NIFI 安全集群：设置 2 个节点，但仅检测到 1 个节点（Apache Nifi 1.11.4）答案

【问题标题】：NIFI secured cluster : Setup 2 nodes but only 1 node detected (Apache Nifi 1.11.4)NIFI 安全集群：设置 2 个节点，但仅检测到 1 个节点（Apache Nifi 1.11.4）
【发布时间】：2020-06-26 09:52:22
【问题描述】：

我需要使用 2 个节点设置安全的 nifi 集群。 我使用的是最新版本的 nifi：1.11.4

我以为我成功构建了集群，因为 UI 成功打开了两个 https url，如果我在其中一个中执行某些操作（例如：放置新处理器），另一个成功同步。但是突然我意识到总节点只检测到1个！另一个在哪里？？？这是证明：

both webs Successfully open and sync

我已经正确设置了所有配置文件（authorizers.xml、nifi.properties、zookeeper、statemanagement、ssl 等）。

您可以通过以下链接查看配置和登录：

NODE 1 配置 - bihdpapp399

bihdpapp399

NODE 2 配置 - bihdpapp394

bihdpapp394

顺便说一句，你知道我使用 2 个节点，但是根据日志，虽然我已经为两个节点设置了最大选举 conf = 2，但另一个节点不想投票选举，我也尝试过增加选举时间从 1m 到 5m，但仍然没有来自其他节点的投票：

2020-06-25 12:08:16,909 INFO [main] o.a.nifi.controller.StandardFlowService Requested by cluster coordinator to retry connection in 5 seconds with explanation: Cluster is still voting on which Flow is the correct flow for the cluster. Election will complete in 59 seconds or after 1 more vote is cast, whichever occurs first.
2020-06-25 12:08:22,012 INFO [main] o.a.nifi.controller.StandardFlowService Requested by cluster coordinator to retry connection in 5 seconds with explanation: Cluster is still voting on which Flow is the correct flow for the cluster. Election will complete in 54 seconds or after 1 more vote is cast, whichever occurs first.
2020-06-25 12:08:27,116 INFO [main] o.a.nifi.controller.StandardFlowService Requested by cluster coordinator to retry connection in 5 seconds with explanation: Cluster is still voting on which Flow is the correct flow for the cluster. Election will complete in 49 seconds or after 1 more vote is cast, whichever occurs first.
2020-06-25 12:08:32,216 INFO [main] o.a.nifi.controller.StandardFlowService Requested by cluster coordinator to retry connection in 5 seconds with explanation: Cluster is still voting on which Flow is the correct flow for the cluster. Election will complete in 44 seconds or after 1 more vote is cast, whichever occurs first.
2020-06-25 12:08:37,317 INFO [main] o.a.nifi.controller.StandardFlowService Requested by cluster coordinator to retry connection in 5 seconds with explanation: Cluster is still voting on which Flow is the correct flow for the cluster. Election will complete in 39 seconds or after 1 more vote is cast, whichever occurs first.
2020-06-25 12:08:42,418 INFO [main] o.a.nifi.controller.StandardFlowService Requested by cluster coordinator to retry connection in 5 seconds with explanation: Cluster is still voting on which Flow is the correct flow for the cluster. Election will complete in 34 seconds or after 1 more vote is cast, whichever occurs first.

【问题讨论】：

我认为您已经部署了两个单节点集群（即两个不同的集群，每个集群都有一个节点）。配置需要指向同一个 ZooKeeper 集群并注册为同一个集群的成员。
嗨@Andy，感谢您的回复，我已经为每个节点调整了相同的zookeeper配置，当我从主菜单中看到集群信息时，它们都只显示一个相同的节点，所以我们可以确保它在同一个集群中，对吧？有关更多详细信息，您可以从上面下载它。请帮助兄弟，我已经坚持了大约 1 周:'(

标签： java hadoop etl apache-nifi

【解决方案1】：

$NIFI_Directory/state/local 文件夹存储集群的状态，即使在添加节点后也会使用它。删除或重命名状态文件夹并重新启动 nifi 集群。这帮助我解决了这个问题。

【讨论】：