Yarn インテグレーションのメトリクス
Yarn の監視に使う Datadog の公式インテグレーションです。収集するメトリクスは 75 件あります。
公式の紹介文 (英語)
Collect cluster-wide health metrics and track application progress.
種別の内訳
- gauge (その時点の値): 66 件
- rate (1 秒あたりの値): 9 件
メトリクスの一覧
説明は公式の定義の英文をそのまま載せています。
| メトリクス | 種別 | 単位 |
|---|---|---|
yarn.Deprecated use yarn.apps.allocated_mb_gauge instead | rate | mebibyte |
yarn.The sum of memory in MB allocated to the applications running containers | gauge | mebibyte |
yarn.Deprecated use yarn.apps.allocated_vcores_gauge instead | rate | core |
yarn.The sum of virtual cores allocated to the applications running containers | gauge | core |
yarn.Deprecated use yarn.apps.elapsed_time_gauge instead | rate | second |
yarn.The elapsed time since the application started (in ms) | gauge | millisecond |
yarn.Deprecated use yarn.apps.finished_time_gauge instead | rate | second |
yarn.The time in which the application finished (in ms since epoch) | gauge | millisecond |
yarn.Deprecated use yarn.apps.memory_seconds_gauge instead | rate | second |
yarn.The amount of memory the application has allocated (megabyte-seconds) | gauge | mebibyte/ |
yarn.Deprecated use yarn.apps.progress_gauge instead | rate | percent |
yarn.The progress of the application, displayed as 0, 10, & 100, which represent the 3 states: hasn't started, in progress, & completed | gauge | percent |
yarn.Deprecated use yarn.apps.running_containers_gauge instead | rate | - |
yarn.The number of containers currently running for the application | gauge | container |
yarn.Deprecated use yarn.apps.started_time_gauge instead | rate | second |
yarn.The time in which application started (in ms since epoch) | gauge | millisecond |
yarn.Deprecated use yarn.apps.vcore_seconds_gauge instead | rate | second |
yarn.The amount of CPU resources the application has allocated (virtual core-seconds) | gauge | core/ |
yarn.The number of active nodes | gauge | node |
yarn.The amount of allocated memory | gauge | mebibyte |
yarn.The number of allocated virtual cores | gauge | core |
yarn.The number of completed apps | gauge | task |
yarn.The number of failed apps | gauge | task |
yarn.The number of killed apps | gauge | task |
yarn.The number of pending apps | gauge | task |
yarn.The number of running apps | gauge | task |
yarn.The number of submitted apps | gauge | task |
yarn.The amount of available memory | gauge | mebibyte |
yarn.The number of available virtual cores | gauge | core |
yarn.The number of containers allocated | gauge | - |
yarn.The number of containers pending | gauge | - |
yarn.The number of containers reserved | gauge | - |
yarn.The number of decommissioned nodes | gauge | node |
yarn.The number of decommissioning nodes | gauge | node |
yarn.The number of lost nodes | gauge | node |
yarn.The number of rebooted nodes | gauge | node |
yarn.The size of reserved memory | gauge | mebibyte |
yarn.The number of reserved virtual cores | gauge | core |
yarn.The amount of total memory | gauge | mebibyte |
yarn.The total number of nodes | gauge | node |
yarn.The total number of virtual cores | gauge | core |
yarn.The number of unhealthy nodes | gauge | node |
yarn.The total amount of memory currently available on the node (in MB) | gauge | mebibyte |
yarn.The total number of vCores available on the node | gauge | core |
yarn.The last time the node reported its health (in ms since epoch) | gauge | millisecond |
yarn.The total number of containers currently running on the node | gauge | - |
yarn.The total amount of memory currently used on the node (in MB) | gauge | mebibyte |
yarn.The total number of vCores currently used on the node | gauge | core |
yarn.The absolute capacity percentage this queue can use of entire cluster | gauge | percent |
yarn.The absolute maximum capacity percentage this queue can use of the entire cluster | gauge | percent |
yarn.The absolute used capacity percentage this queue is using of the entire cluster | gauge | percent |
yarn.The maximum memory resources this queue can use for Application Masters (in MB) | gauge | mebibyte |
yarn.The maximum vCpus this queue can use for Application Masters | gauge | core |
yarn.The configured queue capacity in percentage relative to its parent queue | gauge | percent |
yarn.The maximum number of active applications this queue can have | gauge | task |
yarn.The maximum number of active applications per user this queue can have | gauge | task |
yarn.The maximum number of applications this queue can have | gauge | task |
yarn.The maximum number of applications per user this queue can have | gauge | task |
yarn.The configured maximum queue capacity in percentage relative to its parent queue | gauge | percent |
yarn.The number of active applications in this queue | gauge | task |
yarn.The number of applications currently in the queue | gauge | task |
yarn.The number of containers being used | gauge | - |
yarn.The number of pending applications in this queue | gauge | task |
yarn.The total memory resources this queue is using (in MB) | gauge | mebibyte |
yarn.The total vCpus this queue is using | gauge | core |
yarn.The configured queue capacity in percentage for root queue | gauge | percent |
yarn.The configured maximum queue capacity in percentage for root queue | gauge | percent |
yarn.The used queue capacity in percentage for root queue | gauge | percent |
yarn.The memory resources used for Application Masters (in MB) | gauge | mebibyte |
yarn.The vCpus used for Application Masters | gauge | core |
yarn.The used queue capacity in percentage | gauge | percent |
yarn.The maximum memory resources a user can use for Application Masters (in MB) | gauge | mebibyte |
yarn.The maximum vCpus a user can use for Application Masters | gauge | core |
yarn.The user limit factor set in the configuration | gauge | - |
yarn.The minimum user limit percent set in the configuration | gauge | - |
公式ドキュメント
メトリクスの名前と説明は Datadog の公式インテグレーションの定義 (Copyright Datadog, Inc.・BSD-3-Clause) を英文のまま載せています。利用条件の全文は サイトについてにあります。