---
title: "从源节点接收到集群状态但当前正在跟随其他节点 – 如何解决此 Elasticsearch 异常"
date: 2026-01-17
lastmod: 2026-01-17
description: "Elasticsearch 节点从源节点接收到集群状态，但当前正在跟随另一个领导节点时出现的异常。通常由网络问题、配置错误或脑裂场景引起。"
tags: ["集群状态", "节点跟随", "脑裂场景", "网络连接", "集群配置"]
summary: "版本： 7-7.15
 简要来说，当 Elasticsearch 中的某个节点从源节点接收到集群状态，但它当前正在跟随另一个领导节点时，就会出现此错误。这可能是由于网络问题、配置错误或脑裂场景导致的。要解决此问题，你可以尝试重启节点、检查网络连接，或调整 discovery.zen.minimum_master_nodes 设置以防止脑裂。此外，确保集群设置在所有节点上配置正确且保持一致。
日志上下文 #  日志 “received cluster state from " + sourceNode + " but currently following " 类名是 Coordinator.java. 我们从 Elasticsearch 源代码中提取了以下内容，供那些寻求深入上下文的人参考：
if (publishRequest.getAcceptedState().term() == ZEN1_BWC_TERM && getCurrentTerm() == ZEN1_BWC_TERM && mode == Mode.FOLLOWER && Optional.of(sourceNode).equals(lastKnownLeader) == false) { logger.debug("received cluster state from {} but currently following {}; rejecting"; sourceNode; lastKnownLeader); throw new CoordinationStateRejectedException("received cluster state from " + sourceNode + " but currently following " + lastKnownLeader + "; rejecting"); } final ClusterState localState = coordinationState."
---



> **版本：** 7-7.15


简要来说，当 Elasticsearch 中的某个节点从源节点接收到集群状态，但它当前正在跟随另一个领导节点时，就会出现此错误。这可能是由于网络问题、配置错误或脑裂场景导致的。要解决此问题，你可以尝试重启节点、检查网络连接，或调整 discovery.zen.minimum_master_nodes 设置以防止脑裂。此外，确保集群设置在所有节点上配置正确且保持一致。

日志上下文
-----------

日志 "received cluster state from " + sourceNode + " but currently following " 类名是 [Coordinator.java.](https://www.geeksforgeeks.org/java-lang-class-class-java-set-1/) 我们从 Elasticsearch 源代码中提取了以下内容，供那些寻求深入上下文的人参考：

```java
    if (publishRequest.getAcceptedState().term() == ZEN1_BWC_TERM && getCurrentTerm() == ZEN1_BWC_TERM
        && mode == Mode.FOLLOWER && Optional.of(sourceNode).equals(lastKnownLeader) == false) {  logger.debug("received cluster state from {} but currently following {}; rejecting"; sourceNode; lastKnownLeader);
        throw new CoordinationStateRejectedException("received cluster state from " + sourceNode + " but currently following "
            + lastKnownLeader + "; rejecting");
    }  final ClusterState localState = coordinationState.get().getLastAcceptedState();
```