fix misleading example for cluster admin

[d1egoaz]

fixes #1238

kafka_1        | [2020-01-31 20:01:54,868] INFO [Admin Manager on Broker 1001]: Updating topic test.1.nokey with new configuration org.apache.kafka.common.requests.AlterConfigsRequest$Config@614a2b22 (kafka.server.AdminManager)
zookeeper_1    | [2020-01-31 20:01:54,880] INFO Got user-level KeeperException when processing sessionid:0x100008cfc4f0000 type:create cxid:0x3bf zxid:0x1eb txntype:-1 reqpath:n/a Error Path:/config/changes Error:KeeperErrorCode = NodeExists for /config/changes (org.apache.zookeeper.server.PrepRequestProcessor)
kafka_1        | [2020-01-31 20:01:54,930] INFO Processing notification(s) to /config/changes (kafka.common.ZkNodeChangeNotificationListener)
kafka_1        | [2020-01-31 20:01:54,953] INFO Processing override for entityPath: topics/test.1.nokey with config: Map(retention.ms -> 60000) (kafka.server.DynamicConfigManager)
kafka_1        | [2020-01-31 20:01:54,961] WARN [Log partition=test.1.nokey-0, dir=/kafka/kafka-logs-d40879780272] retention.ms for topic test.1.nokey is set to 60000. It is smaller than message.timestamp.difference.max.ms's value 9223372036854775807. This may result in frequent log rolling. (kafka.log.Log)