[GitHub] [spark] dongjoon-hyun commented on a change in pull request #34734: [SPARK-37480][K8S][DOC] Sync Kubernetes configuration to latest in running-on-k8s.md
dongjoon-hyun commented on a change in pull request #34734: URL: https://github.com/apache/spark/pull/34734#discussion_r758926832 ## File path: docs/running-on-kubernetes.md ## @@ -1322,6 +1347,144 @@ See the [configuration page](configuration.html) for information on Spark config 3.3.0 + + spark.kubernetes.configMap.maxSize + 1572864 + +Max size limit for a config map. This is configurable as per https://etcd.io/docs/v3.4.0/dev-guide/limit/ on k8s server end. + + 3.1.0 + + + spark.kubernetes.allocation.executor.timeout + 600s + +Time to wait before a newly created executor POD request, which does not reached +the POD pending state yet, considered timedout and will be deleted. + + 3.1.0 + + + spark.kubernetes.executor.missingPodDetectDelta + 30s + +When a registered executor's POD is missing from the Kubernetes API server's polled +list of PODs then this delta time is taken as the accepted time difference between the +registration time and the time of the polling. After this time the POD is considered +missing from the cluster and the executor will be removed. + + 3.1.1 + + + spark.kubernetes.allocation.driver.readinessTimeout + 1s + +Time to wait for driver pod to get ready before creating executor pods. This wait +only happens on application start. If timeout happens, executor pods will still be +created. + + 3.1.3 + + + spark.kubernetes.decommission.script + /opt/decom.sh + +The location of the script to use for graceful decommissioning. + + 3.2.0 + + + spark.kubernetes.driver.service.deleteOnTermination + true + +If true, driver service will be deleted on Spark application termination. If false, it will be cleaned up when the driver pod is deletion. + + 3.2.0 + + + spark.kubernetes.driver.ownPersistentVolumeClaim + false + +If true, driver pod becomes the owner of on-demand persistent volume claims instead of the executor pods + + 3.2.0 + + + spark.kubernetes.driver.reusePersistentVolumeClaim + false + +If true, driver pod tries to reuse driver-owned on-demand persistent volume claims +of the deleted executor pods if exists. This can be useful to reduce executor pod +creation delay by skipping persistent volume creations. Note that a pod in +`Terminating` pod status is not a deleted pod by definition and its resources +including persistent volume claims are not reusable yet. Spark will create new +persistent volume claims when there exists no reusable one. In other words, the total +number of persistent volume claims can be larger than the number of running executors +sometimes. This config requires spark.kubernetes.driver.ownPersistentVolumeClaim=true. + + 3.2.0 + + + spark.kubernetes.executor.disableConfigMap + false + +If true, disable ConfigMap creation for executors. + + 3.2.0 + + + spark.kubernetes.driver.pod.featureSteps + (none) + +Class names of an extra driver pod feature step implementing +KubernetesFeatureConfigStep. This is a developer API. Comma separated. +Runs after all of Spark internal feature steps." + + 3.2.0 + + + spark.kubernetes.executor.pod.featureSteps + (none) + +Class name of an extra executor pod feature step implementing +KubernetesFeatureConfigStep. This is a developer API. Comma separated. +Runs after all of Spark internal feature steps. + + 3.2.0 + + + spark.kubernetes.allocation.maxPendingPods Review comment: Is there a reason why you put `spark.kubernetes.allocation` after `spark.kubernetes.executor`? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
[GitHub] [spark] dongjoon-hyun commented on a change in pull request #34734: [SPARK-37480][K8S][DOC] Sync Kubernetes configuration to latest in running-on-k8s.md
dongjoon-hyun commented on a change in pull request #34734: URL: https://github.com/apache/spark/pull/34734#discussion_r758926274 ## File path: docs/running-on-kubernetes.md ## @@ -1322,6 +1347,144 @@ See the [configuration page](configuration.html) for information on Spark config 3.3.0 + + spark.kubernetes.configMap.maxSize + 1572864 + +Max size limit for a config map. This is configurable as per https://etcd.io/docs/v3.4.0/dev-guide/limit/ on k8s server end. + + 3.1.0 + + + spark.kubernetes.allocation.executor.timeout + 600s + +Time to wait before a newly created executor POD request, which does not reached +the POD pending state yet, considered timedout and will be deleted. + + 3.1.0 + + + spark.kubernetes.executor.missingPodDetectDelta + 30s + +When a registered executor's POD is missing from the Kubernetes API server's polled +list of PODs then this delta time is taken as the accepted time difference between the +registration time and the time of the polling. After this time the POD is considered +missing from the cluster and the executor will be removed. + + 3.1.1 + + + spark.kubernetes.allocation.driver.readinessTimeout + 1s + +Time to wait for driver pod to get ready before creating executor pods. This wait +only happens on application start. If timeout happens, executor pods will still be +created. + + 3.1.3 + + + spark.kubernetes.decommission.script + /opt/decom.sh + +The location of the script to use for graceful decommissioning. + + 3.2.0 + + + spark.kubernetes.driver.service.deleteOnTermination + true + +If true, driver service will be deleted on Spark application termination. If false, it will be cleaned up when the driver pod is deletion. + + 3.2.0 + + + spark.kubernetes.driver.ownPersistentVolumeClaim + false + +If true, driver pod becomes the owner of on-demand persistent volume claims instead of the executor pods + + 3.2.0 + + + spark.kubernetes.driver.reusePersistentVolumeClaim + false + +If true, driver pod tries to reuse driver-owned on-demand persistent volume claims +of the deleted executor pods if exists. This can be useful to reduce executor pod +creation delay by skipping persistent volume creations. Note that a pod in +`Terminating` pod status is not a deleted pod by definition and its resources +including persistent volume claims are not reusable yet. Spark will create new +persistent volume claims when there exists no reusable one. In other words, the total +number of persistent volume claims can be larger than the number of running executors +sometimes. This config requires spark.kubernetes.driver.ownPersistentVolumeClaim=true. + + 3.2.0 + + + spark.kubernetes.executor.disableConfigMap + false + +If true, disable ConfigMap creation for executors. + + 3.2.0 + + + spark.kubernetes.driver.pod.featureSteps + (none) + +Class names of an extra driver pod feature step implementing +KubernetesFeatureConfigStep. This is a developer API. Comma separated. +Runs after all of Spark internal feature steps." + + 3.2.0 + + + spark.kubernetes.executor.pod.featureSteps + (none) + +Class name of an extra executor pod feature step implementing +KubernetesFeatureConfigStep. This is a developer API. Comma separated. +Runs after all of Spark internal feature steps. + + 3.2.0 + + + spark.kubernetes.allocation.maxPendingPods + Int.MaxValue + +Maximum number of pending PODs allowed during executor allocation for this +application. Those newly requested executors which are unknown by Kubernetes yet are +also counted into this limit as they will change into pending PODs by time. +This limit is independent from the resource profiles as it limits the sum of all +allocation for all the used resource profiles. + + 3.2.0 + + + spark.kubernetes.allocation.pods.allocator + direct + +Allocator to use for pods. Possible values are direct (the default) and statefulset, Review comment: codify `direct` and `statefulset`? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
[GitHub] [spark] dongjoon-hyun commented on a change in pull request #34734: [SPARK-37480][K8S][DOC] Sync Kubernetes configuration to latest in running-on-k8s.md
dongjoon-hyun commented on a change in pull request #34734: URL: https://github.com/apache/spark/pull/34734#discussion_r758926255 ## File path: docs/running-on-kubernetes.md ## @@ -1322,6 +1347,144 @@ See the [configuration page](configuration.html) for information on Spark config 3.3.0 + + spark.kubernetes.configMap.maxSize + 1572864 + +Max size limit for a config map. This is configurable as per https://etcd.io/docs/v3.4.0/dev-guide/limit/ on k8s server end. + + 3.1.0 + + + spark.kubernetes.allocation.executor.timeout + 600s + +Time to wait before a newly created executor POD request, which does not reached +the POD pending state yet, considered timedout and will be deleted. + + 3.1.0 + + + spark.kubernetes.executor.missingPodDetectDelta + 30s + +When a registered executor's POD is missing from the Kubernetes API server's polled +list of PODs then this delta time is taken as the accepted time difference between the +registration time and the time of the polling. After this time the POD is considered +missing from the cluster and the executor will be removed. + + 3.1.1 + + + spark.kubernetes.allocation.driver.readinessTimeout + 1s + +Time to wait for driver pod to get ready before creating executor pods. This wait +only happens on application start. If timeout happens, executor pods will still be +created. + + 3.1.3 + + + spark.kubernetes.decommission.script + /opt/decom.sh + +The location of the script to use for graceful decommissioning. + + 3.2.0 + + + spark.kubernetes.driver.service.deleteOnTermination + true + +If true, driver service will be deleted on Spark application termination. If false, it will be cleaned up when the driver pod is deletion. + + 3.2.0 + + + spark.kubernetes.driver.ownPersistentVolumeClaim + false + +If true, driver pod becomes the owner of on-demand persistent volume claims instead of the executor pods + + 3.2.0 + + + spark.kubernetes.driver.reusePersistentVolumeClaim + false + +If true, driver pod tries to reuse driver-owned on-demand persistent volume claims +of the deleted executor pods if exists. This can be useful to reduce executor pod +creation delay by skipping persistent volume creations. Note that a pod in +`Terminating` pod status is not a deleted pod by definition and its resources +including persistent volume claims are not reusable yet. Spark will create new +persistent volume claims when there exists no reusable one. In other words, the total +number of persistent volume claims can be larger than the number of running executors +sometimes. This config requires spark.kubernetes.driver.ownPersistentVolumeClaim=true. + + 3.2.0 + + + spark.kubernetes.executor.disableConfigMap + false + +If true, disable ConfigMap creation for executors. + + 3.2.0 + + + spark.kubernetes.driver.pod.featureSteps + (none) + +Class names of an extra driver pod feature step implementing +KubernetesFeatureConfigStep. This is a developer API. Comma separated. +Runs after all of Spark internal feature steps." + + 3.2.0 + + + spark.kubernetes.executor.pod.featureSteps + (none) + +Class name of an extra executor pod feature step implementing +KubernetesFeatureConfigStep. This is a developer API. Comma separated. +Runs after all of Spark internal feature steps. + + 3.2.0 + + + spark.kubernetes.allocation.maxPendingPods + Int.MaxValue + +Maximum number of pending PODs allowed during executor allocation for this +application. Those newly requested executors which are unknown by Kubernetes yet are +also counted into this limit as they will change into pending PODs by time. +This limit is independent from the resource profiles as it limits the sum of all +allocation for all the used resource profiles. + + 3.2.0 + + + spark.kubernetes.allocation.pods.allocator + direct Review comment: codify? ## File path: docs/running-on-kubernetes.md ## @@ -1322,6 +1347,144 @@ See the [configuration page](configuration.html) for information on Spark config 3.3.0 + + spark.kubernetes.configMap.maxSize + 1572864 + +Max size limit for a config map. This is configurable as per https://etcd.io/docs/v3.4.0/dev-guide/limit/ on k8s server end. + + 3.1.0 + + + spark.kubernetes.allocation.executor.timeout + 600s + +Time to wait before a newly created executor POD request, which does not reached +the POD pending state yet, considered timedout and will be deleted. + + 3.1.0 + + + spark.kubernetes.executor.missingPodDetectDelta + 30s + +When a registered executor's POD is missing from the Kubernetes API server's polled +list of PODs then this delta time is taken as the accepted time difference between the +registration time and the time of the polling. After this time the POD is
[GitHub] [spark] dongjoon-hyun commented on a change in pull request #34734: [SPARK-37480][K8S][DOC] Sync Kubernetes configuration to latest in running-on-k8s.md
dongjoon-hyun commented on a change in pull request #34734: URL: https://github.com/apache/spark/pull/34734#discussion_r758926045 ## File path: docs/running-on-kubernetes.md ## @@ -1322,6 +1347,144 @@ See the [configuration page](configuration.html) for information on Spark config 3.3.0 + + spark.kubernetes.configMap.maxSize + 1572864 + +Max size limit for a config map. This is configurable as per https://etcd.io/docs/v3.4.0/dev-guide/limit/ on k8s server end. + + 3.1.0 + + + spark.kubernetes.allocation.executor.timeout + 600s + +Time to wait before a newly created executor POD request, which does not reached +the POD pending state yet, considered timedout and will be deleted. + + 3.1.0 + + + spark.kubernetes.executor.missingPodDetectDelta + 30s + +When a registered executor's POD is missing from the Kubernetes API server's polled +list of PODs then this delta time is taken as the accepted time difference between the +registration time and the time of the polling. After this time the POD is considered +missing from the cluster and the executor will be removed. + + 3.1.1 + + + spark.kubernetes.allocation.driver.readinessTimeout + 1s + +Time to wait for driver pod to get ready before creating executor pods. This wait +only happens on application start. If timeout happens, executor pods will still be +created. + + 3.1.3 + + + spark.kubernetes.decommission.script + /opt/decom.sh + +The location of the script to use for graceful decommissioning. + + 3.2.0 + + + spark.kubernetes.driver.service.deleteOnTermination + true + +If true, driver service will be deleted on Spark application termination. If false, it will be cleaned up when the driver pod is deletion. + + 3.2.0 + + + spark.kubernetes.driver.ownPersistentVolumeClaim + false + +If true, driver pod becomes the owner of on-demand persistent volume claims instead of the executor pods + + 3.2.0 + + + spark.kubernetes.driver.reusePersistentVolumeClaim + false + +If true, driver pod tries to reuse driver-owned on-demand persistent volume claims +of the deleted executor pods if exists. This can be useful to reduce executor pod +creation delay by skipping persistent volume creations. Note that a pod in +`Terminating` pod status is not a deleted pod by definition and its resources +including persistent volume claims are not reusable yet. Spark will create new +persistent volume claims when there exists no reusable one. In other words, the total +number of persistent volume claims can be larger than the number of running executors +sometimes. This config requires spark.kubernetes.driver.ownPersistentVolumeClaim=true. + + 3.2.0 + + + spark.kubernetes.executor.disableConfigMap + false + +If true, disable ConfigMap creation for executors. + + 3.2.0 + + + spark.kubernetes.driver.pod.featureSteps + (none) + +Class names of an extra driver pod feature step implementing +KubernetesFeatureConfigStep. This is a developer API. Comma separated. +Runs after all of Spark internal feature steps." + + 3.2.0 + + + spark.kubernetes.executor.pod.featureSteps + (none) + +Class name of an extra executor pod feature step implementing Review comment: `Class name` -> `Class names`? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
[GitHub] [spark] dongjoon-hyun commented on a change in pull request #34734: [SPARK-37480][K8S][DOC] Sync Kubernetes configuration to latest in running-on-k8s.md
dongjoon-hyun commented on a change in pull request #34734: URL: https://github.com/apache/spark/pull/34734#discussion_r758925884 ## File path: docs/running-on-kubernetes.md ## @@ -1322,6 +1347,144 @@ See the [configuration page](configuration.html) for information on Spark config 3.3.0 + + spark.kubernetes.configMap.maxSize + 1572864 + +Max size limit for a config map. This is configurable as per https://etcd.io/docs/v3.4.0/dev-guide/limit/ on k8s server end. + + 3.1.0 + + + spark.kubernetes.allocation.executor.timeout + 600s + +Time to wait before a newly created executor POD request, which does not reached +the POD pending state yet, considered timedout and will be deleted. + + 3.1.0 + + + spark.kubernetes.executor.missingPodDetectDelta + 30s + +When a registered executor's POD is missing from the Kubernetes API server's polled +list of PODs then this delta time is taken as the accepted time difference between the +registration time and the time of the polling. After this time the POD is considered +missing from the cluster and the executor will be removed. + + 3.1.1 + + + spark.kubernetes.allocation.driver.readinessTimeout + 1s + +Time to wait for driver pod to get ready before creating executor pods. This wait +only happens on application start. If timeout happens, executor pods will still be +created. + + 3.1.3 + + + spark.kubernetes.decommission.script + /opt/decom.sh + +The location of the script to use for graceful decommissioning. + + 3.2.0 + + + spark.kubernetes.driver.service.deleteOnTermination + true + +If true, driver service will be deleted on Spark application termination. If false, it will be cleaned up when the driver pod is deletion. + + 3.2.0 + + + spark.kubernetes.driver.ownPersistentVolumeClaim + false + +If true, driver pod becomes the owner of on-demand persistent volume claims instead of the executor pods + + 3.2.0 + + + spark.kubernetes.driver.reusePersistentVolumeClaim + false + +If true, driver pod tries to reuse driver-owned on-demand persistent volume claims +of the deleted executor pods if exists. This can be useful to reduce executor pod +creation delay by skipping persistent volume creations. Note that a pod in +`Terminating` pod status is not a deleted pod by definition and its resources +including persistent volume claims are not reusable yet. Spark will create new +persistent volume claims when there exists no reusable one. In other words, the total +number of persistent volume claims can be larger than the number of running executors +sometimes. This config requires spark.kubernetes.driver.ownPersistentVolumeClaim=true. + + 3.2.0 + + + spark.kubernetes.executor.disableConfigMap + false + +If true, disable ConfigMap creation for executors. + + 3.2.0 + + + spark.kubernetes.driver.pod.featureSteps + (none) + +Class names of an extra driver pod feature step implementing +KubernetesFeatureConfigStep. This is a developer API. Comma separated. Review comment: Codify with `KubernetesFeatureConfigStep` or back-quotation? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
[GitHub] [spark] dongjoon-hyun commented on a change in pull request #34734: [SPARK-37480][K8S][DOC] Sync Kubernetes configuration to latest in running-on-k8s.md
dongjoon-hyun commented on a change in pull request #34734: URL: https://github.com/apache/spark/pull/34734#discussion_r758925510 ## File path: docs/running-on-kubernetes.md ## @@ -1322,6 +1347,144 @@ See the [configuration page](configuration.html) for information on Spark config 3.3.0 + + spark.kubernetes.configMap.maxSize + 1572864 + +Max size limit for a config map. This is configurable as per https://etcd.io/docs/v3.4.0/dev-guide/limit/ on k8s server end. + + 3.1.0 + + + spark.kubernetes.allocation.executor.timeout + 600s + +Time to wait before a newly created executor POD request, which does not reached +the POD pending state yet, considered timedout and will be deleted. + + 3.1.0 + + + spark.kubernetes.executor.missingPodDetectDelta + 30s + +When a registered executor's POD is missing from the Kubernetes API server's polled +list of PODs then this delta time is taken as the accepted time difference between the +registration time and the time of the polling. After this time the POD is considered +missing from the cluster and the executor will be removed. + + 3.1.1 + + + spark.kubernetes.allocation.driver.readinessTimeout + 1s + +Time to wait for driver pod to get ready before creating executor pods. This wait +only happens on application start. If timeout happens, executor pods will still be +created. + + 3.1.3 + + + spark.kubernetes.decommission.script + /opt/decom.sh + +The location of the script to use for graceful decommissioning. + + 3.2.0 + + + spark.kubernetes.driver.service.deleteOnTermination + true + +If true, driver service will be deleted on Spark application termination. If false, it will be cleaned up when the driver pod is deletion. + + 3.2.0 + + + spark.kubernetes.driver.ownPersistentVolumeClaim + false + +If true, driver pod becomes the owner of on-demand persistent volume claims instead of the executor pods + + 3.2.0 + + + spark.kubernetes.driver.reusePersistentVolumeClaim + false + +If true, driver pod tries to reuse driver-owned on-demand persistent volume claims +of the deleted executor pods if exists. This can be useful to reduce executor pod +creation delay by skipping persistent volume creations. Note that a pod in +`Terminating` pod status is not a deleted pod by definition and its resources +including persistent volume claims are not reusable yet. Spark will create new +persistent volume claims when there exists no reusable one. In other words, the total +number of persistent volume claims can be larger than the number of running executors +sometimes. This config requires spark.kubernetes.driver.ownPersistentVolumeClaim=true. + + 3.2.0 + + + spark.kubernetes.executor.disableConfigMap + false Review comment: ditto -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
[GitHub] [spark] dongjoon-hyun commented on a change in pull request #34734: [SPARK-37480][K8S][DOC] Sync Kubernetes configuration to latest in running-on-k8s.md
dongjoon-hyun commented on a change in pull request #34734: URL: https://github.com/apache/spark/pull/34734#discussion_r758925368 ## File path: docs/running-on-kubernetes.md ## @@ -1322,6 +1347,144 @@ See the [configuration page](configuration.html) for information on Spark config 3.3.0 + + spark.kubernetes.configMap.maxSize + 1572864 + +Max size limit for a config map. This is configurable as per https://etcd.io/docs/v3.4.0/dev-guide/limit/ on k8s server end. + + 3.1.0 + + + spark.kubernetes.allocation.executor.timeout + 600s + +Time to wait before a newly created executor POD request, which does not reached +the POD pending state yet, considered timedout and will be deleted. + + 3.1.0 + + + spark.kubernetes.executor.missingPodDetectDelta + 30s + +When a registered executor's POD is missing from the Kubernetes API server's polled +list of PODs then this delta time is taken as the accepted time difference between the +registration time and the time of the polling. After this time the POD is considered +missing from the cluster and the executor will be removed. + + 3.1.1 + + + spark.kubernetes.allocation.driver.readinessTimeout + 1s + +Time to wait for driver pod to get ready before creating executor pods. This wait +only happens on application start. If timeout happens, executor pods will still be +created. + + 3.1.3 + + + spark.kubernetes.decommission.script + /opt/decom.sh + +The location of the script to use for graceful decommissioning. + + 3.2.0 + + + spark.kubernetes.driver.service.deleteOnTermination + true Review comment: ditto ## File path: docs/running-on-kubernetes.md ## @@ -1322,6 +1347,144 @@ See the [configuration page](configuration.html) for information on Spark config 3.3.0 + + spark.kubernetes.configMap.maxSize + 1572864 + +Max size limit for a config map. This is configurable as per https://etcd.io/docs/v3.4.0/dev-guide/limit/ on k8s server end. + + 3.1.0 + + + spark.kubernetes.allocation.executor.timeout + 600s + +Time to wait before a newly created executor POD request, which does not reached +the POD pending state yet, considered timedout and will be deleted. + + 3.1.0 + + + spark.kubernetes.executor.missingPodDetectDelta + 30s + +When a registered executor's POD is missing from the Kubernetes API server's polled +list of PODs then this delta time is taken as the accepted time difference between the +registration time and the time of the polling. After this time the POD is considered +missing from the cluster and the executor will be removed. + + 3.1.1 + + + spark.kubernetes.allocation.driver.readinessTimeout + 1s + +Time to wait for driver pod to get ready before creating executor pods. This wait +only happens on application start. If timeout happens, executor pods will still be +created. + + 3.1.3 + + + spark.kubernetes.decommission.script + /opt/decom.sh + +The location of the script to use for graceful decommissioning. + + 3.2.0 + + + spark.kubernetes.driver.service.deleteOnTermination + true + +If true, driver service will be deleted on Spark application termination. If false, it will be cleaned up when the driver pod is deletion. + + 3.2.0 + + + spark.kubernetes.driver.ownPersistentVolumeClaim + false Review comment: ditto -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
[GitHub] [spark] dongjoon-hyun commented on a change in pull request #34734: [SPARK-37480][K8S][DOC] Sync Kubernetes configuration to latest in running-on-k8s.md
dongjoon-hyun commented on a change in pull request #34734: URL: https://github.com/apache/spark/pull/34734#discussion_r758925346 ## File path: docs/running-on-kubernetes.md ## @@ -1322,6 +1347,144 @@ See the [configuration page](configuration.html) for information on Spark config 3.3.0 + + spark.kubernetes.configMap.maxSize + 1572864 + +Max size limit for a config map. This is configurable as per https://etcd.io/docs/v3.4.0/dev-guide/limit/ on k8s server end. + + 3.1.0 + + + spark.kubernetes.allocation.executor.timeout + 600s + +Time to wait before a newly created executor POD request, which does not reached +the POD pending state yet, considered timedout and will be deleted. + + 3.1.0 + + + spark.kubernetes.executor.missingPodDetectDelta + 30s + +When a registered executor's POD is missing from the Kubernetes API server's polled +list of PODs then this delta time is taken as the accepted time difference between the +registration time and the time of the polling. After this time the POD is considered +missing from the cluster and the executor will be removed. + + 3.1.1 + + + spark.kubernetes.allocation.driver.readinessTimeout + 1s + +Time to wait for driver pod to get ready before creating executor pods. This wait +only happens on application start. If timeout happens, executor pods will still be +created. + + 3.1.3 + + + spark.kubernetes.decommission.script + /opt/decom.sh Review comment: ditto. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
[GitHub] [spark] dongjoon-hyun commented on a change in pull request #34734: [SPARK-37480][K8S][DOC] Sync Kubernetes configuration to latest in running-on-k8s.md
dongjoon-hyun commented on a change in pull request #34734: URL: https://github.com/apache/spark/pull/34734#discussion_r758925174 ## File path: docs/running-on-kubernetes.md ## @@ -1322,6 +1347,144 @@ See the [configuration page](configuration.html) for information on Spark config 3.3.0 + + spark.kubernetes.configMap.maxSize + 1572864 + +Max size limit for a config map. This is configurable as per https://etcd.io/docs/v3.4.0/dev-guide/limit/ on k8s server end. + + 3.1.0 + + + spark.kubernetes.allocation.executor.timeout + 600s + +Time to wait before a newly created executor POD request, which does not reached +the POD pending state yet, considered timedout and will be deleted. + + 3.1.0 + + + spark.kubernetes.executor.missingPodDetectDelta + 30s Review comment: ditto. ## File path: docs/running-on-kubernetes.md ## @@ -1322,6 +1347,144 @@ See the [configuration page](configuration.html) for information on Spark config 3.3.0 + + spark.kubernetes.configMap.maxSize + 1572864 + +Max size limit for a config map. This is configurable as per https://etcd.io/docs/v3.4.0/dev-guide/limit/ on k8s server end. + + 3.1.0 + + + spark.kubernetes.allocation.executor.timeout + 600s + +Time to wait before a newly created executor POD request, which does not reached +the POD pending state yet, considered timedout and will be deleted. + + 3.1.0 + + + spark.kubernetes.executor.missingPodDetectDelta + 30s + +When a registered executor's POD is missing from the Kubernetes API server's polled +list of PODs then this delta time is taken as the accepted time difference between the +registration time and the time of the polling. After this time the POD is considered +missing from the cluster and the executor will be removed. + + 3.1.1 + + + spark.kubernetes.allocation.driver.readinessTimeout + 1s Review comment: ditto -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
[GitHub] [spark] dongjoon-hyun commented on a change in pull request #34734: [SPARK-37480][K8S][DOC] Sync Kubernetes configuration to latest in running-on-k8s.md
dongjoon-hyun commented on a change in pull request #34734: URL: https://github.com/apache/spark/pull/34734#discussion_r758925122 ## File path: docs/running-on-kubernetes.md ## @@ -1322,6 +1347,144 @@ See the [configuration page](configuration.html) for information on Spark config 3.3.0 + + spark.kubernetes.configMap.maxSize + 1572864 + +Max size limit for a config map. This is configurable as per https://etcd.io/docs/v3.4.0/dev-guide/limit/ on k8s server end. + + 3.1.0 + + + spark.kubernetes.allocation.executor.timeout + 600s Review comment: Codify? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
[GitHub] [spark] dongjoon-hyun commented on a change in pull request #34734: [SPARK-37480][K8S][DOC] Sync Kubernetes configuration to latest in running-on-k8s.md
dongjoon-hyun commented on a change in pull request #34734: URL: https://github.com/apache/spark/pull/34734#discussion_r758925060 ## File path: docs/running-on-kubernetes.md ## @@ -1322,6 +1347,144 @@ See the [configuration page](configuration.html) for information on Spark config 3.3.0 + + spark.kubernetes.configMap.maxSize + 1572864 Review comment: Codify? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
[GitHub] [spark] dongjoon-hyun commented on a change in pull request #34734: [SPARK-37480][K8S][DOC] Sync Kubernetes configuration to latest in running-on-k8s.md
dongjoon-hyun commented on a change in pull request #34734: URL: https://github.com/apache/spark/pull/34734#discussion_r758924807 ## File path: docs/running-on-kubernetes.md ## @@ -1278,6 +1295,14 @@ See the [configuration page](configuration.html) for information on Spark config 3.0.0 + + spark.kubernetes.dynamicAllocation.deleteGracePeriod + 5s Review comment: I guess we need `` tag around this. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
[GitHub] [spark] dongjoon-hyun commented on a change in pull request #34734: [SPARK-37480][K8S][DOC] Sync Kubernetes configuration to latest in running-on-k8s.md
dongjoon-hyun commented on a change in pull request #34734: URL: https://github.com/apache/spark/pull/34734#discussion_r758924421 ## File path: docs/running-on-kubernetes.md ## @@ -1322,6 +1347,144 @@ See the [configuration page](configuration.html) for information on Spark config 3.3.0 + + spark.kubernetes.configMap.maxSize + 1572864 + +Max size limit for a config map. This is configurable as per https://etcd.io/docs/v3.4.0/dev-guide/limit/ on k8s server end. Review comment: Could you make it as a hyper link by using ``? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
[GitHub] [spark] dongjoon-hyun commented on a change in pull request #34734: [SPARK-37480][K8S][DOC] Sync Kubernetes configuration to latest in running-on-k8s.md
dongjoon-hyun commented on a change in pull request #34734: URL: https://github.com/apache/spark/pull/34734#discussion_r758922930 ## File path: docs/running-on-kubernetes.md ## @@ -591,7 +591,7 @@ See the [configuration page](configuration.html) for information on Spark config spark.kubernetes.container.image.pullPolicy IfNotPresent -Container image pull policy used when pulling images within Kubernetes. +Container image pull policy used when pulling images within Kubernetes. Valid values are Always, Never, and IfNotPresent. Review comment: Could you codify the values? For example, `Always`? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org