Re: on k8s 部署taskmanager一直不能启动

Mon, 11 Jul 2022 20:57:48 -0700 

看一下 TM pods 是否启动了？TM log 中是否有异常？看起来是 TM 一直没有注册上来

Best，
Lijie

陈卓宇 <2572805...@qq.com.invalid> 于2022年7月12日周二 10:53写道：

> flink：1.14.5
> on k8s 部署taskmanager一直不能启动，也没有日志
> jobmanager日志：
> 2022-07-12 02:08:22,271 INFO&nbsp;
> org.apache.flink.kubernetes.KubernetesResourceManagerDriver&nbsp; [] -
> Creating new TaskManager pod with name iii5-taskmanager-1-1 and resource
> <1728,1.0&gt;.
> 2022-07-12 02:08:22,286 WARN&nbsp;
> org.apache.flink.kafka.shaded.org.apache.kafka.clients.admin.AdminClientConfig
> [] - The configuration 'key.deserializer' was supplied but isn't a known
> config.
> 2022-07-12 02:08:22,286 WARN&nbsp;
> org.apache.flink.kafka.shaded.org.apache.kafka.clients.admin.AdminClientConfig
> [] - The configuration 'value.deserializer' was supplied but isn't a known
> config.
> 2022-07-12 02:08:22,286 WARN&nbsp;
> org.apache.flink.kafka.shaded.org.apache.kafka.clients.admin.AdminClientConfig
> [] - The configuration 'enable.auto.commit' was supplied but isn't a known
> config.
> 2022-07-12 02:08:22,287 WARN&nbsp;
> org.apache.flink.kafka.shaded.org.apache.kafka.clients.admin.AdminClientConfig
> [] - The configuration 'group.id' was supplied but isn't a known config.
> 2022-07-12 02:08:22,287 WARN&nbsp;
> org.apache.flink.kafka.shaded.org.apache.kafka.clients.admin.AdminClientConfig
> [] - The configuration 'client.id.prefix' was supplied but isn't a known
> config.
> 2022-07-12 02:08:22,287 WARN&nbsp;
> org.apache.flink.kafka.shaded.org.apache.kafka.clients.admin.AdminClientConfig
> [] - The configuration 'partition.discovery.interval.ms' was supplied but
> isn't a known config.
> 2022-07-12 02:08:22,287 WARN&nbsp;
> org.apache.flink.kafka.shaded.org.apache.kafka.clients.admin.AdminClientConfig
> [] - The configuration 'auto.offset.reset' was supplied but isn't a known
> config.
> 2022-07-12 02:08:22,287 INFO&nbsp;
> org.apache.flink.kafka.shaded.org.apache.kafka.common.utils.AppInfoParser
> [] - Kafka version: unknown
> 2022-07-12 02:08:22,287 INFO&nbsp;
> org.apache.flink.kafka.shaded.org.apache.kafka.common.utils.AppInfoParser
> [] - Kafka commitId: unknown
> 2022-07-12 02:08:22,287 INFO&nbsp;
> org.apache.flink.kafka.shaded.org.apache.kafka.common.utils.AppInfoParser
> [] - Kafka startTimeMs: 1657591702287
> 2022-07-12 02:08:22,354 INFO&nbsp;
> org.apache.flink.connector.kafka.source.enumerator.KafkaSourceEnumerator []
> - Starting the KafkaSourceEnumerator for consumer group
> hire_sign_contract_prod without periodic partition discovery.
> 2022-07-12 02:08:23,464 INFO&nbsp;
> org.apache.flink.kubernetes.KubernetesResourceManagerDriver&nbsp; [] - Pod
> iii5-taskmanager-1-1 is created.
> 2022-07-12 02:08:23,467 INFO&nbsp;
> org.apache.flink.connector.kafka.source.enumerator.KafkaSourceEnumerator []
> - Discovered new partitions: [canal_hire_sign_v2-11, canal_hire_sign_v2-9,
> canal_hire_sign_v2-10, canal_hire_sign_v2-0, canal_hire_sign_v2-3,
> canal_hire_sign_v2-4, canal_hire_sign_v2-1, canal_hire_sign_v2-2,
> canal_hire_sign_v2-7, canal_hire_sign_v2-8, canal_hire_sign_v2-5,
> canal_hire_sign_v2-6]
> 2022-07-12 02:08:23,576 INFO&nbsp;
> org.apache.flink.kubernetes.KubernetesResourceManagerDriver&nbsp; [] -
> Received new TaskManager pod: iii5-taskmanager-1-1
> 2022-07-12 02:08:23,578 INFO&nbsp;
> org.apache.flink.runtime.resourcemanager.active.ActiveResourceManager [] -
> Requested worker iii5-taskmanager-1-1 with resource spec WorkerResourceSpec
> {cpuCores=1.0, taskHeapSize=384.000mb (402653174 bytes), taskOffHeapSize=0
> bytes, networkMemSize=128.000mb (134217730 bytes), managedMemSize=512.000mb
> (536870920 bytes), numSlots=1}.
>
> 到这里就卡主了
> 然后过一段时间，会报slot分配的异常，但是机器的资源是够的，之前也是能启动的

回复