.NET Core微服務之路：基於Consul最少集群實現服務的注冊與發現（二）

重溫Consul最少化集群的搭建

 

　　我們再復習一下上一篇的內容，先建立三台consul server節點，兩個consul client節點，分別在每個節點上跑不同（名稱不同而已）的實例。我們先通過vmware啟動這五個節點，並且能成功訪問這個兩個client節點的實例。（具體配置可以見上一篇）

 

　　通過配置文件自動生成服務

{
  "services": [
    {
      "id": "CLIENT_SERVICE_01",
      "name": "CAS Client Service",
      "tags": [
        "urlprefix-/ClientService01"
      ],
      "address": "192.168.153.132",
      "port": 5000,
      "checks": [
        {
          "name": "clientservice_check01",
          "http": "http://192.168.53.132:5000/api/health",
          "interval": "10s",
          "timeout": "5s"
        }
      ]
    },
    {
      "id": "CLIENT_SERVICE_02",
      "name": "CAS Client Service",
      "tags": [
        "urlprefix-/ClientService02"
      ],
      "address": "192.168.153.132",
      "port": 5001,
      "checks": [
        {
          "name": "clientservice_check02",
          "http": "http://192.168.153.132:5001/api/health",
          "interval": "10s",
          "timeout": "5s"
        }
      ]
    }
  ]
}

 

添加KEY/VALUE　　

curl -X PUT -d 'edisonchou' http://192.168.80.100:8500/v1/kv/web/vhallaccount

　　我們通過 http://192.168.153.129:8500/v1/kv/?recurse來查看這個值是否添加到Consul中

 　　再驗證是否已經同步到其他Consul服務端，先看129，注意加粗字體

[root@localhost ~]# ./consul kv get web/vhallaccount
stevelee
[root@localhost ~]# ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: ens32: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast state DOWN group default qlen 1000
    link/ether 00:0c:29:a6:a1:1c brd ff:ff:ff:ff:ff:ff
3: ens34: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether 00:0c:29:a6:a1:26 brd ff:ff:ff:ff:ff:ff
    inet 192.168.153.129/24 brd 192.168.153.255 scope global noprefixroute ens34
       valid_lft forever preferred_lft forever
    inet6 fe80::6d21:aa51:2262:b80f/64 scope link noprefixroute
       valid_lft forever preferred_lft forever

 　　再看130的Consul服務端

[root@localhost ~]# ./consul kv get web/vhallaccount
stevelee
[root@localhost ~]# ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: ens32: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether 00:0c:29:2c:c2:fc brd ff:ff:ff:ff:ff:ff
    inet 192.168.153.130/24 brd 192.168.153.255 scope global noprefixroute ens32
       valid_lft forever preferred_lft forever
    inet6 fe80::6114:ed9c:c49d:649b/64 scope link noprefixroute
       valid_lft forever preferred_lft forever
3: ens34: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether 00:0c:29:2c:c2:06 brd ff:ff:ff:ff:ff:ff
    inet 192.168.153.170/24 brd 192.168.153.255 scope global noprefixroute dynamic ens34
       valid_lft 1216sec preferred_lft 1216sec
    inet6 fe80::a67c:7966:8767:27fb/64 scope link noprefixroute
       valid_lft forever preferred_lft forever

 

 　　我們也可以通過Consul WEB UI來編輯KV值。

 

 Consul服務的Watch機制

 　　熔斷保護在Consul和Ocelot中都有實現，意思就是當一個服務不正常時（比如我們的一個服務實例掛了，Consul的健康檢查機制檢測到了），應該給系統維護人員給以告警。在Consul中，服務告警也是通過配置文件來實現的。

{
  "watches": [
    {
      "type": "checks",
      "handler_type": "http",
      "state": "critical",
      "http_handler_config": {
        "path": "http://192.168.153.132:9000/notice",
        "method": "POST",
        "timeout": "10s",
        "header": { "Authorization": [ "token" ] }
      }
    }
  ]
}

　　我們再新建一個項目，建立一個控制器，用默認的HomController控制器也可以，鍵入如下代碼：

[Produces("application/json")]
public class HomeController : Controller
{
    [HttpPost]
    [Route("/notice")]
    public IActionResult Notice()
    {
        var stream = HttpContext.Request.Body;
        if (HttpContext.Request.ContentLength != null)
        {
            var buffer = new byte[HttpContext.Request.ContentLength.Value];
            stream.Read(buffer, 0, buffer.Length);
            var content = Encoding.UTF8.GetString(buffer);

            var path = $"{AppDomain.CurrentDomain.BaseDirectory}{DateTime.Now:hh_mm_ss_ffff}.log";
            if (!System.IO.File.Exists(path))
            {
                System.IO.File.Create(path).Close();
            }

            using (var sw = new StreamWriter(path))
            {
                sw.Write(content);
                sw.Flush();
                sw.Close();
            }

            return Ok();
        }

        throw new Exception("post is null");
    }
}

　　功能很簡單，做一個接受消息的客戶端，POST方法，代碼不解釋。

　　當我們把5000端口上的服務停掉，會出現什么樣的情況呢？132Consul的客戶端服務器就會出現:5000/api/health無法訪問的問題。

2018/10/17 04:03:10 [WARN] agent: Check "service:CLIENT_SERVICE_01" HTTP request failed: Get http://192.168.153.132:5000/api/health: dial tcp 192.168.153.132:5000: connect: connection refused
2018/10/17 04:03:10 [INFO] agent: Synced check "service:CLIENT_SERVICE_01"
2018/10/17 04:03:20 [WARN] agent: Check "service:CLIENT_SERVICE_01" HTTP request failed: Get http://192.168.153.132:5000/api/health: dial tcp 192.168.153.132:5000: connect: connection refused
2018/10/17 04:03:30 [WARN] agent: Check "service:CLIENT_SERVICE_01" HTTP request failed: Get http://192.168.153.132:5000/api/health: dial tcp 192.168.153.132:5000: connect: connection refused
2018/10/17 04:03:40 [WARN] agent: Check "service:CLIENT_SERVICE_01" HTTP request failed: Get http://192.168.153.132:5000/api/health: dial tcp 192.168.153.132:5000: connect: connection refused
2018/10/17 04:03:50 [WARN] agent: Check "service:CLIENT_SERVICE_01" HTTP request failed: Get http://192.168.153.132:5000/api/health: dial tcp 192.168.153.132:5000: connect: connection refused
2018/10/17 04:04:00 [WARN] agent: Check "service:CLIENT_SERVICE_01" HTTP request failed: Get http://192.168.153.132:5000/api/health: dial tcp 192.168.153.132:5000: connect: connection refused

　　Watch機制會每隔10s向http://192.168.153.132:9000/notice發送一組json格式的消息，內容如下

[{
    "Node": "LZZ.DEV.WebServer",
    "CheckID": "service:CLIENT_SERVICE_01",
    "Name": "clientservice_check01",
    "Status": "critical",
    "Notes": "",
    "Output": "Get http://192.168.153.132:5000/api/health: dial tcp 192.168.153.132:5000: connect: connection refused",
    "ServiceID": "CLIENT_SERVICE_01",
    "ServiceName": "CAS Client Service",
    "ServiceTags": ["urlprefix-/ClientService01"],
    "Definition": {
        "HTTP": "",
        "Header": null,
        "Method": "",
        "TLSSkipVerify": false,
        "TCP": "",
        "Interval": "0s",
        "Timeout": "0s",
        "DeregisterCriticalServiceAfter": "0s"
    }
}]

　　你也可以將上面的notice方法替換為其他消息通知的方法，比如郵件，比如短信，我更推薦用短信模式，這樣能馬上告訴運維人員，有節點掛了，該去加班了。

　　綜合之前上一篇的介紹，目前看Consul的Web控制台，已經出現了10個微服務了（不包括系統）！