服务器之家:专注于VPS、云服务器配置技术及软件下载分享
分类导航

PHP教程|ASP.NET教程|Java教程|ASP教程|编程技术|正则表达式|C/C++|IOS|C#|Swift|Android|VB|R语言|JavaScript|易语言|vb.net|

服务器之家 - 编程语言 - Java教程 - elasticsearch集群cluster discovery可配式模块示例分析

elasticsearch集群cluster discovery可配式模块示例分析

2022-11-16 15:01zziawan Java教程

这篇文章主要为大家介绍了elasticsearch集群cluster discovery可配式模块示例分析,有需要的朋友可以借鉴参考下,希望能够有所帮助,祝大家多多进步,早日升职加薪

前言

elasticsearch cluster实现了自己发现机制zen。Discovery功能主要包括以下几部分内容:master选举,master错误探测,集群中其它节点探测,单播多播ping。本篇会首先概述以下Discovery这一部分的功能,然后介绍节点检测。其它内容会在接下来介绍。

Discovery模块的概述

discovery是可配式模块,官方支持亚马逊的Azure discovery,Google Compute Engine,EC2 Discovery三种发现机制,根据插件规则完全可以自己实现其它的发现机制。整个模块通过实现guice的DiscoveryModule对外提供模块的注册和启动, 默认使用zen discovery。发现模块对外接口为DiscoveryService,它的方法如下所示:

elasticsearch集群cluster discovery可配式模块示例分析

它本质上是discovery的一个代理,所有的功能最终都是由所绑定的discovery所实现的。节点启动时通过DiscoveryModule获取DiscoveryService,然后启动DiscoveryService,DiscoveryService启动绑定的Discovery,整个功能模块就完成了加载和启动。这也是elasticsearch所有模块的实现方式,通过module对外提供绑定和获取,通过service接口对外提供模块的功能,在后面的分析中会经常遇到。

cluster节点探测

接下来分析cluster的一个重要功能就是节点探测。cluster中不能没有master节点,因此集群中所有节点都要周期探测master节点,一旦无法检测到,将会进行master选举。同时作为master,对于节点变动也要时刻关注,因此它需要周期性探测集群中所有节点,确保及时剔除已经宕机的节点。这种相互间的心跳检测就是cluster的faultdetection。下图是faultdetection的继承关系:

elasticsearch集群cluster discovery可配式模块示例分析

有两种实现方式,分别是master探测集群中其它节点和其它节点对master节点的探测。

FaultDetection只要一个抽象方法handleTransportDisconnect,该方法在内部类FDConnectionListener中被调用。在elasticsearch中大量使用了listener的异步方式,异步可以极大提升系统性能。它的代码如下所示:

?
1
2
3
4
5
6
7
8
9
private class FDConnectionListener implements TransportConnectionListener {
        @Override
        public void onNodeConnected(DiscoveryNode node) {
        }
        @Override
        public void onNodeDisconnected(DiscoveryNode node) {
            handleTransportDisconnect(node);
        }
    }

faultdetection启动时会注册相应的FDConnetionListener,当探测到节点丢失,会通过onNodeDisconnected方法回调对于的handleTransportDisconnect进行处理。

MasterFaultDetection的启动代码

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
private void innerStart(final DiscoveryNode masterNode) {
        this.masterNode = masterNode;
        this.retryCount = 0;
        this.notifiedMasterFailure.set(false);
        // 尝试连接master节点
        try {
            transportService.connectToNode(masterNode);
        } catch (final Exception e) {
            // 连接失败通知masterNode失败
            notifyMasterFailure(masterNode, "failed to perform initial connect [" + e.getMessage() + "]");
            return;
        }
    //关闭之前的masterping,重启新的masterping
        if (masterPinger != null) {
            masterPinger.stop();
        }
        this.masterPinger = new MasterPinger();
        // 周期之后启动masterPing,这里并没有周期启动masterPing,只是设定了延迟时间。
        threadPool.schedule(pingInterval, ThreadPool.Names.SAME, masterPinger);
    }

代码有有详细注释,就不再过多解释。

master连接失败的逻辑

代码如下:

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
private void notifyMasterFailure(final DiscoveryNode masterNode, final String reason) {
        if (notifiedMasterFailure.compareAndSet(false, true)) {
            threadPool.generic().execute(new Runnable() {
                @Override
                public void run() {
            //通知所有listener master丢失
                    for (Listener listener : listeners) {
                        listener.onMasterFailure(masterNode, reason);
                    }
                }
            });
            stop("master failure, " + reason);
        }
    }

 在ZenDiscovery中实现了listener.onMasterFailure接口。会进行master丢失的相关处理,在后面再分析。

MasterPing的关键代码

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
private class MasterPinger implements Runnable {
        private volatile boolean running = true;
        public void stop() {
            this.running = false;
        }
        @Override
        public void run() {
            if (!running) {
                // return and don't spawn...
                return;
            }
            final DiscoveryNode masterToPing = masterNode;
   final MasterPingRequest request = new MasterPingRequest(clusterService.localNode().id(), masterToPing.id(), clusterName);
            final TransportRequestOptions options = options().withType(TransportRequestOptions.Type.PING).withTimeout(pingRetryTimeout);
            transportService.sendRequest(masterToPing, MASTER_PING_ACTION_NAME, request, options, new BaseTransportResponseHandler<MasterPingResponseResponse>() {
                        @Override
                        public MasterPingResponseResponse newInstance() {
                            return new MasterPingResponseResponse();
                        }
                        @Override
                        public void handleResponse(MasterPingResponseResponse response) {
                            if (!running) {
                                return;
                            }
                            // reset the counter, we got a good result
                            MasterFaultDetection.this.retryCount = 0;
                            // check if the master node did not get switched on us..., if it did, we simply return with no reschedule
                            if (masterToPing.equals(MasterFaultDetection.this.masterNode())) {
                                // 启动新的ping周期
                                threadPool.schedule(pingInterval, ThreadPool.Names.SAME, MasterPinger.this);
                            }
                        }
                        @Override
                        public void handleException(TransportException exp) {
                            if (!running) {
                                return;
                            }
                            synchronized (masterNodeMutex) {
                                // check if the master node did not get switched on us...
                                if (masterToPing.equals(MasterFaultDetection.this.masterNode())) {
                                    if (exp instanceof ConnectTransportException || exp.getCause() instanceof ConnectTransportException) {
                                        handleTransportDisconnect(masterToPing);
                                        return;
                                    } else if (exp.getCause() instanceof NoLongerMasterException) {
                                        logger.debug("[master] pinging a master {} that is no longer a master", masterNode);
                                        notifyMasterFailure(masterToPing, "no longer master");
                                        return;
                                    } else if (exp.getCause() instanceof NotMasterException) {
                                        logger.debug("[master] pinging a master {} that is not the master", masterNode);
                                        notifyMasterFailure(masterToPing, "not master");
                                        return;
                                    } else if (exp.getCause() instanceof NodeDoesNotExistOnMasterException) {
                                        logger.debug("[master] pinging a master {} but we do not exists on it, act as if its master failure", masterNode);
                                        notifyMasterFailure(masterToPing, "do not exists on master, act as master failure");
                                        return;
                                    }
                                    int retryCount = ++MasterFaultDetection.this.retryCount;
                                    logger.trace("[master] failed to ping [{}], retry [{}] out of [{}]", exp, masterNode, retryCount, pingRetryCount);
                                    if (retryCount >= pingRetryCount) {
                                        logger.debug("[master] failed to ping [{}], tried [{}] times, each with maximum [{}] timeout", masterNode, pingRetryCount, pingRetryTimeout);
                                        // not good, failure
                                        notifyMasterFailure(masterToPing, "failed to ping, tried [" + pingRetryCount + "] times, each with  maximum [" + pingRetryTimeout + "] timeout");
                                    } else {
                                         // resend the request, not reschedule, rely on send timeout
                                        transportService.sendRequest(masterToPing, MASTER_PING_ACTION_NAME, request, options, this);
                                    }
                                }
                            }
                        }
            );
        }
    }

MasterPing是一个线程,在innerStart的方法中没有设定周期启动masterping,但是masterping需要周期进行,这个秘密就在run 方法中,如果ping成功就会重启一个新的ping。这样既保证了ping线程的唯一性同时也保证了ping的顺序和间隔。

ping的方式跟之前一样是也是通过transport发送一个masterpingrequest,进行一个连接。节点收到该请求后,如果已不再是master会抛出NotMasterException,状态更新出差会抛出其它异常,异常会通过。否则会正常响应notifyMasterFailure方法处理跟启动逻辑一样。

对于网络问题导致的无响应情况,会调用handleTransportDisconnect(masterToPing)方法处理。masterfaultDetection对该方法的实现如下:

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
protected void handleTransportDisconnect(DiscoveryNode node) {
    //这里需要同步
        synchronized (masterNodeMutex) {
        //master 已经换成其它节点,就没必要再连接
            if (!node.equals(this.masterNode)) {
                return;
            }
            if (connectOnNetworkDisconnect) {
                try {
            //尝试再次连接
                    transportService.connectToNode(node);
                    // if all is well, make sure we restart the pinger
                    if (masterPinger != null) {
                        masterPinger.stop();
                    }
            //连接成功启动新的masterping
                    this.masterPinger = new MasterPinger();
                    // we use schedule with a 0 time value to run the pinger on the pool as it will run on later
                    threadPool.schedule(TimeValue.timeValueMillis(0), ThreadPool.Names.SAME, masterPinger);
                } catch (Exception e) {
            //连接出现异常,启动master节点丢失通知
                    logger.trace("[master] [{}] transport disconnected (with verified connect)", masterNode);
                    notifyMasterFailure(masterNode, "transport disconnected (with verified connect)");
                }
            } else {
          //不需要重连,通知master丢失。
                logger.trace("[master] [{}] transport disconnected", node);
                notifyMasterFailure(node, "transport disconnected");
            }
        }
    }

这就是masterfaultDetection的整个流程:启动中如果master丢失则通知节点丢失,否则在一定延迟(3s)后启动masterping,masterping线程尝试连接master节点,如果master节点网络失联,尝试再次连接。master节点收到masterpingrequest后首先看一下自己还是不是master,如果不是则抛出异常,否则正常回应。节点如果收到响应是异常则启动master丢失通知,否则此次ping结束。在一定延迟后启动新的masterping线程。

NodeFaultDetection的逻辑跟实现上跟MasterFualtDetetion相似,区别主要在于ping异常处理上。当某个节点出现异常或者没有响应时,会启动节点丢失机制,只是受到通知后的处理逻辑不通。就不再详细分析,有兴趣可以参考具体代码,希望大家以后多多支持服务器之家!

原文链接:https://www.cnblogs.com/zziawanblog/p/6533731.html

延伸 · 阅读

精彩推荐