Hi, Thanks for the quick reply.
We are already using the config max.block.ms (alongwith with other recommended config like request.timeout.ms and others). Although the value we are using is very less at 250 ms but since we have 5 different KafkaProducer running in each individual thread out of which 4 are working without any issue and only 1 is throwing the TimeOutException, so this does not seems to be the issue, Please else us know if anybody had came across this type of behaviour by Kafka. If yes, please help in finding out the root cause and resolving it. Regards, Deepak -----Original Message----- From: 张晓寅 <zhxy0...@gmail.com> Sent: 06 June 2022 19:10 To: users@kafka.apache.org Cc: Luke Chen <show...@gmail.com> Subject: Re: Random continuous TimeoutException with Topic not present on one KafkaProducer out of many in multithreaded env Caution: From Cumulus Systems – IT Department, this email originated from outside of the organization. Please call and confirm with the sender before opening attachments or clicking links inside the email. maybe you can add producer "max.block.ms" config,but you should test your broker look up some logs about leader change ,producer performance,like traffic ,produce "buffer" and "batch.size" On Mon, Jun 6, 2022 at 6:53 PM Deepak Jain <deepak.j...@cumulus-systems.com> wrote: > Hello All, > > Please help me out in this regard as the Customer has reported this on > their production environment and waiting for our reply ASAP. > > Regards, > Deepak > > From: Deepak Jain > Sent: 02 June 2022 20:53 > To: 'users@kafka.apache.org' <users@kafka.apache.org> > Cc: 'Luke Chen' <show...@gmail.com>; Alap Patwardhan < > a...@cumulus-systems.com>; Bhushan Patil < > bhushan.pa...@cumulus-systems.com> > Subject: Random continuous TimeoutException with Topic not present on > one KafkaProducer out of many in multithreaded env > > Hello Everyone, > > We are using Kafka 2.8.1 Broker/Client system in our prod env. > > Getting following exception randomly after 1 hour or so for one > Realtime transfer from Kafka Producer to broker out of 5. (Rest 4 are > working fine.) > > java.util.concurrent.ExecutionException: > org.apache.kafka.common.errors.TimeoutException: Topic > realtimeImport_1 not present in metadata after 250 ms. > at > org.apache.kafka.clients.producer.KafkaProducer$FutureFailure.<init>(KafkaProducer.java:1316) > at > org.apache.kafka.clients.producer.KafkaProducer.doSend(KafkaProducer.java:985) > at > org.apache.kafka.clients.producer.KafkaProducer.send(KafkaProducer.java:885) > at > org.apache.kafka.clients.producer.KafkaProducer.send(KafkaProducer.jav > a:773) > > We are using multithreaded KafkaProducer with their each unique topic > sending data to single broker. Here, we notice that this exception > comes when we reconnect to Kafka using close() (void > org<eclipse-javadoc:%E2%98%82=hdca/D:%5C/git%5C/hdca%5C/megha%5C/app%5 > C/platform%5C/common%5C/tools%5C/lib%5C/kafka-clients-2.8.1.jar%3Corg> > .apache<eclipse-javadoc:%E2%98%82=hdca/D:%5C/git%5C/hdca%5C/megha%5C/a > pp%5C/platform%5C/common%5C/tools%5C/lib%5C/kafka-clients-2.8.1.jar%3C > org.apache>.kafka<eclipse-javadoc:%E2%98%82=hdca/D:%5C/git%5C/hdca%5C/ > megha%5C/app%5C/platform%5C/common%5C/tools%5C/lib%5C/kafka-clients-2. > 8.1.jar%3Corg.apache.kafka>.clients<eclipse-javadoc:%E2%98%82=hdca/D:% > 5C/git%5C/hdca%5C/megha%5C/app%5C/platform%5C/common%5C/tools%5C/lib%5 > C/kafka-clients-2.8.1.jar%3Corg.apache.kafka.clients>.producer<eclipse > -javadoc:%E2%98%82=hdca/D:%5C/git%5C/hdca%5C/megha%5C/app%5C/platform% > 5C/common%5C/tools%5C/lib%5C/kafka-clients-2.8.1.jar%3Corg.apache.kafk > a.clients.producer>.KafkaProducer<eclipse-javadoc:%E2%98%82=hdca/D:%5C > /git%5C/hdca%5C/megha%5C/app%5C/platform%5C/common%5C/tools%5C/lib%5C/ > kafka-clients-2.8.1.jar%3Corg.apache.kafka.clients.producer(KafkaProdu > cer.class%E2%98%83KafkaProducer>.close()) > and > org<eclipse-javadoc:%E2%98%82=hdca/D:%5C/git%5C/hdca%5C/megha%5C/app%5 > C/platform%5C/common%5C/tools%5C/lib%5C/kafka-clients-2.8.1.jar%3Corg> > .apache<eclipse-javadoc:%E2%98%82=hdca/D:%5C/git%5C/hdca%5C/megha%5C/a > pp%5C/platform%5C/common%5C/tools%5C/lib%5C/kafka-clients-2.8.1.jar%3C > org.apache>.kafka<eclipse-javadoc:%E2%98%82=hdca/D:%5C/git%5C/hdca%5C/ > megha%5C/app%5C/platform%5C/common%5C/tools%5C/lib%5C/kafka-clients-2. > 8.1.jar%3Corg.apache.kafka>.clients<eclipse-javadoc:%E2%98%82=hdca/D:% > 5C/git%5C/hdca%5C/megha%5C/app%5C/platform%5C/common%5C/tools%5C/lib%5 > C/kafka-clients-2.8.1.jar%3Corg.apache.kafka.clients>.producer<eclipse > -javadoc:%E2%98%82=hdca/D:%5C/git%5C/hdca%5C/megha%5C/app%5C/platform% > 5C/common%5C/tools%5C/lib%5C/kafka-clients-2.8.1.jar%3Corg.apache.kafk > a.clients.producer>.KafkaProducer<eclipse-javadoc:%E2%98%82=hdca/D:%5C > /git%5C/hdca%5C/megha%5C/app%5C/platform%5C/common%5C/tools%5C/lib%5C/ > kafka-clients-2.8.1.jar%3Corg.apache.kafka.clients.producer(KafkaProdu > cer.class%E2%98%83KafkaProducer>.KafkaProducer<Object<eclipse-javadoc: > %E2%98%82=hdca/D:%5C/git%5C/hdca%5C/megha%5C/app%5C/platform%5C/common > %5C/tools%5C/lib%5C/kafka-clients-2.8.1.jar%3Corg.apache.kafka.clients > .producer(KafkaProducer.class%E2%98%83KafkaProducer~KafkaProducer~Ljav > a.util.Properties;%E2%98%82java.lang.Object>, > Object<eclipse-javadoc:%E2%98%82=hdca/D:%5C/git%5C/hdca%5C/megha%5C/ap > p%5C/platform%5C/common%5C/tools%5C/lib%5C/kafka-clients-2.8.1.jar%3Co > rg.apache.kafka.clients.producer(KafkaProducer.class%E2%98%83KafkaProd > ucer~KafkaProducer~Ljava.util.Properties;%E2%98%82java.lang.Object>>(P > roperties<eclipse-javadoc:%E2%98%82=hdca/D:%5C/git%5C/hdca%5C/megha%5C > /app%5C/platform%5C/common%5C/tools%5C/lib%5C/kafka-clients-2.8.1.jar% > 3Corg.apache.kafka.clients.producer(KafkaProducer.class%E2%98%83KafkaP > roducer~KafkaProducer~Ljava.util.Properties;%E2%98%82java.util.Propert > ies> > properties) calls. Not sure whether this is the culprit or not. > > Due to this exception the realtime resources are not getting transfer > to Kafka Consumer. We are using following config on Kafka Broker: > > group.initial.rebalance.delay.ms=0 > listeners=SASL_PLAINTEXT://0.0.0.0:9092 > log.retention.minutes=15 > delete.topic.enable=true > auto.create.topics.enable=true > security.inter.broker.protocol=SASL_PLAINTEXT > sasl.mechanism.inter.broker.protocol=PLAIN > sasl.enabled.mechanisms=PLAIN > security.protocol=SASL_PLAINTEXT > authorizer.class.name=kafka.security.auth.SimpleAclAuthorizer > allow.everyone.if.no.acl.found=true > zookeeper.set.acl=true > > Can anyone please help us in finding the root cause for it? > > Regards, > Deepak Jain > Cumulus Systems >
