[KAFKA-10569] Running aggregate queries on KSQL client side is getting to ERROR Shutdown broker because all log dirs in ...

• Type: Bug
• Status: Resolved
• Priority: Critical
• Resolution: Invalid
• Affects Version/s: 2.5.0
• Component/s: clients
• Environment: local
• Flags: Important
• Assignee: Unassigned
• Reporter: Petre Gordan
• Created: 02/Oct/20 17:06
• Updated: 03/Oct/20 21:31
• Resolved: 03/Oct/20 21:31

Description

Working on Windows 10 and confluent-5.5.0-2.12.zip and kafka_2.12-2.5.0. I'm running locally:

• in powershell, zookeeper with: bin\windows\zookeeper-server-start.bat config\zookeeper.properties

• in powershell, kafka-server

bin\windows\kafka-server-start.bat config\server.properties

• in bash (with ubuntu) ksqldb server

sudo bin/ksql-server-start etc/ksqldb/ksql-server.properties

• in bash (with ubuntu) ksql client

sudo bin/ksql http://0.0.0.0:8088

After all of these are sorted, than, I'm starting to practice the Kafka. So, I'm creating tables, streams, making inserts, and all good. I can do small queries like: 

select * from products emit changes;, etc.

All good until at this step.

When, I'm trying to run every type of aggregate query, than is showing the result after a while, but into the end, after I will press Ctrl+C to terminate that and to do other query, everything is down.

For example, see the attached .sql script, and after I will run that script the products table and orders stream are created with success and populated with success.

After that if I run this query:

select ProductRowKey, count(ProductRowKey) from orders group by ProductRowKey emit changes;

I can see the results, all good, but into the end if I will press Ctrl + C, than everything is down.

 

Looking into the logs and taking based on the time history the main raised warning and issues are:

• first is raised this: 

Query terminated due to exception:org.eclipse.jetty.io.EofException (io.confluent.ksql.rest.server.resources.streaming.QueryStreamWriter:95)

• than this: INFO stream-client [_confluent-ksql-default_transient State transition from RUNNING to PENDING_SHUTDOWN (org.apache.kafka.streams.KafkaStreams:285)
• than these:

INFO stream-thread [_confluent-ksql-default_transient Informed to shut down (org.apache.kafka.streams.processor.internals.StreamThread:1116)
State transition from RUNNING to PENDING_SHUTDOWN (org.apache.kafka.streams.processor.internals.StreamThread:221)
Shutdown complete (org.apache.kafka.streams.processor.internals.StreamThread:1150)

• than this:

INFO stream-thread [qtp2032891036-47] Deleting obsolete state directory 0_0 for task 0_0 as 1ms has elapsed

• than this:

WARN Could not clean up the schema registry for query: confluent-ksql-default_transient

• than this:

WARN [Producer clientId=producer-1] Connection to node 0 (localhost/127.0.0.1:9092) could not be established. Broker may not be available. (org.apache.kafka.clients.NetworkClient:763)

all of these from above on the KSQLDB server side logs,

• finally this on the Kafka cluster side:

ERROR Shutdown broker because all log dirs in ........ have failed (kafka.log.LogManager)

 

And after that everything is down. Please see all the attached files to get all the info.

 

Please help me with these.