命名空间
Namespace=QCE/TSTREAM
监控指标
指标英文名 | 指标中文名 | 指标说明 | 单位 | 维度 | 统计规则
[period, statType] |
JobRestartingtime | 作业重启耗时 | 作业最近一次重启耗时 | ms | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
JobmanagerLastcheckpointrestoretimestamp | 作业最近一次恢复的时间戳 | 作业最近一次从快照恢复的 Unix 时间戳（以毫秒为单位，如果未恢复过则是 -1） | ms | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
TaskmanagerJvmThreadsCount | TaskManager 活动线程数 | 作业中所有 TaskManager 中活动的线程数之和，含 Daemon 和非 Daemon 线程。 | 个 | tjob_id | [60s, sum]
[300s, avg]
[3600s, avg]
[86400s, avg] |
TaskmanagerMemoryNonheapCommitted | TaskManager 已提交的非堆内存容量 | 作业中所有 TaskManager 已提交（committed）的非堆内存（JVM 元空间、代码缓存等）用量总和 | Bytes | tjob_id | [60s, sum]
[300s, avg]
[3600s, avg]
[86400s, avg] |
TaskmanagerMemoryHeapCommitted | TaskManager 已提交的堆内存容量 | 作业中所有 TaskManager 已提交（committed）的堆内存容量总和 | Bytes | tjob_id | [60s, sum]
[300s, avg]
[3600s, avg]
[86400s, avg] |
TaskmanagerNetworkTotalmemorysegments | TaskManager 已分配的 MemorySegment 总数 | 作业中所有 TaskManager 已分配的 MemorySegment 个数总和 | Bytes | tjob_id | [60s, sum]
[300s, avg]
[3600s, avg]
[86400s, avg] |
TaskmanagerJvmYoungGcTime | TaskManager 年轻代 GC 时间 | 作业中所有 TaskManager 年轻代 GC 时间之和 | ms | tjob_id | [60s, sum]
[300s, last]
[3600s, last]
[86400s, last] |
TaskmanagerJvmYoungGcCount | TaskManager 年轻代 GC 次数 | 作业中所有 TaskManager 年轻代 GC 次数之和 | 次 | tjob_id | [60s, sum]
[300s, last]
[3600s, last]
[86400s, last] |
TaskmanagerJvmOldGcTime | TaskManager 老年代 GC 时间 | 作业中所有 TaskManager 老年代 GC 时间之和 | ms | tjob_id | [60s, sum]
[300s, last]
[3600s, last]
[86400s, last] |
TaskmanagerJvmOldGcCount | TaskManager 老年代 GC 次数 | 作业中所有 TaskManager 老年代 GC 次数之和 | 次 | tjob_id | [60s, sum]
[300s, last]
[3600s, last]
[86400s, last] |
TaskmanagerStatusJvmMemoryNonheapMax | TaskManager 非堆内存最大容量 | 作业中所有 TaskManager 非堆内存（JVM 元空间、代码缓存等）最大容量总和 | Bytes | tjob_id | [60s, sum]
[300s, avg]
[3600s, avg]
[86400s, avg] |
TaskmanagerMemoryDirectCount | TaskManager 堆外直接内存缓存数 | 作业中所有 TaskManager 堆外直接内存（Direct Buffer Pool）中的缓存（Buffer）个数之和 | 个 | tjob_id | [60s, sum]
[300s, avg]
[3600s, avg]
[86400s, avg] |
TaskmanagerMemoryDirectTotalcapacity | TaskManager 堆外直接内存总容量 | 作业中所有 TaskManager 堆外直接内存（Direct Buffer Pool）的最大容量之和 | Bytes | tjob_id | [60s, sum]
[300s, avg]
[3600s, avg]
[86400s, avg] |
TaskmanagerMemoryMappedCount | TaskManager 堆外映射内存缓存数 | 作业中所有 TaskManager 堆外映射内存（Mapped Buffer Pool）中的缓存（Buffer）个数之和 | 个 | tjob_id | [60s, sum]
[300s, avg]
[3600s, avg]
[86400s, avg] |
TaskmanagerMemoryMappedTotalcapacity | TaskManager 堆外映射内存总容量 | 作业中所有 TaskManager 堆外映射内存（Mapped Buffer Pool）的最大容量之和 | Bytes | tjob_id | [60s, sum]
[300s, avg]
[3600s, avg]
[86400s, avg] |
TaskmanagerStatusJvmMemoryHeapUsedPercentage | TaskManager 堆内存使用率 | 作业中所有 TaskManager 的平均堆内存使用率 | % | tjob_id | [60s, expr]
[300s, expr] |
TaskmanagerNetworkAvailablememorysegments | TaskManager 可用的 MemorySegment 个数 | 作业中所有 TaskManager 的可用 MemorySegment 个数之和 | Bytes | tjob_id | [60s, sum]
[300s, avg]
[3600s, avg]
[86400s, avg] |
JobMemoryHeapMax | TaskManager 堆内存最大容量 | 作业中所有 TaskManager 的堆内存最大（max）容量总和 | Bytes | tjob_id | [60s, sum]
[300s, avg]
[3600s, avg]
[86400s, avg] |
TaskmanagerCpuTime | TaskManager CPU 使用时长 | 作业中所有 TaskManager CPU 使用时长总和（毫秒） | ms | tjob_id | [60s, sum]
[300s, avg]
[3600s, avg]
[86400s, avg] |
JobBytesInPerSecond | 作业每秒输入的数据量 | 作业所有数据源（Source）每秒输入的数据总量（仅对 Kafka Source 有效） | Bytes/s | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
JobBytesOutPerSecond | 作业每秒输出的数据量 | 作业所有数据目的（Sink）每秒输出的数据总量（仅对 Kafka Sink 有效） | Bytes/s | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
JobmanagerNumrunningjobs | 运行中的作业数 | 正在运行中作业数。如果作业正常运行，则值为 1. 如果作业崩溃则值为 0 | 个 | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
TaskmanagerStatusJvmMemoryProcessMemoryused | 所有 TaskManager JVM 的物理内存用量的最大值 | 所有 TaskManager JVM 的物理内存用量的最大值 | Bytes | tjob_id | [60s, sum]
[300s, max]
[3600s, max]
[86400s, max] |
JobNumrecordsinbutfailed | 严重异常数据个数 | 算子中发生严重异常（例如抛出各种 Exception）的数据个数，如果大于 1 则会影响 Exactly-Once 语义（试验参数，仅供参考） | 个 | tjob_id | [60s, sum]
[300s, sum]
[3600s, sum]
[86400s, sum] |
Syndelay | 数据源同步百分比 | 数据源同步百分比 | % | tjob_id | [60s, last]
[300s, avg]
[3600s, avg]
[86400s, avg] |
Binlogpos | 数据源日志位点信息 | 数据源日志位点信息 | - | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
TaskmanagerJobTaskOperatorKafkaSwitch | 是否包含 Kafka connector | 是否包含 Kafka connector | - | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
JobmanagerTaskslotsavailable | 可用任务槽数量 | 如果作业正常运行，则可用的任务槽（Task Slot）数为 0. 如果不为 0 则说明作业可能出现短时间的非运行状态 | 个 | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
JobUptime | 作业无中断持续执行的时间 | 对于运行中的作业，表示当次作业持续处于运行中的时长 | ms | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
JobmanagerDowntime | 注册的 TaskManager 数 | 对于失败或恢复等非运行状态的作业，表示本次中断运行的时长。对于正在运行中的作业，值为 0. | ms | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
JobLastcheckpointduration | 最近一次的 Checkpoint 耗时 | 当前作业最近一次的 Checkpoint 耗时 | ms | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
JobLastcheckpointsize | 最近一次的 Checkpoint 大小 | 当前作业最近一次的 Checkpoint 大小 | Bytes | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
JobmanagerNumregisteredtaskmanagers | JobManager 已注册的 TaskManager 数 | 当前作业已注册的 TaskManager 数，通常等于所有算子并行度的最大值。如果 TaskManager 个数减少，说明存在 TaskManager 失联，作业可能崩溃并尝试恢复。 | 个 | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
JobmanagerMemoryNonheapCommitted | JobManager 已提交的非堆内存容量 | 当前作业已提交（committed）的 JobManager 非堆内存（JVM 元空间、代码缓存等）容量 | Bytes | tjob_id | [60s, max]
[300s, max]
[3600s, max]
[86400s, max] |
JobmanagerNumberofinprogresscheckpoints | 正在进行的 Checkpoint 个数 | 当前作业进行中（未完成）的 Checkpoint 个数 | 个 | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
JobmanagerThreadsCount | JobManager 中活动的线程数 | 当前作业 JobManager 中活动的线程数，含 Daemon 和非 Daemon 线程。 | 个 | tjob_id | [60s, max]
[300s, max]
[3600s, max]
[86400s, max] |
JobmanagerMemoryHeapCommitted | JobManager 已提交的堆内存容量 | 当前作业 JobManager 已提交（committed）的堆内存容量 | Bytes | tjob_id | [60s, max]
[300s, max]
[3600s, max]
[86400s, max] |
JobmanagerJvmYoungGcTime | JobManager 年轻代 GC 时间 | 当前作业 JobManager 年轻代 GC 时间 | ms | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
JobmanagerJvmYoungGcCount | JobManager 年轻代 GC 次数 | 当前作业 JobManager 年轻代 GC 次数 | 次 | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
JobmanagerJvmOldGcTime | JobManager 老年代 GC 时间 | 当前作业 JobManager 老年代 GC 时间 | ms | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
JobmanagerJvmOldGcCount | JobManager 老年代 GC 次数 | 当前作业 JobManager 老年代 GC 次数 | 次 | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
JobmanagerStatusJvmMemoryNonheapUsed | JobManager 非堆内存用量 | 当前作业 JobManager 非堆内存（JVM 元空间、代码缓存等）用量 | Bytes | tjob_id | [60s, max]
[300s, max]
[3600s, max]
[86400s, max] |
JobmanagerStatusJvmMemoryNonheapMax | JobManager 非堆内存最大容量 | 当前作业 JobManager 非堆内存（JVM 元空间、代码缓存等）的最大容量 | Bytes | tjob_id | [60s, max]
[300s, max]
[3600s, max]
[86400s, max] |
JobmanagerMemoryHeapMax | JobManager 堆内存最大容量 | 当前作业 JobManager 堆内存最大容量 | Bytes | tjob_id | [60s, max]
[300s, max]
[3600s, max]
[86400s, max] |
JobmanagerStatusJvmMemoryHeapUsedPercentage | JobManager 堆内存使用率 | 当前作业 JobManager 堆内存使用率 | % | tjob_id | [60s, expr]
[300s, expr] |
JobmanagerStatusJvmMemoryHeapUsed | JobManager 堆内存的用量 | 当前作业 JobManager 堆内存的用量 | Bytes | tjob_id | [60s, max]
[300s, max]
[3600s, max]
[86400s, max] |
JobmanagerCpuLoad | JobManager CPU 使用率 | 当前作业 JobManager 的 CPU 使用率 | % | tjob_id | [60s, avg]
[300s, max]
[3600s, max]
[86400s, max] |
JobmanagerCpuTime | JobManager CPU 使用时长 | 当前作业 JobManager CPU 使用时长（毫秒） | ms | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
JobNumberoffailedcheckpoints | Checkpoint 失败次数 | 当前作业 Checkpoint 失败（例如超时、遇到异常等）的次数 | 次 | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
JobNumberofcompletedcheckpoints | Checkpoint 成功次数 | 当前作业 Checkpoint 成功完成的次数 | 次 | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
JobmanagerJobNumrestarts | 当前实例崩溃重启次数 | 当前实例 JobManager 记录的任务崩溃重启次数（不含 JobManager 退出后作业重新拉起的场景） | 次 | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
RecordsLagMaxMin | Kafka - Records_Lag 最小值 | TaskManger 上报的 Kafka 最大 lag 指标最小值 | - | tjob_id | [60s, min]
[300s, min]
[3600s, min]
[86400s, min] |
RecordsLagMaxSum | Kafka - Records_Lag 最大值 | TaskManger 上报的 Kafka 最大 lag 指标的求和值 | - | tjob_id | [60s, sum]
[300s, sum]
[3600s, sum]
[86400s, sum] |
RecordsLagMaxAvg | Kafka - Records_Lag 均值 | TaskManger 上报的 Kafka 最大 lag 指标的均值 | - | tjob_id | [60s, avg]
[300s, avg]
[3600s, avg]
[86400s, avg] |
RecordsLagMax | Kafka - Records_Lag 最大指标 | TaskManger 上报的 Kafka 最大 lag 指标 | - | tjob_id | [60s, max]
[300s, max]
[3600s, max]
[86400s, max] |
Sourceidletime | Source 处理的空闲时间 | Source 处理的空闲时间 | ms | tjob_id | [60s, last]
[300s, avg]
[3600s, avg]
[86400s, avg] |
Currentfetcheventtimelag | Source Fetch消息的延迟时间 | Source Fetch消息的延迟时间(EmitTime-messageTimestamp) | ms | tjob_id | [60s, last]
[300s, avg]
[3600s, avg]
[86400s, avg] |
Dbflushdelay | Sink 刷新延迟 | Sink 刷新延迟 | ms | tjob_id | [60s, last]
[300s, avg]
[3600s, avg]
[86400s, avg] |
JobmanagerTaskslotstotal | 任务槽总数 | Oceanus 中一个 TaskManager 只有一个任务槽，因此任务槽总数等于注册的 TaskManager 数 | 个 | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
JobmanagerStatusJvmMemoryProcessMemoryused | JobManager 所在的 JVM 的物理内存用量 | JobManager 所在的 JVM 的物理内存用量 | Bytes | tjob_id | [60s, max]
[300s, max]
[3600s, max]
[86400s, max] |
JobmanagerMemoryDirectCount | JobManager 堆外直接内存中的缓存数 | JobManager 堆外直接内存（Direct Buffer Pool）中的缓存（Buffer）个数 | 个 | tjob_id | [60s, max]
[300s, max]
[3600s, max]
[86400s, max] |
JobmanagerMemoryDirectTotalcapacity | JobManager 堆外直接内存总容量 | JobManager 堆外直接内存（Direct Buffer Pool）的最大用量 | Bytes | tjob_id | [60s, max]
[300s, max]
[3600s, max]
[86400s, max] |
JobmanagerMemoryDirectMemoryused | JobManager 堆外直接内存使用量 | JobManager 堆外直接内存（Direct Buffer Pool）的用量 | Bytes | tjob_id | [60s, max]
[300s, max]
[3600s, max]
[86400s, max] |
JobmanagerMemoryMappedCount | JobManager 堆外存缓存个数 | JobManager 堆外映射内存（Mapped Buffer Pool）中的缓存（Buffer）个数之和 | 个 | tjob_id | [60s, max]
[300s, max]
[3600s, max]
[86400s, max] |
JobmanagerMemoryMappedTotalcapacity | JobManager 堆外映射内存的总容量 | JobManager 堆外映射内存（Mapped Buffer Pool）的最大用量 | Bytes | tjob_id | [60s, max]
[300s, max]
[3600s, max]
[86400s, max] |
JobmanagerMemoryMappedMemoryused | JobManager 堆外映射内存的使用量 | JobManager 堆外映射内存（Mapped Buffer Pool）的用量 | Bytes | tjob_id | [60s, max]
[300s, max]
[3600s, max]
[86400s, max] |
JobTotalnumberofcheckpoints | Checkpoint 总次数 | Checkpoint 总次数（进行中、已完成和失败的总和） | 次 | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
Currentemiteventtimelag | CDC Source 处理发送下游的时间与消息本身时间差 | CDC Source 处理发送下游的时间与消息本身时间差 | ms | tjob_id | [60s, last]
[300s, avg]
[3600s, avg]
[86400s, avg] |
TaskmanagerJobTaskOperatorSchemachange | CDC Schema 变更次数 | CDC Schema 变更次数 | 次 | tjob_id | [60s, last]
[300s, last]
[3600s, last]
[86400s, last] |
各维度对应参数总览
参数名称 | 维度名称 | 维度解释 | 格式 |
Instances.N.Dimensions.0.Name | tjob_id | 作业 ID 的维度名称 | 输入 String 类型维度名称：tjob_id |
Instances.N.Dimensions.0.Value | tjob_id | 具体作业 ID |
入参说明
查询云服务器监控数据，入参取值如下：
&Namespace=QCE/TSTREAM
&Instances.N.Dimensions.0.Name=tjob_id
&Instances.N.Dimensions.0.Value=具体作业 ID
