XX:+HeapDumpOnOutOfMemoryError -XX:HeapDumpPath=./logs -verbose:gc - XX:+PrintGCDetails -XX:+PrintGCDateStamps - XX:+PrintTenuringDistribution -Xloggc:./logs/kyuubi-server-gc-%t.log - XX:+UseGCLogFileRotation -XX:NumberOfGCLogFiles=10 kyuubi.delegation.key .update.interval PT24H unused yet dura tion 1.0 .0 kyuubi.delegation .token.gc.interval PT1H unused yet dura tion 1.0 .0 kyuubi.delegation .token.max.lifetime PT168H unused yet worker event loop of MySQL frontend service. int 1.4 .0 Key Default Meaning Type Since Use min(cpu_cores, 8) in default. kyuubi.frontend.mysql .worker.keepalive.time PT1M Time(ms) that an idle async

XX:+HeapDumpOnOutOfMemoryError -XX:HeapDumpPath=./logs -verbose:gc - XX:+PrintGCDetails -XX:+PrintGCDateStamps - XX:+PrintTenuringDistribution -Xloggc:./logs/kyuubi-server-gc-%t.log - XX:+UseGCLogFileRotation -XX:NumberOfGCLogFiles=10 kyuubi.delegation.key .update.interval PT24H unused yet dura tion 1.0 .0 kyuubi.delegation .token.gc.interval PT1H unused yet dura tion 1.0 .0 kyuubi.delegation .token.max.lifetime PT168H unused yet Number of thread in the netty worker event loop of MySQL frontend service. Use min(cpu_cores, 8) in default. int 1.4 .0 kyuubi.frontend.mysql .worker.keepalive.time PT1M Time(ms) that

to specify replicaCount and resources: replicaCount: 1 resources: requests: cpu: 2 memory: 4Gi limits: cpu: 4 memory: 10Gi and use it to override default chart values with -f flag: helm XX:+HeapDumpOnOutOfMemoryError -XX:HeapDumpPath=./logs -verbose:gc - XX:+PrintGCDetails -XX:+PrintGCDateStamps - XX:+PrintTenuringDistribution -Xloggc:./logs/kyuubi-server-gc-%t.log - XX:+UseGCLogFileRotation -XX:NumberOfGCLogFiles=10 Since kyuubi.delegation. key.update.interval PT24H unused yet duration 1.0.0 kyuubi.delegation.t oken.gc.interval PT1H unused yet duration 1.0.0 kyuubi.delegation.t oken.max.lifetime PT168H unused yet duration

yaml to specify replicaCount and resources: replicaCount: 1 resources: requests: cpu: 2 memory: 4Gi limits: cpu: 4 memory: 10Gi and use it to override default chart values with -f flag: helm install emoryError -XX:HeapDumpPath=./logs -verbose:gc -XX:+PrintGCDetails␣ ˓→-XX:+PrintGCDateStamps -XX:+PrintTenuringDistribution -Xloggc:./logs/kyuubi-server-gc- ˓→%t.log -XX:+UseGCLogFileRotation -XX:NumberOfGCLogFiles=10 dynamicAllocation. shuffleTracking.enabled is enabled, as we can tell Spark to be more active for shuffle data GC. Setting User Default Settings On the server-side, the workloads for different users might be different

XX:+HeapDumpOnOutOfMemoryError -XX:HeapDumpPath=./logs -verbose:gc -XX:+PrintGCDetails - XX:+PrintGCDateStamps -XX:+PrintTenuringDistribution -Xloggc:./logs/kyuubi-server-gc-%t.log -XX:+UseGCLogFileRotation -XX:NumberOfGCLogFiles=10 kyuubi.delegation.key.update.interval PT24H unused yet dura tion 1.0 .0 kyuubi.delegation.token.gc.interval PT1H unused yet dura tion 1.0 .0 kyuubi.delegation.token.max.lifetime PT168H unused yet threads Number of thread in the netty worker event loop of MySQL frontend service. Use min(cpu_cores, 8) in default. int 1.4 .0 kyuubi.frontend.mysql.worker.keepalive.time PT1M Time(ms) that

XX:+HeapDumpOnOutOfMemoryError -XX:HeapDumpPath=./logs -verbose:gc -XX:+PrintGCDetails - XX:+PrintGCDateStamps -XX:+PrintTenuringDistribution -Xloggc:./logs/kyuubi-server-gc-%t.log -XX:+UseGCLogFileRotation -XX:NumberOfGCLogFiles=10 kyuubi.delegation.key.update.interval PT24H unused yet dura tion 1.0 .0 kyuubi.delegation.token.gc.interval PT1H unused yet dura tion 1.0 .0 kyuubi.delegation.token.max.lifetime PT168H unused yet threads Number of thread in the netty worker event loop of MySQL frontend service. Use min(cpu_cores, 8) in default. int 1.4 .0 kyuubi.frontend.mysql.worker.keepalive.time PT1M Time(ms) that

XX:+HeapDumpOnOutOfMemoryError -XX:HeapDumpPath=./logs -verbose:gc -XX:+PrintGCDetails - XX:+PrintGCDateStamps -XX:+PrintTenuringDistribution -Xloggc:./logs/kyuubi-server-gc-%t.log -XX:+UseGCLogFileRotation -XX:NumberOfGCLogFiles=10 kyuubi.delegation.key.update.interval PT24H unused yet dura tion 1.0 .0 kyuubi.delegation.token.gc.interval PT1H unused yet dura tion 1.0 .0 kyuubi.delegation.token.max.lifetime PT168H unused yet threads Number of thread in the netty worker event loop of MySQL frontend service. Use min(cpu_cores, 8) in default. int 1.4 .0 kyuubi.frontend.mysql.worker.keepalive.time PT1M Time(ms) that

oryError -XX:HeapDumpPath=./logs -verbose:gc - ˓→XX:+PrintGCDetails -XX:+PrintGCDateStamps -XX:+PrintTenuringDistribution -Xloggc:./ ˓→logs/kyuubi-server-gc-%t.log -XX:+UseGCLogFileRotation -XX:NumberOfGCLogFiles=10 dynamicAllocation.shuffleTracking.enabled is enabled, as we can tell Spark to be more active for shuffle data GC. Setting User Default Settings On the server-side, the workloads for different users might be different to a high-weight pool to get more executors for execution. In essence, resource isolation such as CPU/memory/IO should be something that resource managers like YARN and Kubernetes should do. Doing logical

oryError -XX:HeapDumpPath=./logs -verbose:gc - ˓→XX:+PrintGCDetails -XX:+PrintGCDateStamps -XX:+PrintTenuringDistribution -Xloggc:./ ˓→logs/kyuubi-server-gc-%t.log -XX:+UseGCLogFileRotation -XX:NumberOfGCLogFiles=10 dynamicAllocation.shuffleTracking.enabled is enabled, as we can tell Spark to be more active for shuffle data GC. Setting User Default Settings On the server-side, the workloads for different users might be different to a high-weight pool to get more executors for execution. In essence, resource isolation such as CPU/memory/IO should be something that resource managers like YARN and Kubernetes should do. Doing logical

XX:+HeapDumpOnOutOfMemoryError -XX:HeapDumpPath=./logs -verbose:gc - XX:+PrintGCDetails -XX:+PrintGCDateStamps - XX:+PrintTenuringDistribution -Xloggc:./logs/kyuubi-server-gc-%t.log - XX:+UseGCLogFileRotation -XX:NumberOfGCLogFiles=10 kyuubi.delegation.key .update.interval PT24H unused yet dura tion 1.0 .0 kyuubi.delegation .token.gc.interval PT1H unused yet dura tion 1.0 .0 kyuubi.delegation .token.max.lifetime PT168H unused yet dynamicAllocation.shuffleTracking.enabled is enabled, as we can tell Spark to be more active for shuffle data GC. Setting User Default Settings On the server-side, the workloads for different users might be different