Why is Cassandra crashing whenever I try to run DataStax Kafka Connector? - apache-kafka

Question

Goal: My goal is to use Kafka to send messages to a Cassandra sink using Kafka Connect.
I've deployed Kafka and Cassandra and I am able to work with each of them individually - I have no problem sending data to Kafka, using producers to pass messages, and using consumers to consume them. I have no problem using cqlsh to create tables and insert data into them. However, whenever I try to deploy the DataStax Apache Kafka Connector, Cassandra seems to crash.
I am trying to learn how to use Kafka Connect using just one Kafka producer, broker, and one Cassandra keyspace using the standalone mode. I've configured both connect-standalone.properties and the cassandra-sink-standalone.properties following the instructions shown on DataStax: https://docs.datastax.com/en/kafka/doc/kafka/kafkaStringJson.html
connect-standalone.properties
bootstrap.servers=localhost:9092
key.converter=org.apache.kafka.connect.storage.StringConverter
value.converter=org.apache.kafka.connect.json.JsonConverter
key.converter.schemas.enable=false
value.converter.schemas.enable=false
offset.storage.file.filename=/tmp/connect.offsets
offset.flush.interval.ms=10000
plugin.path= *install_location*/kafka-connect-cassandra-sink-1.4.0.jar
cassandra-sink-standalone.properties
name=stocks-sink
connector.class=com.datastax.kafkaconnector.DseSinkConnector
tasks.max=1
topics=stocks_topic
topic.stocks_topic.stocks_keyspace.stocks_table.mapping = symbol=value.symbol, ts=value.ts, exchange=value.exchange, industry=value.industry, name=key, value=value.value
Then, the Kafka Connector is started using bin/connect-standalone.sh connect-standalone.properties cassandra-sink-standalone.properties.
About 95% of the time I attempt to launch Kafka Connector, Cassandra crashes. Running bin/nodetool status shows the message:
nodetool: Failed to connect to '127.0.0.1:7199' - ConnectException: 'Connection refused (Connection refused)'
In the system.log and debug.log logs, there is no indication that Cassandra has even crashed. The last line just remains as:
INFO [main] 2023-01-31 00:00:00,143 StorageService.java:2806 - Node localhost/127.0.0.1:7000 state jump to NORMAL
And in the Kafka Connect logs, the error messages states:
[2023-01-31 15:24:47,803] INFO [plc-sink|task-0] DataStax Java driver for Apache Cassandra(R) (com.datastax.oss:java-driver-core) version 4.6.0 (com.datastax.oss.driver.internal.core.DefaultMavenCoordinates:37)
[2023-01-31 15:24:47,947] INFO [plc-sink|task-0] Could not register Graph extensions; this is normal if Tinkerpop was explicitly excluded from classpath (com.datastax.oss.driver.internal.core.context.InternalDriverContext:540)
[2023-01-31 15:24:47,948] INFO [plc-sink|task-0] Could not register Reactive extensions; this is normal if Reactive Streams was explicitly excluded from classpath (com.datastax.oss.driver.internal.core.context.InternalDriverContext:559)
[2023-01-31 15:24:47,997] INFO [plc-sink|task-0] Using native clock for microsecond precision (com.datastax.oss.driver.internal.core.time.Clock:40)
[2023-01-31 15:24:47,999] INFO [plc-sink|task-0] [s0] No contact points provided, defaulting to /127.0.0.1:9042 (com.datastax.oss.driver.internal.core.metadata.MetadataManager:134)
[2023-01-31 15:24:48,190] WARN [plc-sink|task-0] [s0] Error connecting to Node(endPoint=/127.0.0.1:9042, hostId=null, hashCode=3247c5e4), trying next node (ConnectionInitException: [s0|control|connecting...] Protocol initialization request, step 1 (OPTIONS): failed to send request (java.nio.channels.ClosedChannelException)) (com.datastax.oss.driver.internal.core.control.ControlConnection:34)
[2023-01-31 15:24:48,200] ERROR [plc-sink|task-0] WorkerSinkTask{id=plc-sink-0} Task threw an uncaught and unrecoverable exception. Task is being killed and will not recover until manually restarted (org.apache.kafka.connect.runtime.WorkerTask:196)
com.datastax.oss.driver.api.core.AllNodesFailedException: Could not reach any contact point, make sure you've provided valid addresses (showing first 1 nodes, use getAllErrors() for more): Node(endPoint=/127.0.0.1:9042, hostId=null, hashCode=3247c5e4): [com.datastax.oss.driver.api.core.connection.ConnectionInitException: [s0|control|connecting...] Protocol initialization request, step 1 (OPTIONS): failed to send request (java.nio.channels.ClosedChannelException)]
at com.datastax.oss.driver.api.core.AllNodesFailedException.copy(AllNodesFailedException.java:141)
at com.datastax.oss.driver.internal.core.util.concurrent.CompletableFutures.getUninterruptibly(CompletableFutures.java:149)
at com.datastax.oss.driver.api.core.session.SessionBuilder.build(SessionBuilder.java:612)
at com.datastax.oss.kafka.sink.state.LifeCycleManager.buildCqlSession(LifeCycleManager.java:518)
at com.datastax.oss.kafka.sink.state.LifeCycleManager.lambda$startTask$0(LifeCycleManager.java:113)
at java.util.concurrent.ConcurrentHashMap.computeIfAbsent(ConcurrentHashMap.java:1660)
at com.datastax.oss.kafka.sink.state.LifeCycleManager.startTask(LifeCycleManager.java:109)
at com.datastax.oss.kafka.sink.CassandraSinkTask.start(CassandraSinkTask.java:83)
at org.apache.kafka.connect.runtime.WorkerSinkTask.initializeAndStart(WorkerSinkTask.java:312)
at org.apache.kafka.connect.runtime.WorkerTask.doRun(WorkerTask.java:187)
at org.apache.kafka.connect.runtime.WorkerTask.run(WorkerTask.java:244)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:750)
Suppressed: com.datastax.oss.driver.api.core.connection.ConnectionInitException: [s0|control|connecting...] Protocol initialization request, step 1 (OPTIONS): failed to send request (java.nio.channels.ClosedChannelException)
at com.datastax.oss.driver.internal.core.channel.ProtocolInitHandler$InitRequest.fail(ProtocolInitHandler.java:342)
at com.datastax.oss.driver.internal.core.channel.ChannelHandlerRequest.writeListener(ChannelHandlerRequest.java:87)
at io.netty.util.concurrent.DefaultPromise.notifyListener0(DefaultPromise.java:577)
at io.netty.util.concurrent.DefaultPromise.notifyListenersNow(DefaultPromise.java:551)
at io.netty.util.concurrent.DefaultPromise.notifyListeners(DefaultPromise.java:490)
at io.netty.util.concurrent.DefaultPromise.addListener(DefaultPromise.java:183)
at io.netty.channel.DefaultChannelPromise.addListener(DefaultChannelPromise.java:95)
at io.netty.channel.DefaultChannelPromise.addListener(DefaultChannelPromise.java:30)
at com.datastax.oss.driver.internal.core.channel.ChannelHandlerRequest.send(ChannelHandlerRequest.java:76)
at com.datastax.oss.driver.internal.core.channel.ProtocolInitHandler$InitRequest.send(ProtocolInitHandler.java:183)
at com.datastax.oss.driver.internal.core.channel.ProtocolInitHandler.onRealConnect(ProtocolInitHandler.java:118)
at com.datastax.oss.driver.internal.core.channel.ConnectInitHandler.lambda$connect$0(ConnectInitHandler.java:57)
at io.netty.util.concurrent.DefaultPromise.notifyListener0(DefaultPromise.java:577)
at io.netty.util.concurrent.DefaultPromise.notifyListeners0(DefaultPromise.java:570)
at io.netty.util.concurrent.DefaultPromise.notifyListenersNow(DefaultPromise.java:549)
at io.netty.util.concurrent.DefaultPromise.notifyListeners(DefaultPromise.java:490)
at io.netty.util.concurrent.DefaultPromise.setValue0(DefaultPromise.java:615)
at io.netty.util.concurrent.DefaultPromise.setFailure0(DefaultPromise.java:608)
at io.netty.util.concurrent.DefaultPromise.tryFailure(DefaultPromise.java:117)
at io.netty.channel.nio.AbstractNioChannel$AbstractNioUnsafe.fulfillConnectPromise(AbstractNioChannel.java:321)
at io.netty.channel.nio.AbstractNioChannel$AbstractNioUnsafe.finishConnect(AbstractNioChannel.java:337)
at io.netty.channel.nio.NioEventLoop.processSelectedKey(NioEventLoop.java:702)
at io.netty.channel.nio.NioEventLoop.processSelectedKeysOptimized(NioEventLoop.java:650)
at io.netty.channel.nio.NioEventLoop.processSelectedKeys(NioEventLoop.java:576)
at io.netty.channel.nio.NioEventLoop.run(NioEventLoop.java:493)
at io.netty.util.concurrent.SingleThreadEventExecutor$4.run(SingleThreadEventExecutor.java:989)
at io.netty.util.internal.ThreadExecutorMap$2.run(ThreadExecutorMap.java:74)
at io.netty.util.concurrent.FastThreadLocalRunnable.run(FastThreadLocalRunnable.java:30)
... 1 more
Suppressed: io.netty.channel.AbstractChannel$AnnotatedConnectException: Connection refused: /127.0.0.1:9042
Caused by: java.net.ConnectException: Connection refused
at sun.nio.ch.SocketChannelImpl.checkConnect(Native Method)
at sun.nio.ch.SocketChannelImpl.finishConnect(SocketChannelImpl.java:716)
at io.netty.channel.socket.nio.NioSocketChannel.doFinishConnect(NioSocketChannel.java:330)
at io.netty.channel.nio.AbstractNioChannel$AbstractNioUnsafe.finishConnect(AbstractNioChannel.java:334)
at io.netty.channel.nio.NioEventLoop.processSelectedKey(NioEventLoop.java:702)
at io.netty.channel.nio.NioEventLoop.processSelectedKeysOptimized(NioEventLoop.java:650)
at io.netty.channel.nio.NioEventLoop.processSelectedKeys(NioEventLoop.java:576)
at io.netty.channel.nio.NioEventLoop.run(NioEventLoop.java:493)
at io.netty.util.concurrent.SingleThreadEventExecutor$4.run(SingleThreadEventExecutor.java:989)
at io.netty.util.internal.ThreadExecutorMap$2.run(ThreadExecutorMap.java:74)
at io.netty.util.concurrent.FastThreadLocalRunnable.run(FastThreadLocalRunnable.java:30)
at java.lang.Thread.run(Thread.java:750)
Caused by: java.nio.channels.ClosedChannelException
at io.netty.channel.AbstractChannel$AbstractUnsafe.newClosedChannelException(AbstractChannel.java:957)
at io.netty.channel.AbstractChannel$AbstractUnsafe.flush0(AbstractChannel.java:921)
at io.netty.channel.nio.AbstractNioChannel$AbstractNioUnsafe.flush0(AbstractNioChannel.java:354)
at io.netty.channel.AbstractChannel$AbstractUnsafe.flush(AbstractChannel.java:897)
at io.netty.channel.DefaultChannelPipeline$HeadContext.flush(DefaultChannelPipeline.java:1372)
at io.netty.channel.AbstractChannelHandlerContext.invokeFlush0(AbstractChannelHandlerContext.java:748)
at io.netty.channel.AbstractChannelHandlerContext.invokeFlush(AbstractChannelHandlerContext.java:740)
at io.netty.channel.AbstractChannelHandlerContext.flush(AbstractChannelHandlerContext.java:726)
at io.netty.channel.ChannelDuplexHandler.flush(ChannelDuplexHandler.java:127)
at io.netty.channel.AbstractChannelHandlerContext.invokeFlush0(AbstractChannelHandlerContext.java:748)
at io.netty.channel.AbstractChannelHandlerContext.invokeWriteAndFlush(AbstractChannelHandlerContext.java:763)
at io.netty.channel.AbstractChannelHandlerContext.write(AbstractChannelHandlerContext.java:788)
at io.netty.channel.AbstractChannelHandlerContext.writeAndFlush(AbstractChannelHandlerContext.java:756)
at io.netty.channel.AbstractChannelHandlerContext.writeAndFlush(AbstractChannelHandlerContext.java:806)
at io.netty.channel.DefaultChannelPipeline.writeAndFlush(DefaultChannelPipeline.java:1025)
at io.netty.channel.AbstractChannel.writeAndFlush(AbstractChannel.java:294)
at com.datastax.oss.driver.internal.core.channel.ChannelHandlerRequest.send(ChannelHandlerRequest.java:75)
... 20 more
In the 5% of the time that Cassandra doesn't actually crash, the following message shows up in Kafka Connect's logs:
[2023-01-31 15:41:32,839] INFO [plc-sink|task-0] DataStax Java driver for Apache Cassandra(R) (com.datastax.oss:java-driver-core) version 4.6.0 (com.datastax.oss.driver.internal.core.DefaultMavenCoordinates:37)
[2023-01-31 15:41:32,981] INFO [plc-sink|task-0] Could not register Graph extensions; this is normal if Tinkerpop was explicitly excluded from classpath (com.datastax.oss.driver.internal.core.context.InternalDriverContext:540)
[2023-01-31 15:41:32,982] INFO [plc-sink|task-0] Could not register Reactive extensions; this is normal if Reactive Streams was explicitly excluded from classpath (com.datastax.oss.driver.internal.core.context.InternalDriverContext:559)
[2023-01-31 15:41:33,037] INFO [plc-sink|task-0] Using native clock for microsecond precision (com.datastax.oss.driver.internal.core.time.Clock:40)
[2023-01-31 15:41:33,040] INFO [plc-sink|task-0] [s0] No contact points provided, defaulting to /127.0.0.1:9042 (com.datastax.oss.driver.internal.core.metadata.MetadataManager:134)
[2023-01-31 15:41:33,254] INFO [plc-sink|task-0] [s0] Failed to connect with protocol DSE_V2, retrying with DSE_V1 (com.datastax.oss.driver.internal.core.channel.ChannelFactory:224)
[2023-01-31 15:41:33,263] INFO [plc-sink|task-0] [s0] Failed to connect with protocol DSE_V1, retrying with V4 (com.datastax.oss.driver.internal.core.channel.ChannelFactory:224)
[2023-01-31 15:41:34,091] INFO [plc-sink|task-0] WorkerSinkTask{id=plc-sink-0} Sink task finished initialization and start (org.apache.kafka.connect.runtime.WorkerSinkTask:313)
[2023-01-31 15:41:34,092] INFO [plc-sink|task-0] WorkerSinkTask{id=plc-sink-0} Executing sink task (org.apache.kafka.connect.runtime.WorkerSinkTask:198)
...
Versions:
Apache Cassandra 4.0.7
Apache Kafka 3.3.1
DataStax Apache Kafka Connector 1.4.0
I am currently using WSL2 Ubuntu 20.04.5 on Windows 11, with the following specs:
CPU: 4 Cores
Memory: 8GB RAM
Disk (SSD): 250 GB
Seeing that it actually works 5% of the time, I suspect that it's an OOM problem as outlined in https://community.datastax.com/questions/6947/index.html (and I sometimes just happen to have enough memory?). I've tried the solution in this article but it didn't help. How can I configure Cassandra / Kafka Connect to avoid this problem? Is this just a matter of needing a computer with more memory?

Answer 1

I think you're on the right track when you suggested that memory is an issue.
I have a "small" Windows Surface Pro I use to replicate issues like yours. I'm also running Ubuntu 20.04 with WSL2 on this laptop.
By default, Windows allocates half of system RAM to WSL2 so on my sub-8GB RAM installation, my Ubuntu installation takes up 3.7GB of memory. A vanilla installation of Cassandra (out-of-the-box zero configuration), starts with 1.3GB of memory allocated to it so there's only 2.4GB left for everything else.
What I suspect is happening is that when you start Kafka on the same node, Ubuntu runs out of memory and it triggers the Linux oom-killer. Although the end result is similar, the trigger is slightly different to what I described in the post you linked so my recommendation to explicitly set disk_access_mode doesn't help in this situation.
As a workaround, configure Cassandra to only allocate 1GB of memory by setting the MAX_HEAP_SIZE in conf/cassandra-env.sh:
MAX_HEAP_SIZE="1G"
Kafka is configured with 1GB by default but if it isn't, set the following in bin/kafka-server-start.sh:
export KAFKA_HEAP_OPTS="-Xmx1G -Xms1G"
By setting both, there should be over 1GB left for Ubuntu and hopefully allow you to run your tests. Cheers!