http://www.cnblogs.com/codingexperience/p/5939059.html
一、心跳机制简介
在分布式系统中,分布在不同主机上的节点需要检测其他节点的状态,如服务器节点需要检测从节点是否失效。为了检测对方节点的有效性,每隔固定时间就发送一个固定信息给对方,对方回复一个固定信息,如果长时间没有收到对方的回复,则断开与对方的连接。
发包方既可以是服务端,也可以是客户端,这要看具体实现。因为是每隔固定时间发送一次,类似心跳,所以发送的固定信息称为心跳包。心跳包一般为比较小的包,可根据具体实现。心跳包主要应用于长连接的保持与短线链接。
一般而言,应该客户端主动向服务器发送心跳包,因为服务器向客户端发送心跳包会影响服务器的性能。
二、心跳机制实现方式
心跳机制有两种实现方式,一种基于TCP自带的心跳包,TCP的SO_KEEPALIVE选项可以,系统默认的默认跳帧频率为2小时,超过2小时后,本地的TCP 实现会发送一个数据包给远程的 Socket. 如果远程Socket 没有发回响应, TCP实现就会持续尝试 11 分钟, 直到接收到响应为止。 否则就会自动断开Socket连接。但TCP自带的心跳包无法检测比较敏感地知道对方的状态,默认2小时的空闲时间,对于大多数的应用而言太长了。可以手工开启KeepAlive功能并设置合理的KeepAlive参数。
另一种在应用层自己进行实现,基本步骤如下:
- Client使用定时器,不断发送心跳;
- Server收到心跳后,回复一个包;
- Server为每个Client启动超时定时器,如果在指定时间内没有收到Client的心跳包,则Client失效。
三、Java实现心跳机制
这里基于Java实现的简单RPC框架实现心跳机制。Java实现代码如下所示:
心跳客户端类:
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
|
public
class
HeartbeatClient
implements
Runnable {
private
String serverIP =
"127.0.0.1"
;
private
int
serverPort =
8089
;
private
String nodeID = UUID.randomUUID().toString();
private
boolean
isRunning =
true
;
// 最近的心跳时间
private
long
lastHeartbeat;
// 心跳间隔时间
private
long
heartBeatInterval =
10
*
1000
;
public
void
run() {
try
{
while
(isRunning) {
HeartbeatHandler handler = RPClient.getRemoteProxyObj(HeartbeatHandler.
class
,
new
InetSocketAddress(serverIP, serverPort));
long
startTime = System.currentTimeMillis();
// 是否达到发送心跳的周期时间
if
(startTime - lastHeartbeat > heartBeatInterval) {
System.out.println(
"send a heart beat"
);
lastHeartbeat = startTime;
HeartbeatEntity entity =
new
HeartbeatEntity();
entity.setTime(startTime);
entity.setNodeID(nodeID);
// 向服务器发送心跳,并返回需要执行的命令
Cmder cmds = handler.sendHeartBeat(entity);
if
(!processCommand(cmds))
continue
;
}
}
}
catch
(Exception e) {
e.printStackTrace();
}
}
private
boolean
processCommand(Cmder cmds) {
// ...
return
true
;
}
}
|
心跳包实体类:
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
|
public
class
HeartbeatEntity
implements
Serializable {
private
long
time;
private
String nodeID;
private
String error;
private
Map<String, Object> info =
new
HashMap<String, Object>();
public
String getNodeID() {
return
nodeID;
}
public
void
setNodeID(String nodeID) {
this
.nodeID = nodeID;
}
public
String getError() {
return
error;
}
public
void
setError(String error) {
this
.error = error;
}
public
Map<String, Object> getInfo() {
return
info;
}
public
void
setInfo(Map<String, Object> info) {
this
.info = info;
}
public
long
getTime() {
return
time;
}
public
void
setTime(
long
time) {
this
.time = time;
}
}
|
服务器接受心跳包返回的命令对象类:
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
|
public
class
Cmder
implements
Serializable {
private
String nodeID;
private
String error;
private
Map<String, Object> info =
new
HashMap<String, Object>();
public
String getNodeID() {
return
nodeID;
}
public
void
setNodeID(String nodeID) {
this
.nodeID = nodeID;
}
public
String getError() {
return
error;
}
public
void
setError(String error) {
this
.error = error;
}
public
Map<String, Object> getInfo() {
return
info;
}
public
void
setInfo(Map<String, Object> info) {
this
.info = info;
}
}
|
RPC服务注册中心:
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
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
|
public
class
ServiceCenter {
private
ExecutorService executor = Executors.newFixedThreadPool(
20
);
private
final
ConcurrentHashMap<String, Class> serviceRegistry =
new
ConcurrentHashMap<String, Class>();
private
AtomicBoolean isRunning =
new
AtomicBoolean(
true
);
// 服务器监听端口
private
int
port =
8089
;
// 心跳监听器
HeartbeatLinstener linstener;
// 单例模式
private
static
class
SingleHolder {
private
static
final
ServiceCenter INSTANCE =
new
ServiceCenter();
}
private
ServiceCenter() {
}
public
static
ServiceCenter getInstance() {
return
SingleHolder.INSTANCE;
}
public
void
register(Class serviceInterface, Class impl) {
System.out.println(
"regeist service "
+ serviceInterface.getName());
serviceRegistry.put(serviceInterface.getName(), impl);
}
public
void
start()
throws
IOException {
ServerSocket server =
new
ServerSocket();
server.bind(
new
InetSocketAddress(port));
System.out.println(
"start server"
);
linstener = HeartbeatLinstener.getInstance();
System.out.println(
"start listen heart beat"
);
try
{
while
(
true
) {
// 1.监听客户端的TCP连接,接到TCP连接后将其封装成task,由线程池执行
executor.execute(
new
ServiceTask(server.accept()));
}
}
finally
{
server.close();
}
}
public
void
stop() {
isRunning.set(
false
);
executor.shutdown();
}
public
boolean
isRunning() {
return
isRunning.get();
}
public
int
getPort() {
return
port;
}
public
void
settPort(
int
port) {
this
.port = port;
}
public
ConcurrentHashMap<String, Class> getServiceRegistry() {
return
serviceRegistry;
}
private
class
ServiceTask
implements
Runnable {
Socket clent =
null
;
public
ServiceTask(Socket client) {
this
.clent = client;
}
public
void
run() {
ObjectInputStream input =
null
;
ObjectOutputStream output =
null
;
try
{
// 2.将客户端发送的码流反序列化成对象,反射调用服务实现者,获取执行结果
input =
new
ObjectInputStream(clent.getInputStream());
String serviceName = input.readUTF();
String methodName = input.readUTF();
Class<?>[] parameterTypes = (Class<?>[]) input.readObject();
Object[] arguments = (Object[]) input.readObject();
Class serviceClass = serviceRegistry.get(serviceName);
if
(serviceClass ==
null
) {
throw
new
ClassNotFoundException(serviceName +
" not found"
);
}
Method method = serviceClass.getMethod(methodName, parameterTypes);
Object result = method.invoke(serviceClass.newInstance(), arguments);
// 3.将执行结果反序列化,通过socket发送给客户端
output =
new
ObjectOutputStream(clent.getOutputStream());
output.writeObject(result);
}
catch
(Exception e) {
e.printStackTrace();
}
finally
{
if
(output !=
null
) {
try
{
output.close();
}
catch
(IOException e) {
e.printStackTrace();
}
}
if
(input !=
null
) {
try
{
input.close();
}
catch
(IOException e) {
e.printStackTrace();
}
}
if
(clent !=
null
) {
try
{
clent.close();
}
catch
(IOException e) {
e.printStackTrace();
}
}
}
}
}
}
|
心跳监听类:
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
73
74
75
76
77
78
79
80
|
package
com.cang.heartbeat;
import
java.io.IOException;
import
java.io.ObjectInputStream;
import
java.io.ObjectOutputStream;
import
java.lang.reflect.Method;
import
java.net.InetSocketAddress;
import
java.net.ServerSocket;
import
java.net.Socket;
import
java.util.Iterator;
import
java.util.Map;
import
java.util.concurrent.ConcurrentHashMap;
import
java.util.concurrent.ExecutorService;
import
java.util.concurrent.Executors;
import
java.util.concurrent.atomic.AtomicBoolean;
/**
* 心跳监听保存信息
*
* @author cang
* @create_time 2016-09-28 11:40
*/
public
class
HeartbeatLinstener {
private
ExecutorService executor = Executors.newFixedThreadPool(
20
);
private
final
ConcurrentHashMap<String, Object> nodes =
new
ConcurrentHashMap<String, Object>();
private
final
ConcurrentHashMap<String, Long> nodeStatus =
new
ConcurrentHashMap<String, Long>();
private
long
timeout =
10
*
1000
;
// 服务器监听端口
private
int
port =
8089
;
// 单例模式
private
static
class
SingleHolder {
private
static
final
HeartbeatLinstener INSTANCE =
new
HeartbeatLinstener();
}
private
HeartbeatLinstener() {
}
public
static
HeartbeatLinstener getInstance() {
return
SingleHolder.INSTANCE;
}
public
ConcurrentHashMap<String, Object> getNodes() {
return
nodes;
}
public
void
registerNode(String nodeId, Object nodeInfo) {
nodes.put(nodeId, nodeInfo);
nodeStatus.put(nodeId, System.currentTimeMillis());
}
public
void
removeNode(String nodeID) {
if
(nodes.containsKey(nodeID)) {
nodes.remove(nodeID);
}
}
// 检测节点是否有效
public
boolean
checkNodeValid(String key) {
if
(!nodes.containsKey(key) || !nodeStatus.containsKey(key))
return
false
;
if
((System.currentTimeMillis() - nodeStatus.get(key)) > timeout)
return
false
;
return
true
;
}
// 删除所有失效节点
public
void
removeInValidNode() {
Iterator<Map.Entry<String, Long>> it = nodeStatus.entrySet().iterator();
while
(it.hasNext()) {
Map.Entry<String, Long> e = it.next();
if
((System.currentTimeMillis() - nodeStatus.get(e.getKey())) > timeout) {
nodes.remove(e.getKey());
}
}
}
}
|
心跳处理类接口:
1
2
3
|
public
interface
HeartbeatHandler {
public
Cmder sendHeartBeat(HeartbeatEntity info);
}
|
心跳处理实现类:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
|
public
class
HeartbeatHandlerImpl
implements
HeartbeatHandler {
public
Cmder sendHeartBeat(HeartbeatEntity info) {
HeartbeatLinstener linstener = HeartbeatLinstener.getInstance();
// 添加节点
if
(!linstener.checkNodeValid(info.getNodeID())) {
linstener.registerNode(info.getNodeID(), info);
}
// 其他操作
Cmder cmder =
new
Cmder();
cmder.setNodeID(info.getNodeID());
// ...
System.out.println(
"current all the nodes: "
);
Map<String, Object> nodes = linstener.getNodes();
for
(Map.Entry e : nodes.entrySet()) {
System.out.println(e.getKey() +
" : "
+ e.getValue());
}
System.out.println(
"hadle a heartbeat"
);
return
cmder;
}
}
|
测试类:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
|
public
class
HeartbeatTest {
public
static
void
main(String[] args) {
new
Thread(
new
Runnable() {
public
void
run() {
try
{
ServiceCenter serviceServer = ServiceCenter.getInstance();
serviceServer.register(HeartbeatHandler.
class
, HeartbeatHandlerImpl.
class
);
serviceServer.start();
}
catch
(IOException e) {
e.printStackTrace();
}
}
}).start();
Thread client1 =
new
Thread(
new
HeartbeatClient());
client1.start();
Thread client2 =
new
Thread(
new
HeartbeatClient());
client2.start();
}
}
|
四、总结
上面的代码还有很多不足的地方,希望有空能进行改善:
- 配置为硬编码;
- 命令类Cmder没有实际实现,返回的Cmder对象没有实际进行处理;
其他小问题就暂时不管了,希望以后能重写上面的代码。