(rate(ob_system_event_total_waits{event_group="CONFIGURATION",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件 | system_event_administrative_total_waits | 租户每秒发生的 administrative 等待事件次数 | sum(rate(ob_system_event_total_waits{event_group="ADMINISTRATIVE",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件 | system_event_concurrency_total_waits | 租户每秒发生的 concurrency 等待事件次数 | sum(rate(ob_system_event_total_waits{event_group="CONCURRENCY",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件 | system_event_commit_total_waits | 租户每秒发生的 commit 等待事件次数 | sum(rate(ob_system_event_total_waits{event_group="COMMIT",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件 | system_event_network_total_waits | 租户每秒发生的 network 等待事件次数 | sum(rate(ob_system_event_total_waits{event_group="NETWORK",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件 | system_event_user_io_total_waits | 租户每秒发生的 user_io 等待事件次数 | sum(rate(ob_system_event_total_waits{event_group="USER_IO",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件 | ystem_event_system_io_total_waits | 租户每秒发生的 system_io 等待事件次数 | sum(rate(ob_system_event_total_waits{event_group="SYSTEM_IO",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件 | system_event_cluster_total_waits | 租户每秒发生的 cluster 等待事件次数 | sum(rate(ob_system_event_total_waits{event_group="CLUSTER",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件 | system_event_other_total_waits | 租户每秒发生的其他等待事件次数 | sum(rate(ob_system_event_total_waits{event_group="OTHER",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件耗时 | wait_event_rt | 租户的等待事件平均等待耗时 | sum(rate(ob_waitevent_wait_seconds_total{@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_waitevent_wait_total{@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件耗时 | application_wait | 租户的 application 等待事件的平均耗时 | sum(rate(ob_system_event_time_waited{event_group="APPLICATION",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件耗时 | system_event_configuration_time_waited | 租户的 configuration 等待事件平均耗时 | sum(rate(ob_system_event_time_waited{event_group="CONFIGURATION",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件耗时 | system_event_administrative_time_waited | 租户的 administrative 等待事件平均耗时 | sum(rate(ob_system_event_time_waited{event_group="ADMINISTRATIVE",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件耗时 | system_event_concurrency_time_waited | 租户的 concurrency 等待事件平均耗时 | sum(rate(ob_system_event_time_waited{event_group="CONCURRENCY",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件耗时 | system_event_commit_time_waited | 租户的 commit 等待事件平均耗时 | sum(rate(ob_system_event_time_waited{event_group="COMMIT",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件耗时 | system_event_network_time_waited | 租户的 network 等待事件平均耗时 | sum(rate(ob_system_event_time_waited{event_group="NETWORK",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件耗时 | system_event_user_io_time_waited | 租户的 user_io 等待事件平均耗时 | sum(rate(ob_system_event_time_waited{event_group="USER_IO",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件耗时 | system_event_system_io_time_waited | 租户的 system_io 等待事件平均耗时 | sum(rate(ob_system_event_time_waited{event_group="SYSTEM_IO",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件耗时 | system_event_cluster_time_waited | 租户的 cluster 等待事件平均耗时 | sum(rate(ob_system_event_time_waited{event_group="CLUSTER",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 等待事件耗时 | system_event_other_time_waited | 租户其他等待事件的平均等待耗时 | sum(rate(ob_system_event_total_waits{event_group="OTHER",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 请求等待队列 | request_queue_count | 平均每秒进入等待队列的 SQL 数 | sum(rate(ob_sysstat{stat_id="20000",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 请求等待队列 | request_queue_count | SQL 在等待队列中平均等待耗时 | sum(rate(ob_sysstat{stat_id="20001",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 请求等待队列耗时 | request_queue_time | SQL 在等待队列中等待耗时 | sum(rate(ob_sysstat{stat_id="20002",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="20001",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 请求等待队列大小 | ob_tenant_request_queue_size | 租户请求队列长度 | sum(rate(ob_tenant_request_queue_total_size{@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 请求等待队列大小 | ob_tenant_request_queue_large_size | 租户大查询队列大小 | sum(rate(ob_tenant_request_queue_large_size{@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 租户 CPU 消耗 | ob_tenant_cpu_percent | 租户大查询队列大小 | sum(rate(ob_sysstat{stat_id="140013",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(ob_sysstat{stat_id="140005",@LABELS}) by (@GBLABELS) / 100 |
| 租户线程使用率 | ob_tenant_thread_percent | 活跃线程数 / 租户的最大线程数 | 100 * sum(ob_sysstat{stat_id="140006",@LABELS}) by (@GBLABELS) / sum(ob_sysstat{stat_id="140005",@LABELS}) by (@GBLABELS) |
| 内存使用率 | ob_memory_percent | 租户的内存使用率 | 100 * avg(ob_sysstat{stat_id="140003",@LABELS}) by (@GBLABELS) / avg(ob_sysstat{stat_id="140002",@LABELS}) by (@GBLABELS) |
| MEMStore 使用百分比 | Memstore_percent | 租户 MEMStore 的使用率 | 100 * sum(ob_sysstat{stat_id="130001",@LABELS}) by (@GBLABELS) / sum(ob_sysstat{stat_id="130004",@LABELS}) by (@GBLABELS) |
| MEMStore 使用百分比 | Active_memstore_percent | 租户 MEMStore 使用占触发冻结的百分比 | 100 * sum(ob_sysstat{stat_id="130000",@LABELS}) by (@GBLABELS) / sum(ob_sysstat{stat_id="130002",@LABELS}) by (@GBLABELS) |
| RPC 包耗时 | rpc_packet_in_rt | 租户收 RPC 包平均耗时 | (sum(rate(ob_sysstat{stat_id="10005",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="10006",@LABELS}[@INTERVAL])) by (@GBLABELS)) / sum(rate(ob_sysstat{stat_id="10000",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| RPC 包耗时 | rpc_packet_out_rt | 租户发 RPC 包平均耗时 | (sum(rate(ob_sysstat{stat_id="10005",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="10006",@LABELS}[@INTERVAL])) by (@GBLABELS)) / sum(rate(ob_sysstat{stat_id="10002",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| RPC 包吞吐量 | rpc_packet_in | 租户收 RPC 包的总数据量 | sum(rate(ob_sysstat{stat_id="10001",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| RPC 包吞吐量 | rpc_packet_out | 租户发 RPC 包的总数据量 | sum(rate(ob_sysstat{stat_id="10003",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 打开的游标数 | opened_cursors_count | 租户 session 缓存的游标数 | sum(delta(ob_sysstat{stat_id="40030",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| clog 同步延迟 | max_full_clog_sync_delay_seconds | 租户日志型副本 clog 同步最大延迟 | max(ob_clog_max_sync_delay_seconds{replica_type="0",@LABELS}) by (@GBLABELS) |
| clog 同步延迟 | max_readonly_clog_sync_delay_seconds | 租户全能型副本 clog 同步最大延迟 | max(ob_clog_max_sync_delay_seconds{replica_type="16",@LABELS}) by (@GBLABELS) |
| DB Time | db_cpu | 数据库实例执行数据库命令消耗的 ON CPU 时间,不包含后台任务 | sum(rate(ob_sysstat{stat_id="200002",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| DB Time | non_idle_wait_time | 数据库实例执行数据库命令消耗非空闲等待时间,不包含后台任务 | sum(rate(ob_sysstat{stat_id="200010",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| DB Time | background_db_cpu | 数据库后台任务消耗的 ON CPU 时间 | sum(rate(ob_sysstat{stat_id="200006",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| DB Time | background_non_idle_wait_time | 数据库后台任务消耗的非空闲等待时间 | sum(rate(ob_sysstat{stat_id="200013",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| SQL 执行阶段 | sql_event_in_parse | 正在执行 SQL PARSE 的活跃会话数 | sum(ob_sql_event_in_parse{@LABELS}) by (@GBLABELS) |
| SQL 执行阶段 | sql_event_in_pl_parse | 正在执行 PL PARSE 的活跃会话数 | sum(ob_sql_event_in_pl_parse{@LABELS}) by (@GBLABELS) |
| SQL 执行阶段 | sql_event_in_plan_cache | 正在访问 plan cache 的活跃会话数 | sum(ob_sql_event_in_plan_cache{@LABELS}) by (@GBLABELS) |
| SQL 执行阶段 | sql_event_in_sql_optimize | 正在执行 SQL 优化的活跃会话数 | sum(ob_sql_event_in_sql_optimize{@LABELS}) by (@GBLABELS) |
| SQL 执行阶段 | sql_event_in_sql_execution | 正在执行 SQL 语句的活跃会话数 | sum(ob_sql_event_in_sql_execution{@LABELS}) by (@GBLABELS) |
| SQL 执行阶段 | sql_event_in_px_execution | 正在进行 PX SQL 并行执行的活跃会话数 | sum(ob_sql_event_in_px_execution{@LABELS}) by (@GBLABELS) |
| SQL 执行阶段 | sql_event_in_sequence_load | 正在进行自增列或 SEQUENCE 取值的活跃会话数 | sum(ob_sql_event_in_sequence_load{@LABELS}) by (@GBLABELS) |
| SQL 执行阶段 | sql_event_in_committing | 正处于事务提交阶段的活跃会话数 | sum(ob_sql_event_in_committing{@LABELS}) by (@GBLABELS) |
| SQL 执行阶段 | sql_event_in_storage_read | 正在执行存储层读的活跃会话数 | sum(ob_sql_event_in_storage_read{@LABELS}) by (@GBLABELS) |
| SQL 执行阶段 | sql_event_in_storage_write | 正在执行存储层写的活跃会话数 | sum(ob_sql_event_in_storage_write{@LABELS}) by (@GBLABELS) |
| SQL 执行阶段 | sql_event_in_remote_das_execution | 正在远程 DAS 执行的活跃会话数 | sum(ob_sql_event_in_remote_das_execution{@LABELS}) by (@GBLABELS) |
| SQL 执行阶段 | sql_event_in_plsql_compilation | 租户处于 plsql compilation 状态的执行事件数 | sum(ob_sql_event_in_plsql_compilation{@LABELS}) by (@GBLABELS) |
| SQL 执行阶段 | sql_event_in_plsql_execution | 正在 PL 编译的活跃会话数 | sum(ob_sql_event_in_plsql_execution{@LABELS}) by (@GBLABELS) |
| SQL 执行阶段 | sql_event_in_plsql_execution | 正在 PL 执行的活跃会话数 | sum(ob_sql_event_in_plsql_execution{@LABELS}) by (@GBLABELS) |
| SQL 执行阶段 | sql_event_in_filter_rows | 正在执行算子下压的活跃会话数 | sum(ob_sql_event_in_filter_rows{@LABELS}) by (@GBLABELS) |
| 前台工作负载 | ob_foreground_on_cpu_waitevent_cnt | 占用 CPU 的执行数据库命令的活跃会话数 | sum(ob_foreground_waitevent_cnt{wait_class="ON_CPU",@LABELS}) by (@GBLABELS) |
| 前台工作负载 | ob_foreground_application_waitevent_cnt | 处于 APPLICATION 类型等待事件的活跃会话数 | sum(ob_foreground_waitevent_cnt{wait_class="APPLICATION",@LABELS}) by (@GBLABELS) |
| 前台工作负载 | ob_foreground_configuration_waitevent_cnt | 处于 CONFIGURATION 类型等待事件的活跃会话数 | sum(ob_foreground_waitevent_cnt{wait_class="CONFIGURATION",@LABELS}) by (@GBLABELS) |
| 前台工作负载 | ob_foreground_administrative_waitevent_cnt | 处于 ADMINISTRATIVE 类型等待事件的活跃会话数 | sum(ob_foreground_waitevent_cnt{wait_class="ADMINISTRATIVE",@LABELS}) by (@GBLABELS) |
| 前台工作负载 | ob_foreground_concurrency_waitevent_cnt | 处于 CONCURRENCY 类型等待事件的活跃会话数 | sum(ob_foreground_waitevent_cnt{wait_class="CONCURRENCY",@LABELS}) by (@GBLABELS) |
| 前台工作负载 | ob_foreground_commit_waitevent_cnt | 处于 COMMIT 类型等待事件的活跃会话数 | sum(ob_foreground_waitevent_cnt{wait_class="COMMIT",@LABELS}) by (@GBLABELS) |
| 前台工作负载 | ob_foreground_network_waitevent_cnt | 处于 NETWORK 类型等待事件的活跃会话数 | sum(ob_foreground_waitevent_cnt{wait_class="NETWORK",@LABELS}) by (@GBLABELS) |
| 前台工作负载 | ob_foreground_user_io_waitevent_cnt | 处于 USER_IO 类型等待事件的活跃会话数 | sum(ob_foreground_waitevent_cnt{wait_class="USER_IO",@LABELS}) by (@GBLABELS) |
| 前台工作负载 | ob_foreground_system_io_waitevent_cnt | 处于 SYSTEM_IO 类型等待事件的活跃会话数 | sum(ob_foreground_waitevent_cnt{wait_class="SYSTEM_IO",@LABELS}) by (@GBLABELS) |
| 前台工作负载 | ob_foreground_cluster_waitevent_cnt | 处于 CLUSTER 类型等待事件的活跃会话数 | sum(ob_foreground_waitevent_cnt{wait_class="CLUSTER",@LABELS}) by (@GBLABELS) |
| 前台工作负载 | ob_foreground_other_waitevent_cnt | 处于 OTHER 类型等待事件的活跃会话数 | sum(ob_foreground_waitevent_cnt{wait_class="OTHER",@LABELS}) by (@GBLABELS) |
| 后台工作负载 | ob_background_on_cpu_waitevent_cnt | 占用 CPU 的执行数据库命令的活跃会话数 | sum(ob_background_waitevent_cnt{wait_class="ON_CPU",@LABELS}) by (@GBLABELS) |
| 后台工作负载 | ob_background_application_waitevent_cnt | 处于 APPLICATION 类型等待事件的活跃会话数 | sum(ob_background_waitevent_cnt{wait_class="APPLICATION",@LABELS}) by (@GBLABELS) |
| 后台工作负载 | ob_background_configuration_waitevent_cnt | CONFIGURATION 类型等待事件的活跃会话数 | sum(ob_background_waitevent_cnt{wait_class="CONFIGURATION",@LABELS}) by (@GBLABELS) |
| 后台工作负载 | ob_background_administrative_waitevent_cnt | 处于 ADMINISTRATIVE 类型等待事件的活跃会话数 | sum(ob_background_waitevent_cnt{wait_class="ADMINISTRATIVE",@LABELS}) by (@GBLABELS) |
| 后台工作负载 | ob_background_concurrency_waitevent_cnt | 处于 CONCURRENCY 类型等待事件的活跃会话数 | sum(ob_background_waitevent_cnt{wait_class="CONCURRENCY",@LABELS}) by (@GBLABELS) |
| 后台工作负载 | ob_background_commit_waitevent_cnt | 处于 COMMIT 类型等待事件的活跃会话数 | sum(ob_background_waitevent_cnt{wait_class="COMMIT",@LABELS}) by (@GBLABELS) |
| 后台工作负载 | ob_background_network_waitevent_cnt | 处于 NETWORK 类型等待事件的活跃会话数 | sum(ob_background_waitevent_cnt{wait_class="NETWORK",@LABELS}) by (@GBLABELS) |
| 后台工作负载 | ob_background_user_io_waitevent_cnt | 处于 USER_IO 类型等待事件的活跃会话数 | sum(ob_background_waitevent_cnt{wait_class="USER_IO",@LABELS}) by (@GBLABELS) |
| 后台工作负载 | ob_background_system_io_waitevent_cnt | 处于 SYSTEM_IO 类型等待事件的活跃会话数 | sum(ob_background_waitevent_cnt{wait_class="SYSTEM_IO",@LABELS}) by (@GBLABELS) |
| 后台工作负载 | ob_background_cluster_waitevent_cnt | 处于 CLUSTER 类型等待事件的活跃会话数 | sum(ob_background_waitevent_cnt{wait_class="CLUSTER",@LABELS}) by (@GBLABELS) |
| 后台工作负载 | ob_background_other_waitevent_cnt | 处于 OTHER 类型等待事件的活跃会话数 | sum(ob_background_waitevent_cnt{wait_class="OTHER",@LABELS}) by (@GBLABELS) |
| TPS | tps | 租户每秒处理总事务数 | sum(rate(ob_sysstat{stat_id="30005",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| TPS | transaction_commit_count | 租户每秒提交事务数 | sum(rate(ob_sysstat{stat_id="30007",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| TPS | transaction_rollback_count | 租户每秒回滚事务数 | sum(rate(ob_sysstat{stat_id="30009",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| TPS | transaction_timeout_count | 租户每秒超时事务数 | sum(rate(ob_sysstat{stat_id="30011",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务响应时间 | trans_sp_total_used_time | 本地事务响应时间 | sum(rate(ob_sysstat{stat_id="30080",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="30012",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务响应时间 | trans_distributed_total_used_time | 租户处理所有事务的平均耗时 | sum(rate(ob_sysstat{stat_id="30081",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="30013",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务响应时间明细 | tps_rt | 分布式事务响应时间 | sum(rate(ob_sysstat{stat_id="30006",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="30005",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务响应时间明细 | transaction_commit_rt | 租户成功提交的事务的平均处理耗时 | sum(rate(ob_sysstat{stat_id="30008",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="30007",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务响应时间明细 | transaction_rollback_rt | 租户回滚事务的平均处理耗时 | sum(rate(ob_sysstat{stat_id="30010",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="30009",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务日志数 | trans_commit_log_count | 每秒提交的事务日志数 | sum(rate(ob_sysstat{stat_id="30002",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务日志量 | clog_trans_log_total_size | 每秒提交的事务日志大小 | sum(rate(ob_sysstat{stat_id="80057",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务日志耗时 | trans_commit_log_sync_rt | 每次事务日志网络同步平均耗时 | sum(rate(ob_sysstat{stat_id="30000",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="30001",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务日志耗时 | clog_write_disk_rt | 每次事务日志写盘的平均耗时 | sum(rate(ob_sysstat{stat_id="80041",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="80040",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务锁等待次数 | memstore_write_lock_succ_count | 写锁等待成功次数 | sum(rate(ob_sysstat{stat_id="60021",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务锁等待次数 | memstore_write_lock_fail_count | 写锁等待失败次数 | ssum(rate(ob_sysstat{stat_id="60022",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务锁平均等待时间 | memstore_write_lock_wait_time | 事务锁平均等待耗时 | sum(rate(ob_sysstat{stat_id="60023",@LABELS}[@INTERVAL])) by (@GBLABELS) / (sum(rate(ob_sysstat{stat_id="60021",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(rate(ob_sysstat{stat_id="60022",@LABELS}[@INTERVAL])) by (@GBLABELS)) |
| 事务数量 | transaction_single_partition_count | 租户每秒处理普通事务次数 | sum(rate(ob_sysstat{stat_id="30012",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务数量 | transaction_multi_partition_count | 租户每秒处理分布式事务次数 | sum(rate(ob_sysstat{stat_id="30013",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务表读请求命中数 | tx_data_hit_mini_cache_count | 事务表读请求命中 mini cache 数量 | sum(rate(ob_sysstat{stat_id="30082",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务表读请求命中数 | tx_data_hit_kv_cache_count | 事务表读请求命中 mini cache 数量 | sum(rate(ob_sysstat{stat_id="30083",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务表读请求命中数 | tx_data_read_tx_ctx_count | 事务表读请求命中事务上下文表数量 | sum(rate(ob_sysstat{stat_id="30084",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务表读请求命中数 | tx_data_read_tx_data_memtable_count | 事务表读请求命中 memtable 数量 | sum(rate(ob_sysstat{stat_id="30085",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 事务表读请求命中数 | tx_data_read_tx_data_sstable_count | 事务表读请求命中 sstable 数量 | sum(rate(ob_sysstat{stat_id="30086",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 分区 | tenant_partition_leader_absent_count | 租户无主分区个数量 | max(partition_leader_absent_count{@LABELS}) by (@GBLABELS) |
| 分区 | tenant_partition_replica_absent_count | 租户缺副本分区个数 | max(partition_replica_absent_count{@LABELS}) by (@GBLABELS) |
| XA 事务数 | xa_trans_start_count | 租户中已经成功开启的 XA 事务总个数(包含已经结束的) | sum(rate(ob_sysstat{stat_id="30220",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| XA 事务数 | xa_read_only_trans_total_count | 租户中只读的 XA 事务总个数 | sum(rate(ob_sysstat{stat_id="30221",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| XA 事务数 | xa_one_phase_commit_total_count | 租户中采用一阶段提交的 XA 事务总个数 | sum(rate(ob_sysstat{stat_id="30222",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| XA 语句数 | xa_start_total_count | 租户中 xa_start 语句个数 | sum(rate(ob_sysstat{stat_id="30200",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| XA 语句数 | xa_end_total_count | 租户中 xa_end 语句个数 | sum(rate(ob_sysstat{stat_id="30204",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| XA 语句数 | xa_prepare_total_count | 租户中 xa_prepare 语句个数 | sum(rate(ob_sysstat{stat_id="30208",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| XA 语句数 | xa_commit_total_count | 租户中 xa_commit 语句个数 | sum(rate(ob_sysstat{stat_id="30212",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| XA 语句数 | xa_rollback_total_count | 租户中 xa_rollback 语句个数 | sum(rate(ob_sysstat{stat_id="30216",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| XA 语句耗时 | xa_start_total_rt | 租户 xa_start 语句耗时 | sum(rate(ob_sysstat{stat_id="30201",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="30200",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| XA 语句耗时 | xa_end_total_rt | 租户 xa_end 语句耗时 | sum(rate(ob_sysstat{stat_id="30205",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="30204",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| XA 语句耗时 | xa_prepare_total_rt | 租户 xa_prepare 语句耗时 | sum(rate(ob_sysstat{stat_id="30209",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="30208",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| XA 语句耗时 | xa_commit_total_rt | 租户 xa_commit 语句耗时 | sum(rate(ob_sysstat{stat_id="30213",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="30212",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| XA 语句耗时 | xa_rollback_total_rt | 租户 xa_rollback 语句耗时 | sum(rate(ob_sysstat{stat_id="30217",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="30216",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| MEMStore | total_memstore_used | 为租户准备的 MEMStore 的大小 | sum(ob_sysstat{stat_id="130001",@LABELS}) by (@GBLABELS) / 1048576 |
| MEMStore | active_memstore_used | 活跃 MemStore 大小 | sum(ob_sysstat{stat_id="130000",@LABELS}) by (@GBLABELS) / 1048576 |
| MEMStore | major_freeze_trigger | 触发合并的 MEMStore 大小 | sum(ob_sysstat{stat_id="130002",@LABELS}) by (@GBLABELS) / 1048576 |
| MEMStore | memstore_limit | MEMStore 的 最大使用限制 | sum(ob_sysstat{stat_id="130004",@LABELS}) by (@GBLABELS) / 1048576 |
| 物理 IO 次数 | io_read_count | 租户每秒读 SSStore 次数 | sum(rate(ob_sysstat{stat_id="60000",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 物理 IO 次数 | io_write_count | 租户每秒写 SSStore 次数 | sum(rate(ob_sysstat{stat_id="60003",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 物理 IO 次数 | clog_io_read_count | 租户每秒读 clog 次数 | sum(rate(ob_sysstat{stat_id="80007",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 物理 IO 次数 | clog_io_write_count | 租户每秒写 clog 次数 | sum(rate(ob_sysstat{stat_id="80001",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 物理 IO 耗时 | io_read_rt | 租户每次读取 SSStore 的平均耗时 | sum(rate(ob_sysstat{stat_id="60001",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="60000",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 物理 IO 耗时 | io_write_rt | 租户每次写入 SSStore 的平均耗时 | sum(rate(ob_sysstat{stat_id="60004",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="60003",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 物理 IO 耗时 | clog_io_read_rt | 租户每次读取 clog 的平均耗时 | sum(rate(ob_sysstat{stat_id="80009",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="80007",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 物理 IO 耗时 | clog_io_write_rt | 租户每次写入 clog 的平均耗时 | sum(rate(ob_sysstat{stat_id="80003",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="80001",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| IO 吞吐量 | io_read_size | 租户每秒从 SSStore 读取的数据量 | sum(rate(ob_sysstat{stat_id="60002",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| IO 吞吐量 | io_write_size | 租户每秒向 SSStore 写入的数据量 | sum(rate(ob_sysstat{stat_id="60005",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| IO 吞吐量 | clog_io_read_size | 租户每秒从 clog 读取的数据量 | sum(rate(ob_sysstat{stat_id="80008",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| IO 吞吐量 | clog_io_write_size | 租户每秒向 clog 写入的数据量 | sum(rate(ob_sysstat{stat_id="80002",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 缓存大小 | block_cache_size | 块缓存大小 | sum(ob_cache_size_bytes{cache_name="user_block_cache",@LABELS}) by (@GBLABELS) |
| 缓存大小 | row_cache_size | 行缓存大小 | sum(ob_sysstat{stat_id="120008",@LABELS}) by (@GBLABELS) |
| 缓存大小 | plan_cache_size | 执行计划缓存大小 | sum(ob_plan_cache_memory_bytes{@LABELS}) by (@GBLABELS) |
| 缓存大小 | clog_cache_size | 提交日志缓存大小 | sum(ob_sysstat{stat_id="120001",@LABELS}) by (@GBLABELS) |
| 缓存大小 | bloom_filter_cache_size | bloomfilter 缓存大小 | sum(ob_sysstat{stat_id="120009",@LABELS}) by (@GBLABELS) |
| 缓存大小 | location_cache_size | location 缓存大小 | sum(ob_sysstat{stat_id="120000",@LABELS}) by (@GBLABELS) |
| 缓存命中率 | block_cache_hit_ratio | 块缓存命中率 | 100 * 1 / (1 + sum(rate(ob_sysstat{stat_id="50009",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="50008",@LABELS}[@INTERVAL])) by (@GBLABELS)) |
| 缓存命中率 | row_cache_hit_ratio | 行缓存命中率 | 100 * 1 / (1 + sum(rate(ob_sysstat{stat_id="50001",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="50000",@LABELS}[@INTERVAL])) by (@GBLABELS)) |
| 缓存命中率 | plan_cache_hit_ratio | 执行计划缓存命中率 | 100 * sum(rate(ob_plan_cache_hit_total{@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_plan_cache_access_total{@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 缓存命中率 | clog_cache_hit_ratio | 提交日志缓存命中率 | 100 * 1 / (1 + sum(rate(ob_sysstat{stat_id="50038",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="50037",@LABELS}[@INTERVAL])) by (@GBLABELS)) |
| 缓存命中率 | bloom_filter_cache_hit_ratio | bloomfilter 缓存命中率 | 100 * 1 / (1 + sum(rate(ob_sysstat{stat_id="50005",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="50004",@LABELS}[@INTERVAL])) by (@GBLABELS)) |
| 缓存命中率 | location_cache_hit_ratio | location 缓存命中率 | 100 * 1 / (1 + sum(rate(ob_sysstat{stat_id="50011",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(ob_sysstat{stat_id="50010",@LABELS}[@INTERVAL])) by (@GBLABELS)) |
| 缓存请求次数 | block_cache_req_total | 块缓存请求次数 | sum(delta(ob_sysstat{stat_id="50008",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(delta(ob_sysstat{stat_id="50009",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 缓存请求次数 | row_cache_req_total | 行缓存请求次数 | sum(delta(ob_sysstat{stat_id="50000",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(delta(ob_sysstat{stat_id="50001",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 缓存请求次数 | clog_cache_req_total | 提交日志缓存请求次数 | sum(delta(ob_sysstat{stat_id="50037",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(delta(ob_sysstat{stat_id="50038",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 缓存请求次数 | bloom_filter_cache_req_total | bloomfilter 缓存请求次数 | sum(delta(ob_sysstat{stat_id="50004",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(delta(ob_sysstat{stat_id="50005",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 缓存请求次数 | location_cache_req_total | location 缓存请求次数 | sum(delta(ob_sysstat{stat_id="50010",@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(delta(ob_sysstat{stat_id="50011",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 冻结 MEMStore | ob_tenant_partition_frozen_memstore_count | 租户冻结 MEMStore 的个数 | max(ob_partition_frozen_memstore_count{@LABELS}) by (@GBLABELS) |
| TPS | transaction_total | OBProxy 集群平均每秒处理事务数 | sum(rate(odp_transaction_total{@LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | request_total_commit | 每秒处理 SQL 语句数 | sum(rate(odp_sql_request_total{sqltype="COMMIT", @LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | request_total_delete | OBProxy 集群平均每秒转发的 delete 请求数 | sum(rate(odp_sql_request_total{sqltype="DELETE", @LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | request_total_insert | OBProxy 集群平均每秒转发的 insert 请求数 | sum(rate(odp_sql_request_total{sqltype="INSERT", @LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | request_total_large | OBProxy 集群平均每秒转发的大请求数 | sum(rate(odp_sql_request_total{sqlLarge = "large", @LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | request_total_select | OBProxy 集群平均每秒转发的 select 请求数 | sum(rate(odp_sql_request_total{sqltype="SELECT", @LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | request_total_sequence | OBProxy 集群平均每秒转发的 sequence 请求数 | sum(rate(odp_sql_request_total{sqltype="SEQUENCE", @LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | request_total_slow | OBProxy 集群平均每秒转发的慢请求数 | sum(rate(odp_sql_request_total{slowQuery = "true", @LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | request_total_update | OBProxy 集群平均每秒转发的 update 请求数 | sum(rate(odp_sql_request_total{sqltype="UPDATE", @LABELS}[@INTERVAL])) by (@GBLABELS) |
| QPS | request_total_others | OBProxy 集群平均每秒转发的其他请求数 | sum(rate(odp_sql_request_total{sqltype="OTHERS", @LABELS}[@INTERVAL])) by (@GBLABELS) |
| 客户端连接数 | current_session_client | OBProxy 集群当前时间点与客户端建立的连接数 | sum(odp_current_session{sessionType="client",@LABELS}) by (@GBLABELS) |
| 客户端连接数 | client_connection_used_percent | OBProxy 客户端连接数 / client_max_connections | 100 * sum(odp_current_session{sessionType="client",@LABELS}) by (@GBLABELS) / sum(obproxy_client_max_connections{@LABELS}) by (@GBLABELS) |
| 服务端连接数 | current_session_server | OBProxy 集群当前时间点与服务端建立的连接数 | sum(odp_current_session{sessionType="server",@LABELS}) by (@GBLABELS) |
| 服务端连接数 | server_connection_used_percent | OBProxy 服务端连接数 / max_connections | 100 * sum(odp_current_session{sessionType="server",@LABELS}) by (@GBLABELS) / sum(obproxy_server_max_connections{@LABELS}) by (@GBLABELS) |
| 平均每条 SQL 响应时间 | request_cost_commit | 平均每次 commit 请求耗时 | sum(rate(odp_sql_cost_total{timeType="total",sqltype="COMMIT", @LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(odp_sql_request_total{sqltype="COMMIT", @LABELS}[@INTERVAL])) by (@GBLABELS) |
| 平均每条 SQL 响应时间 | request_cost_delete | 平均每次 delete 请求耗时 | sum(rate(odp_sql_cost_total{timeType="total",sqltype="DELETE", @LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(odp_sql_request_total{sqltype="DELETE", @LABELS}[@INTERVAL])) by (@GBLABELS) |
| 平均每条 SQL 响应时间 | request_cost_insert | 平均每次 insert 请求耗时 | sum(rate(odp_sql_cost_total{timeType="total",sqltype="INSERT", @LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(odp_sql_request_total{sqltype="INSERT", @LABELS}[@INTERVAL])) by (@GBLABELS) |
| 平均每条 SQL 响应时间 | request_cost_prepare_send_request_time | 平均每次请求发送准备耗时 | sum(rate(odp_sql_cost_total{timeType="prepare", @LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(odp_sql_request_total{@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 平均每条 SQL 响应时间 | request_cost_server_process_request_time | 平均每次服务端处理请求耗时 | sum(rate(odp_sql_cost_total{timeType="server", @LABELS}[@INTERVAL])) by (@GBLABELS) / sum(rate(odp_sql_request_total{@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 平均每条 SQL 响应时间 | request_cost_select | 平均每次 select 请求耗时 | sum(rate(odp_sql_cost_total{timeType="total",sqltype="SELECT", @LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / sum(rate(odp_sql_request_total{sqltype="SELECT", @LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) |
| 平均每条 SQL 响应时间 | request_cost_sequence | 平均每次 sequence 请求耗时 | sum(rate(odp_sql_cost_total{timeType="total",sqltype="SEQUENCE", @LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / sum(rate(odp_sql_request_total{sqltype="SEQUENCE", @LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) |
| 平均每条 SQL 响应时间 | request_cost_update | 平均每次 update 请求耗时 | sum(rate(odp_sql_cost_total{timeType="total",sqltype="UPDATE", @LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / sum(rate(odp_sql_request_total{sqltype="UPDATE", @LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) |
| 平均每条 SQL 响应时间 | request_cost_others | 平均每次其他请求耗时 | sum(rate(odp_sql_cost_total{sqltype="OTHERS", @LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) / sum(rate(odp_sql_request_total{sqltype="OTHERS", @LABELS}[@INTERVAL]) by (@GBLABELS)) by (@GBLABELS) |
| OBProxy 内存 | obproxy_rss_memory | OBProxy 进程常驻内存大小 | sum(process_memory_rss_bytes{name="obproxy",@LABELS}) by (@GBLABELS) |
| OBProxy 内存 | obproxy_rss_mem_used_percent | OBProxy 进程常驻内存 / proxy_mem_limited | 100 * sum(process_memory_rss_bytes{name="obproxy",@LABELS}) by (@GBLABELS) / sum(obproxy_memory_limit_bytes{@LABELS}) by (@GBLABELS) |
| OBProxy 进程 CPU 使用率 | obproxy_process_cpu_percent | OBProxy 进程 CPU 使用率 | avg(process_cpu_percent{name="obproxy",@LABELS}) by (@GBLABELS) |
| OBProxy 文件句柄数 | obproxy_fd_count | OBProxy 文件句柄数 | avg(process_fd_count{name="obproxy",@LABELS}) by (@GBLABELS) |
| OBProxy 线程数 | obproxy_thread_count | OBProxy 进程线程数 | avg(process_thread_count{name="obproxy",@LABELS}) by (@GBLABELS) |
| 错误日志个数 | obproxy_log_error_count | OBProxy 错误日志个数 | sum(rate(obproxy_log_level_count{level="ERROR",logSourceType="obproxy_error",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 平均每秒路由表查询次数 | entry_total_hit | 平均每秒路由表命中次数 | sum(rate(odp_entry_total{routeHit="true",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 平均每秒路由表查询次数 | entry_total_partition | 平均每秒查询 partition entry 的次数 | sum(rate(odp_entry_total{entryType="partition_entry",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 平均每秒路由表查询次数 | entry_total_route | 平均每秒查询 route entry 的次数 | sum(rate(odp_entry_total{entryType="route",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 平均每秒路由表查询次数 | entry_total_success | 平均每秒路由表查询成功次数 | sum(rate(odp_entry_total{routeResult="true",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 平均每秒路由表查询次数 | entry_total_table | 平均每秒查询 table entry 的次数 | sum(rate(odp_entry_total{entryType="table_entry",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 平均每次网络请求或响应字节数 | request_byte_client_request | 客户端 request 字节数 | sum(rate(odp_request_byte{sessionType="client",transType="request",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 平均每次网络请求或响应字节数 | request_byte_client_response | 客户端 reponse 字节数 | sum(rate(odp_request_byte{sessionType="client",transType="response",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 平均每次网络请求或响应字节数 | request_byte_server_request | 服务端 request 字节数 | sum(rate(odp_request_byte{sessionType="server",transType="request",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 平均每次网络请求或响应字节数 | request_byte_server_response | 服务端 response 字节数 | sum(rate(odp_request_byte{sessionType="server",transType="response",@LABELS}[@INTERVAL])) by (@GBLABELS) |
| Linux 系统负载 | load1 | 过去 1 分钟系统平均负载 | avg(node_load1{@LABELS}) by (@GBLABELS) |
| Linux 系统负载 | load5 | 过去 5 分钟系统平均负载 | avg(node_load5{@LABELS}) by (@GBLABELS) |
| Linux 系统负载 | load15 | 过去 15 分钟系统平均负载 | avg(node_load15{@LABELS}) by (@GBLABELS) |
| Linux 系统负载 | load1_per_cpu | 过去 1 分钟系统平均负载/CPU 核数 | sum(node_load1{@LABELS}) by (@GBLABELS) / sum(cpu_count{@LABELS}) by (@GBLABELS) |
| CPU 使用率 | ob_cpu_percent | 租户的 CPU 使用率 | 100 * (1 - sum(rate(node_cpu_seconds_total{mode=~"idle|iowait|steal",@LABELS}[@INTERVAL])) by (@GBLABELS) / sum(cpu_count{@LABELS}) by (@GBLABELS)) |
| IO 使用率 | io_util | IO 使用率 | avg(io_util{@LABELS}) by (@GBLABELS) |
| IO 队列长度 | io_qusize | IO 队列长度 | sum(delta(node_disk_io_time_weighted_seconds_total{@LABELS}[@INTERVAL])) by (@GBLABELS) |
| IOPS | io_read | 每秒读盘次数 | avg(rate(node_disk_reads_completed_total{@LABELS}[@INTERVAL])) by (@GBLABELS) |
| IOPS | io_write | 每秒写盘次数 | avg(rate(node_disk_writes_completed_total{@LABELS}[@INTERVAL])) by (@GBLABELS) |
| IO 耗时 | io_read_time | 平均每次读取耗时 | 1000000 * (avg(rate(node_disk_read_time_seconds_total{@LABELS}[@INTERVAL])) by (@GBLABELS)) / (avg(rate(node_disk_reads_completed_total{@LABELS}[@INTERVAL])) by (@GBLABELS)) |
| IO 耗时 | io_read_time | 平均每次读取耗时 | 1000000 * (avg(rate(node_disk_read_time_seconds_total{@LABELS}[@INTERVAL])) by (@GBLABELS)) / (avg(rate(node_disk_reads_completed_total{@LABELS}[@INTERVAL])) by (@GBLABELS)) |
| IO 吞吐量 | io_read_byte | 平均每秒从磁盘读取的数据量 | avg(rate(node_disk_read_bytes_total{@LABELS}[@INTERVAL])) by (@GBLABELS) / 1048576 |
| IO 吞吐量 | io_write_byte | 平均每秒向磁盘写入的数据量 | avg(rate(node_disk_written_bytes_total{@LABELS}[@INTERVAL])) by (@GBLABELS) / 1048576 |
| 网络吞吐率 | net_recv | 每秒接收数据量 | avg(rate(node_network_receive_bytes_total{device!="lo",@LABELS}[@INTERVAL])) by (@GBLABELS) / 1048576 |
| 网络吞吐率 | net_send | 每秒发送数据量 | avg(rate(node_network_transmit_bytes_total{device!="lo",@LABELS}[@INTERVAL])) by (@GBLABELS) / 1048576 |
| TCP 重传率 | tcp_retrans_percent | 主机 TCP 的重传率 | max(tcp_retrans{@LABELS}) by (@GBLABELS) |
| 网络收发包 | net_receive_drop_count | 服务器网络丢包次数 | sum(rate(node_network_receive_drop_total{@LABELS}[@INTERVAL])) by (@GBLABELS) |
| 网络收发包 | net_traffic_err_count | 服务器网络收发包出错次数 | sum(delta(node_network_receive_errs_total{@LABELS}[@INTERVAL])) by (@GBLABELS) + sum(delta(node_network_transmit_errs_total{@LABELS}[@INTERVAL])) by (@GBLABELS) |
| NTP 时钟偏移 | ntp_offset_milliseconds | 服务器 NTP 时钟偏移数据 | max(abs(node_ntp_offset_seconds{@LABELS})) by (@GBLABELS) * 1000 |
| Inode 使用率 | file_inode_usage | 服务器 inode 使用率 | round(100 - 100 * sum(node_filesystem_files_free{@LABELS}) by (@GBLABELS) / sum(node_filesystem_files{@LABELS}) by (@GBLABELS)) |
| 内存 | memory_used | 主机中已使用物理内存的大小 | (avg(node_memory_MemTotal_bytes{@LABELS}) by (@GBLABELS) - avg(node_memory_MemFree_bytes{@LABELS}) by (@GBLABELS) - avg(node_memory_Cached_bytes{@LABELS}) by (@GBLABELS) - avg(node_memory_Buffers_bytes{@LABELS}) by (@GBLABELS)) / 1073741824 |
| 内存 | memory_buffers | 主机中内核 Buffer Cache 的大小 | avg(node_memory_Buffers_bytes{@LABELS}) by (@GBLABELS) / 1073741824 |
| 内存 | memory_free | 主机中空闲可用物理内存的大小 | avg(node_memory_MemFree_bytes{@LABELS}) by (@GBLABELS) / 1073741824 |
| 内存 | memory_cached | 主机中 cached 的大小 | avg(node_memory_Cached_bytes{@LABELS}) by (@GBLABELS) / 1073741824 |
| 内存使用率 | memory_percent | 主机中已使用物理内存的使用率 | (1 - (avg(node_memory_MemFree_bytes{@LABELS}) by (@GBLABELS) + avg(node_memory_Cached_bytes{@LABELS}) by (@GBLABELS) + avg(node_memory_Buffers_bytes{@LABELS}) by (@GBLABELS) + avg(node_memory_SReclaimable_bytes{@LABELS}) by (@GBLABELS)) / avg(node_memory_MemTotal_bytes{@LABELS}) by (@GBLABELS)) * 100 |
| 磁盘 | host_disk_used | 磁盘分区已使用容量 | sum(node_filesystem_size_bytes{@LABELS} - node_filesystem_avail_bytes{@LABELS}) by (@GBLABELS) / 1073741824 |
| 进程常驻内存 | observer_rss_memory | observer 进程常驻内存大小 | sum(process_memory_rss_bytes{name="observer",@LABELS}) by (@GBLABELS) |
| 进程常驻内存 | obproxy_rss_memory | obproxy 进程常驻内存大小 | sum(process_memory_rss_bytes{name="obproxy",@LABELS}) by (@GBLABELS) |
| 进程常驻内存 | monagent_resident_memory | monagent 进程常驻内存大小 | sum(process_memory_rss_bytes{name="ocp_monagent",@LABELS}) by (@GBLABELS) |
| 进程常驻内存 | mgragent_resident_memory | mgragent 进程常驻内存大小 | sum(process_memory_rss_bytes{name="ocp_mgragent",@LABELS}) by (@GBLABELS) |
| 进程 CPU 使用率 | observer_process_cpu_percent | obproxy 进程 CPU 使用率 | avg(process_cpu_percent{name="observer",@LABELS}) by (@GBLABELS) |
| 进程 CPU 使用率 | obproxy_process_cpu_percent | observer 进程 CPU 使用率 | avg(process_cpu_percent{name="obproxy",@LABELS}) by (@GBLABELS) |
| 进程 CPU 使用率 | monagent_process_cpu_percent | OCP 监控 Agent 进程 CPU 使用率 | avg(process_cpu_percent{name="ocp_monagent",@LABELS}) by (@GBLABELS) |
| 进程 CPU 使用率 | mgragent_process_cpu_percent | OCP 管理 Agent 进程 CPU 使用率 | avg(process_cpu_percent{name="ocp_mgragent",@LABELS}) by (@GBLABELS) |
| 进程文件句柄数 | observer_fd_count | OBServer 文件句柄数 | avg(process_fd_count{name="observer",@LABELS}) by (@GBLABELS) |
| 进程文件句柄数 | oobproxy_fd_count | monagent 进程文件句柄数 | avg(process_fd_count{name="obproxy",@LABELS}) by (@GBLABELS) |
| 进程文件句柄数 | monagent_open_fd_count | monagent 进程文件句柄数 | avg(process_fd_count{name="ocp_monagent",@LABELS}) by (@GBLABELS) |
| 进程文件句柄数 | mgragent_open_fd_count | mgragent 进程文件句柄数 | avg(process_fd_count{name="ocp_mgragent",@LABELS}) by (@GBLABELS) |
| 进程线程数 | observer_thread_count | observer 进程的线程数量 | avg(process_thread_count{name="observer",@LABELS}) by (@GBLABELS) |
| 进程线程数 | obproxy_thread_count | OBProxy 进程线程数量 | avg(process_thread_count{name="obproxy",@LABELS}) by (@GBLABELS) |
| 进程线程数 | monagent_thread_count | monagent 进程线程数量 | avg(process_thread_count{name="ocp_monagent",@LABELS}) by (@GBLABELS) |
| 进程线程数 | mgragent_thread_count | mgragent 进程线程数量 | avg(process_thread_count{name="ocp_mgragent",@LABELS}) by (@GBLABELS) |
| OCP-Agent 进程协程数 | monagent_goroutine_count | monagent 进程 GO 协程数量 | max(go_goroutines{process="ocp_monagent",@LABELS}) by (@GBLABELS) |
| OCP-Agent 进程协程数 | mgragent_goroutine_count | mgragent 进程 GO 协程数量 | max(go_goroutines{process="ocp_mgragent",@LABELS}) by (@GBLABELS) |