[jira] [Created] (FLINK-30218) [Kafka Connector] java.lang.OutOfMemoryError: Metaspace

["Lukas Mahl (Jira)" <ji...@apache.org>]

Lukas Mahl created FLINK-30218:
----------------------------------

             Summary: [Kafka Connector] java.lang.OutOfMemoryError: Metaspace
                 Key: FLINK-30218
                 URL: https://issues.apache.org/jira/browse/FLINK-30218
             Project: Flink
          Issue Type: Bug
          Components: Connectors / Kafka
    Affects Versions: 1.15.1
         Environment: +*AWS EMR*+
 * Standard AWS EMR Cluster (1 master YARN node, 1 core node) - 2 vCore, 8 GB memory each
 * JDK 11 Coretto

+*Kafka Consumer Config*+
{code:java}
    acks = 1
    batch.size = 16384
    bootstrap.servers = [...]
    buffer.memory = 33554432
    client.dns.lookup = use_all_dns_ips
    client.id = producer-1
    compression.type = none
    connections.max.idle.ms = 540000
    delivery.timeout.ms = 120000
    enable.idempotence = false
    interceptor.classes = []
    internal.auto.downgrade.txn.commit = false
    key.serializer = class org.apache.kafka.common.serialization.ByteArraySerializer
    linger.ms = 0
    max.block.ms = 60000
    max.in.flight.requests.per.connection = 5
    max.request.size = 1048576
    metadata.max.age.ms = 300000
    metadata.max.idle.ms = 300000
    metric.reporters = []
    metrics.num.samples = 2
    metrics.recording.level = INFO
    metrics.sample.window.ms = 30000
    partitioner.class = class org.apache.kafka.clients.producer.internals.DefaultPartitioner
    receive.buffer.bytes = 32768
    reconnect.backoff.max.ms = 1000
    reconnect.backoff.ms = 50
    request.timeout.ms = 30000
    retries = 2147483647
    retry.backoff.ms = 100
    sasl.client.callback.handler.class = class software.amazon.msk.auth.iam.IAMClientCallbackHandler
    sasl.jaas.config = [hidden]
    sasl.kerberos.kinit.cmd = /usr/bin/kinit
    sasl.kerberos.min.time.before.relogin = 60000
    sasl.kerberos.service.name = null
    sasl.kerberos.ticket.renew.jitter = 0.05
    sasl.kerberos.ticket.renew.window.factor = 0.8
    sasl.login.callback.handler.class = null
    sasl.login.class = null
    sasl.login.refresh.buffer.seconds = 300
    sasl.login.refresh.min.period.seconds = 60
    sasl.login.refresh.window.factor = 0.8
    sasl.login.refresh.window.jitter = 0.05
    sasl.mechanism = AWS_MSK_IAM
    security.protocol = SASL_SSL
    security.providers = null
    send.buffer.bytes = 131072
    socket.connection.setup.timeout.max.ms = 30000
    socket.connection.setup.timeout.ms = 10000
    ssl.cipher.suites = null
    ssl.enabled.protocols = [TLSv1.2, TLSv1.3]
    ssl.endpoint.identification.algorithm = https
    ssl.engine.factory.class = null
    ssl.key.password = null
    ssl.keymanager.algorithm = SunX509
    ssl.keystore.certificate.chain = null
    ssl.keystore.key = null
    ssl.keystore.location = null
    ssl.keystore.password = null
    ssl.keystore.type = JKS
    ssl.protocol = TLSv1.3
    ssl.provider = null
    ssl.secure.random.implementation = null
    ssl.trustmanager.algorithm = PKIX
    ssl.truststore.certificates = null
    ssl.truststore.location = null
    ssl.truststore.password = null
    ssl.truststore.type = JKS
    transaction.timeout.ms = 3600000
    transactional.id = null
    value.serializer = class org.apache.kafka.common.serialization.ByteArraySerializer{code}
+*Flink Config*+
{code:java}
taskmanager.memory.process.size=3g
taskmanager.memory.jvm-metaspace.size=512m
taskmanager.numberOfTaskSlots=2
jobmanager.memory.process.size=3g
jobmanager.memory.jvm-metaspace.size=256m
jobmanager.web.address0.0.0.0
env.java.opts.all-XX:+HeapDumpOnOutOfMemoryError -XX:HeapDumpPath=${FLINK_LOG_PREFIX}.hprof {code}
 
            Reporter: Lukas Mahl
         Attachments: dump.hprof.zip, image-2022-11-25-15-55-43-559.png

Hello!

I'm running a Flink application on AWS EMR which consumes from a Kafka Topic using the official Flink Kafka consumer. I'm running the application as a Flink batch job every 30 minutes and I see that the JobManager's metaspace is increasing every time I submit a new job and doesn't reduce once a job has finished executing. Eventually the metaspace overflows and I get a OutOfMemory metaspace exception. I've tried increasing the metaspace to 512m, but this just delays the problem - hence it's definitely a classloading leak.

I debugged the issue by creating a simple Flink application with a Kafka consumer only and the issue still occurred, hence I suppose the issue is somewhere in the Kafka consumer. Only other third party plugin I was using when doing so was the AWS IAM Kafka dependency (software.amazon.msk:aws-msk-iam-auth:1.1.5). 

I also tried debugging the issue by generating a heap dump as described here ([https://cwiki.apache.org/confluence/display/FLINK/Debugging+ClassLoader+leak|[https://cwiki.apache.org/confluence/display/FLINK/Debugging+ClassLoader+leaks]),] but I wasn't really able to spot the origin of the memory leak. I can see references to org.apache.kafka.common.utils.AppInfoParser$AppInfo though:

!image-2022-11-25-15-55-43-559.png|width=666,height=278!

I have attached the heap dump - if you need any more information from my set up feel free to ask, I can provide anything you need.

Many thanks in advance! 



--
This message was sent by Atlassian Jira
(v8.20.10#820010)