在所有Kafka经纪人重新启动以升级offset.retention.minutes设置（将其增加到60天）后，使用该卡夫卡Streams应用程序的用户被卡住，并且消费组显示了重新平衡：
bin/kafka-consumer-groups.sh --bootstrap-server ${BOOTSTRAP_SERVERS} --describe --group stream-processor | sort
Warning: Consumer group 'stream-processor' is rebalancing.
TOPIC PARTITION CURRENT-OFFSET LOG-END-OFFSET LAG CONSUMER-ID HOST CLIENT-ID
customers 0 84263 84288 25 - - -
customers 1 71731 85068 13337 - - -
customers 10 71841 84801 12960 - - -
customers 11 84273 84336 63 - - -
customers 12 84189 84297 108 - - -
customers 13 83969 84051 82 - - -
customers 14 84693 84767 74 - - -
customers 15 84472 84556 84 - - -
customers 2 84175 84239 64 - - -
customers 3 71719 71719 0 - - -
customers 4 71446 84499 13053 - - -
customers 5 84291 84361 70 - - -
customers 6 71700 71700 0 - - -
customers 7 72003 85235 13232 - - -
customers 8 84521 84587 66 - - -
customers 9 71513 71513 0 - - -
customers-intermediate 0 102774 102792 18 - - -
customers-intermediate 1 102931 103028 97 - - -
customers-intermediate 10 102883 102965 82 - - -
customers-intermediate 11 102723 102861 138 - - -
customers-intermediate 12 102797 102931 134 - - -
customers-intermediate 13 102339 102460 121 - - -
customers-intermediate 14 103227 103321 94 - - -
customers-intermediate 15 103231 103366 135 - - -
customers-intermediate 2 102637 102723 86 - - -
customers-intermediate 3 84620 103297 18677 - - -
customers-intermediate 4 102596 102687 91 - - -
customers-intermediate 5 102980 103071 91 - - -
customers-intermediate 6 84385 103058 18673 - - -
customers-intermediate 7 103559 103652 93 - - -
customers-intermediate 8 103243 103312 69 - - -
customers-intermediate 9 84211 102772 18561 - - -
events 15 11958555 15231834 3273279 - - -
events 3 1393386 16534651 15141265 - - -
events 4 1149540 15390069 14240529 - - -
visitors 15 2774874 2778873 3999 - - -
visitors 3 603242 603324 82 - - -
visitors 4 565266 565834 568
流媒体应用程序也重新启动，之后我可以看到一些处理日志大约20小时，然后停止处理。
已经有两天了。但也值得一提的是，您在上面看到的所有主题都有16个分区，但有些显示其中三个分区（访问者，事件）。但是，我可以描述这些主题，并且它们的分区像往常一样分布良好，在那里我找不到任何奇怪的地方。
可能会发生什么？
重新启动应用程序后，我可以再次看到所有分区，并且应用程序正在使用主题分区。但是，许多（大多数）分区丢失了偏移量。自从我更改了offset.retention.minutes设置以来，就不会发生这种情况。
events 0 - 14538261 - stream-processor-01f7ecea-4e50-4505-a8e7-8c536058b7bc-StreamThread-1-consumer-a8a6e989-d6c1-472f-aec5-3ae637d87b9e /ip2 stream-processor-01f7ecea-4e50-4505-a8e7-8c536058b7bc-StreamThread-1-consumer
events 1 49070 13276094 13227024 stream-processor-44fefb47-23b1-4597-8a6b-d7f1c364c316-StreamThread-1-consumer-69da45eb-b0b2-4ad8-a831-8aecf6849892 /ip1 stream-processor-44fefb47-23b1-4597-8a6b-d7f1c364c316-StreamThread-1-consumer
events 10 - 15593746 - stream-processor-44fefb47-23b1-4597-8a6b-d7f1c364c316-StreamThread-1-consumer-69da45eb-b0b2-4ad8-a831-8aecf6849892 /ip1 stream-processor-44fefb47-23b1-4597-8a6b-d7f1c364c316-StreamThread-1-consumer
events 11 - 15525487 - stream-processor-44fefb47-23b1-4597-8a6b-d7f1c364c316-StreamThread-1-consumer-69da45eb-b0b2-4ad8-a831-8aecf6849892 /ip1 stream-processor-44fefb47-23b1-4597-8a6b-d7f1c364c316-StreamThread-1-consumer
events 12 - 21863908 - stream-processor-44fefb47-23b1-4597-8a6b-d7f1c364c316-StreamThread-1-consumer-69da45eb-b0b2-4ad8-a831-8aecf6849892 /ip1 stream-processor-44fefb47-23b1-4597-8a6b-d7f1c364c316-StreamThread-1-consumer
events 13 - 15810925 - stream-processor-44fefb47-23b1-4597-8a6b-d7f1c364c316-StreamThread-1-consumer-69da45eb-b0b2-4ad8-a831-8aecf6849892 /ip1 stream-processor-44fefb47-23b1-4597-8a6b-d7f1c364c316-StreamThread-1-consumer
events 14 - 13509742 - stream-processor-44fefb47-23b1-4597-8a6b-d7f1c364c316-StreamThread-1-consumer-69da45eb-b0b2-4ad8-a831-8aecf6849892 /ip1 stream-processor-44fefb47-23b1-4597-8a6b-d7f1c364c316-StreamThread-1-consumer
events 15 11958555 15231834 3273279 stream-processor-01f7ecea-4e50-4505-a8e7-8c536058b7bc-StreamThread-1-consumer-a8a6e989-d6c1-472f-aec5-3ae637d87b9e /ip2 stream-processor-01f7ecea-4e50-4505-a8e7-8c536058b7bc-StreamThread-1-consumer
...
Kafka 1.1.0，Kafka Streams 1.0