From df40cdc11641ad49b96cc8047d732fa33092f670 Mon Sep 17 00:00:00 2001 From: zwZjut Date: Tue, 28 Dec 2021 10:31:36 +0800 Subject: [PATCH] [Bug] [deploy on k8s] add default config of quartz and zookeeper in common configmap && pick #7562 fix to worker #7639 (#7641) * [Feature][dolphinscheduler-api] parse traceId in http header for Cross system delivery to #7237 (#7238) * to #7237 * rerun test Co-authored-by: honghuo.zw * chery-pick 05aef27 and handle conflicts * to #7065: fix ExecutorService and schedulerService (#7072) Co-authored-by: honghuo.zw * [Feature][dolphinscheduler-api] access control of taskDefinition and taskInstance in project to #7081 (#7082) * to #7081 * fix #7081 * to #7081 Co-authored-by: honghuo.zw * chery-pick 8ebe060 and handle conflicts * cherry-pick 1f18444 and handle conflicts * fix #6807: dolphinscheduler.zookeeper.env_vars - > dolphinscheduler.registry.env_vars (#6808) Co-authored-by: honghuo.zw Co-authored-by: Kirs * add default constructor (#6780) Co-authored-by: honghuo.zw * to #7108 (#7109) * add conf && pick #7562 to worker Co-authored-by: honghuo.zw Co-authored-by: Kirs --- docker/kubernetes/dolphinscheduler/values.yaml | 5 +++-- .../server/worker/registry/WorkerRegistryClient.java | 9 ++++----- 2 files changed, 7 insertions(+), 7 deletions(-) diff --git a/docker/kubernetes/dolphinscheduler/values.yaml b/docker/kubernetes/dolphinscheduler/values.yaml index 52431f9cad..d87f09e34a 100644 --- a/docker/kubernetes/dolphinscheduler/values.yaml +++ b/docker/kubernetes/dolphinscheduler/values.yaml @@ -104,6 +104,9 @@ common: HIVE_HOME: "/opt/soft/hive" FLINK_HOME: "/opt/soft/flink" DATAX_HOME: "/opt/soft/datax" + SESSION_TIMEOUT_MS: 60000 + ORG_QUARTZ_THREADPOOL_THREADCOUNT: "25" + ORG_QUARTZ_SCHEDULER_BATCHTRIGGERACQUISTITIONMAXCOUNT: "1" ## Shared storage persistence mounted into api, master and worker, such as Hadoop, Spark, Flink and DataX binary package sharedStoragePersistence: enabled: false @@ -165,7 +168,6 @@ master: MASTER_KILL_YARN_JOB_WHEN_HANDLE_FAILOVER: "true" ORG_QUARTZ_THREADPOOL_THREADCOUNT: "25" ORG_QUARTZ_SCHEDULER_BATCHTRIGGERACQUISTITIONMAXCOUNT: "1" - SESSION_TIMEOUT_MS: 60000 MASTER_PERSIST_EVENT_STATE_THREADS: 10 ## Periodic probe of container liveness. Container will be restarted if the probe fails. Cannot be updated. ## More info: https://kubernetes.io/docs/concepts/workloads/pods/pod-lifecycle#container-probes @@ -234,7 +236,6 @@ worker: WORKER_MAX_CPULOAD_AVG: "-1" WORKER_RESERVED_MEMORY: "0.3" WORKER_GROUPS: "default" - SESSION_TIMEOUT_MS: 60000 WORKER_RETRY_REPORT_TASK_STATUS_INTERVAL: 600 ## Periodic probe of container liveness. Container will be restarted if the probe fails. Cannot be updated. ## More info: https://kubernetes.io/docs/concepts/workloads/pods/pod-lifecycle#container-probes diff --git a/dolphinscheduler-server/src/main/java/org/apache/dolphinscheduler/server/worker/registry/WorkerRegistryClient.java b/dolphinscheduler-server/src/main/java/org/apache/dolphinscheduler/server/worker/registry/WorkerRegistryClient.java index 326fcb2bb6..74af482e5b 100644 --- a/dolphinscheduler-server/src/main/java/org/apache/dolphinscheduler/server/worker/registry/WorkerRegistryClient.java +++ b/dolphinscheduler-server/src/main/java/org/apache/dolphinscheduler/server/worker/registry/WorkerRegistryClient.java @@ -133,14 +133,13 @@ public class WorkerRegistryClient { public void handleConnectionState(ConnectionState state) { switch (state) { case CONNECTED: - logger.info("registry connection state is {}", state); + logger.debug("registry connection state is {}", state); break; case SUSPENDED: - logger.info("registry connection state is {}, ready to stop myself", state); - registryClient.getStoppable().stop("registry connection state is SUSPENDED, stop myself"); + logger.warn("registry connection state is {}, ready to retry connection", state); break; case RECONNECTED: - logger.info("registry connection state is {}, clean the node info", state); + logger.debug("registry connection state is {}, clean the node info", state); String address = NetUtils.getAddr(workerConfig.getListenPort()); Set workerZkPaths = getWorkerZkPaths(); for (String workerZKPath : workerZkPaths) { @@ -149,7 +148,7 @@ public class WorkerRegistryClient { } break; case DISCONNECTED: - logger.info("registry connection state is {}, ready to stop myself", state); + logger.warn("registry connection state is {}, ready to stop myself", state); registryClient.getStoppable().stop("registry connection state is DISCONNECTED, stop myself"); break; default: