[ https://issues.apache.org/jira/browse/KAFKA-12621?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jibitesh Prasad updated KAFKA-12621: ------------------------------------ Description: While configuring kafka with an znode apart from "/", the configuration is created in the wrong znode. Fo example, I have the following entry in my server.properties _zookeeper.connect=10.114.103.207:2181/kafka_secondary_cluster,10.114.103.206:2181/kafka_secondary_cluster,10.114.103.205:2181/kafka_secondary_cluster_ The IPs are the IP addresses of the nodes of zookeeper cluster. I expect the kafka server to use _kafka_secondary_cluster_ as the znode in the zookeeper nodes. But, the znode which is created is actually _/kafka_secondary_cluster,10.114.103.206:2181/kafka_secondary_cluster,10.114.103.205:2181/kafka_secondary_cluster_ Executing ls on the above path shows me the necessary znodes being created in that path _[zk: localhost:2181(CONNECTED) 1] ls /kafka_secondary_cluster,10.114.103.206:2181/kafka_secondary_cluster,10.114.103.205:2181/kafka_secondary_cluster_ Output: _[admin, brokers, cluster, config, consumers, controller, controller_epoch, isr_change_notification, latest_producer_id_block, log_dir_event_notification]_ Shouldn't these configurations be created in _/kafka_secondary_cluster_. It seems the comma separated values are not being split correctly. Or am I doing something wrong? was: While configuring kafka with an znode apart from "/", the configuration is created in the wrong znode. Fo example, I have the following entry in my server.properties _zookeeper.connect=10.114.103.207:2181/kafka_secondary_cluster,10.114.103.206:2181/kafka_secondary_cluster,10.114.103.205:2181/kafka_secondary_cluster_ The IPs are the IP addresses of the nodes of zookeeper cluster. I expect the kafka server to use _kafka_secondary_cluster_ as the znode in the zookeeper nodes. But, the znode which is created is actually _/kafka_secondary_cluster,10.114.103.206:2181/kafka_secondary_cluster,10.114.103.205:2181/kafka_secondary_cluster_ Executing ls on the above path shows me the necessary znodes being created in that path _[zk: localhost:2181(CONNECTED) 1] ls /kafka_secondary_cluster,10.114.103.206:2181/kafka_secondary_cluster,10.114.103.205:2181/kafka_secondary_cluster_ Output: _[admin, brokers, cluster, config, consumers, controller, controller_epoch, isr_change_notification, latest_producer_id_block, log_dir_event_notification]_ Shouldn't these configurations be created in _/kafka_secondary_cluster_. It seems the comma separated values are not being split correctly. Or am I doing something wrong? > Kafka setup with Zookeeper specifying an alternate znode as root fails > ---------------------------------------------------------------------- > > Key: KAFKA-12621 > URL: https://issues.apache.org/jira/browse/KAFKA-12621 > Project: Kafka > Issue Type: Bug > Components: config > Affects Versions: 2.6.1 > Environment: Linux > OS: 16.04.1-Ubuntu SMP > Architecture: x86_64 > Kernel Version: 4.15.0-1108-azure > Reporter: Jibitesh Prasad > Priority: Major > Labels: config, newbie, server.properties > > While configuring kafka with an znode apart from "/", the configuration is > created in the wrong znode. Fo example, I have the following entry in my > server.properties > _zookeeper.connect=10.114.103.207:2181/kafka_secondary_cluster,10.114.103.206:2181/kafka_secondary_cluster,10.114.103.205:2181/kafka_secondary_cluster_ > The IPs are the IP addresses of the nodes of zookeeper cluster. I expect the > kafka server to use _kafka_secondary_cluster_ as the znode in the zookeeper > nodes. But, the znode which is created is actually > _/kafka_secondary_cluster,10.114.103.206:2181/kafka_secondary_cluster,10.114.103.205:2181/kafka_secondary_cluster_ > Executing ls on the above path shows me the necessary znodes being created in > that path > _[zk: localhost:2181(CONNECTED) 1] ls > /kafka_secondary_cluster,10.114.103.206:2181/kafka_secondary_cluster,10.114.103.205:2181/kafka_secondary_cluster_ > Output: > _[admin, brokers, cluster, config, consumers, controller, controller_epoch, > isr_change_notification, latest_producer_id_block, > log_dir_event_notification]_ > Shouldn't these configurations be created in _/kafka_secondary_cluster_. It > seems the comma separated values are not being split correctly. Or am I doing > something wrong? -- This message was sent by Atlassian Jira (v8.3.4#803005)