詳解PHP Swoole長連線常見問題
連線失效問題
例子
其中,Redis常見的報錯就是:
配置項:timeout
報錯資訊:
Error while reading line from the server
Redis可以配置如果客戶端經過多少秒還不給Redis伺服器傳送資料,那麼就會把連線close掉。
mysql常見的報錯:
配置項:wait_timeout & interactive_timeout
報錯資訊:
has gone away
和Redis伺服器一樣,MySQL也會定時的去清理掉沒用的連線。
如何解決
1、用的時候進行重連 。優點是簡單,缺點是面臨短連線的問題。
2、定時傳送心跳維持連線(推薦)。
如何維持長連線
tcp協議中實現的tcp_keepalive
作業系統底層提供了一組tcp的keepalive配置:
tcp_keepalive_time (integer; default: 7200; since linux 2.2) The number of seconds a connection needs to be idle before TCP begins sending out keep-alive probes. Keep-alives are sent only when the SO_KEEPALIVE socket option is enabled. The default value is 7200 seconds (2 hours). An idle connection is terminated after approximately an additional 11 minutes (9 probes an interval of 75 seconds apart) when keep-alive is enabled. Note that underlying connection tracking mechanisms and application timeouts may be much shorter. tcp_keepalive_intvl (integer; default: 75; since Linux 2.4) The number of seconds between TCP keep-alive probes. tcp_keepalive_probes (integer; default: 9; since Linux 2.2) The maximum number of TCP keep-alive probes to send before giving up and killing the connection if no response is obtained from the other end. 8
Swoole底層把這些配置開放出來了,例如:
?php $server = new \Swoole\Server('127.0.0.1',6666,SWOOLE_PROCESS); $server->set([ 'worker_num' => 1,'open_tcp_keepalive' => 1,'tcp_keepidle' => 4,// 對應tcp_kee程式設計客棧palive_time 'tcp_keepinterval' => 1,// 對應tcp_keepalive_intvl 'tcp_keepcount' => 5,// 對應tcp_keepalive_probes ]);
其中:
'open_tcp_keepalive' => 1,// 總開關,用來開啟tcp_keepalive 'tcp_keepidle' => 4,// 4s沒有資料傳輸就進行檢測 // 檢測的策略如下: 'tcp_keepinterval' => 1,// 1s探測一次,即每隔1s給客戶端發一個包(然後客戶端可能會回一個ack的包,如果服務端收到了這個ack包,那麼說明這個連線是活著的) 'tcp_keepcount' => 5,// 探測的次數,超過5次後客戶端還沒有回ack包,那麼close此連線
我們來實戰測試體驗一下,服務端指令碼如下:
<?php
$server = new \Swoole\Server('127.0.0.1',// 開啟tcp_keepalive
'tcp_keepidle' => 4,// 4s沒有資料傳輸就進行檢測
'tcp_keepin程式設計客棧terval' => 1,// 1s探測一次
'tcp_keepcount' => 5,// 探測的次數,超過5次後還沒有回包close此連線
]);
$server->on('connect',function ($server,$fd) {
var_dump("Client: Connect $fd");
});
$server->on('receive',$fd,$reactor_id,$data) {
var_dump($data);
});
$server->on('close',$fd) {
var_dump("close fd $fd");
});
$server->start();
我們啟動這個伺服器:
~/codeDir/phpCode/hyperf-skeleton # php server.php
然後通過tcpdump進行抓包:
~/codeDir/phpCode/hyperf-skeleton # tcpdump -i lo port 6666 tcpdump: verbose output suppressed,use -v or -vv for full protocol decode listening on lo,link-type EN10MB (Ethernet),capture size 262144 bytes
我們此時正在監聽lo上的6666埠的資料包。
然後我們用客戶端去連線它:
~/codeDir/phpCode/hyperf-skeleton # nc 127.0.0.1 6666
此時服務端會打印出訊息:
~/codeDir/phpCode/hyperf-skeleton # php server.php
string(17) "Client: Connect 1"
tcpdump的輸出資訊如下:
01:48:40.178439 IP localhost.33933 > localhost.6666: Flags [S],seq 43162537,win 43690,options [mss 65495,sackOK,TS val 9833698 ecr 0,nop,wscale 7],length 0
01:48:40.178484 IP localhost.6666 > localhost.33933: Flags [S.],seq 1327460565,ack 43162538,TS val 9833698 ecr 9833698,length 0
01:48:40.178519 IP localhost.33933 > localhost.6666: Flags [.],ack 1,win 342,options [nop,TS val 9833698 ecr 9833698],length 0
01:48:44.229926 IP localhost.6666 > localhost.33933: Flags [.],TS val 9834104 ecr 9833698],length 0
01:48:44.229951 IP localhost.33933 > localhost.6666: Flags [.],length 0
// 省略了其他的輸出
我們會發現最開始的時候,會列印三次握手的包:
01:48:40.178439 IP localhost.33933 > localhost.6666: Flags [S],length 0
01:48:40.178519 IP localhost.33933 www.cppcns.com> localhost.6666: Flags [.],length 0
然後,停留了4s沒有任何包的輸出。
之後,每隔1s左右就會打印出一組:
01:52:54.359341 IP localhost.6666 > localhost.43101: Flags [.],TS val 9859144 ecr 9858736],length 0
01:52:54.359377 IP localhost.43101 > localhost.6666: Flags [.],TS val 9859144 ecr 9855887],length 0
其實這就是我們配置的策略:
'tcp_keepinterval' => 1,// 探測的次數,超過5次後還沒有回包close此連線
因為我們作業系統底層會自動的給客戶端回ack,所以這個連線不會在5次探測後被關閉。作業系統底層會持續不斷的傳送這樣的一組包:
01:52:54.359341 IP localhost.6666 > localhost.43101: Flags [.],win程式設計客棧 342,length 0
如果我們要測試5次探測後關閉這個連線,可以禁掉6666埠的包:
~/codeDir/phpCode/hyperf-skeleton # iptables -A INPUT -p tcp --dport 6666 -j DROP
這樣會把所有從6666埠進來的包給禁掉,自然,伺服器就接收不到從客戶端那一邊發來的ack包了。
然後伺服器過5秒就會打印出close(服務端主動的呼叫了close方法,給客戶端傳送了FIN包):
~/codeDir/phpCode/hyperf-skeleton # php server.php
string(17) "Client: Connect 1"
string(10) "close fd 1"
我們恢復一下iptables的規則:
~/codeDir/phpCode # iptables -D INPUT -p tcp -m tcp --dport 6666 -j DROP
即把我們設定的規則給刪除了。
通過tcp_keepalive的方式實現心跳的功能,優點是簡單,不要寫程式碼就可以完成這個功能,並且傳送的心跳包小。缺點是依賴於系統的網路環境,必須保證伺服器和客戶端都實現了這樣的功能,需要客戶端配合發心跳包。還有一個更為嚴重的缺點是如果客戶端和伺服器不是直連的,而是通過代理來進行連線的,例如socks5代理,它只會轉發應用層的包,不會轉發更為底層的tcp探測包,那這個心跳功能就失效了。
所以,Swoole就提供了其他的解決方案,一組檢測死連線的配置。
'heartbeat_check_interval' => 1,// 1s探測一次 'heartbeat_idle_time' => 5,// 5s未傳送資料包就close此連線
swoole實現的heartbeat
我們來測試一下:
<?php $server = new \Swoole\Server('127.0.0.1',SWOOLE_PROCESS); $server->set([ 'worker程式設計客棧_num' => 1,'heartbeat_check_interval' => 1,// 5s未傳送資料包就close此連線 ]); $server->on('connect',$fd) { var_dump("close fd $fd"); }); $server->start();
然後啟動伺服器:
~/codeDir/phpCode/hyperf-skeleton # php server.php
然後啟動tcpdump:
~/codeDir/phpCode # tcpdump -i lo port 6666
tcpdump: verbose output suppressed,use -v or -vv for full protocol decode
listening on lo,capture size 262144 bytes
然後再啟動客戶端:
~/codeDir/phpCode/hyperf-skeleton # nc 127.0.0.1 6666
此時伺服器端列印:
~/codeDir/phpCode/hyperf-skeleton # php server.php
string(17) "Client: Connect 1"
然後tcpdump列印:
02:48:32.516093 IP localhost.42123 > localhost.6666: Flags [S],seq 1088388248,TS val 10193342 ecr 0,length 0
02:48:32.516133 IP localhost.6666 > localhost.42123: Flags [S.],seq 80508236,ack 1088388249,TS val 10193342 ecr 10193342,length 0
02:48:32.516156 IP localhost.42123 > localhost.6666: Flags [.],TS val 10193342 ecr 10193342],length 0
這是三次握手資訊。
然後過了5s後,tcpdump會打印出:
02:48:36.985027 IP localhost.6666 > localhost.42123: Flags [F.],seq 1,TS val 10193789 ecr 10193342],length 0
02:48:36.992172 IP localhost.42123 > localhost.6666: Flags [.],ack 2,TS val 10193790 ecr 10193789],length 0
也就是服務端傳送了FIN包。因為客戶端沒有傳送資料,所以Swoole關閉了連線。
然後伺服器端會列印:
~/codeDir/phpCode/hyperf-skeleton # php server.php
string(17) "Client: Connect 1"
string(10) "close fd 1"
所以,heartbeat和tcp keepalive還是有一定的區別的,tcp keepalive有保活連線的功能,但是heartbeat存粹是檢測沒有資料的連線,然後關閉它,並且只可以在服務端這邊配置,如果需要保活,也可以讓客戶端配合傳送心跳。
如果我們不想讓服務端close掉連線,那麼就得在應用層裡面不斷的傳送資料包來進行保活,例如我在nc客戶端裡面不斷的傳送包:
~/codeDir/phpCode/hyperf-skeleton # nc 127.0.0.1 6666
ping
ping
ping
ping
ping
ping
ping
ping
ping
我傳送了9個ping包給伺服器,tcpdump的輸出如下:
// 省略了三次握手的包
02:57:53.697363 IP localhost.44195 > localhost.6666: Flags [P.],seq 1:6,TS val 10249525 ecr 10249307],length 5
02:57:53.697390 IP localhost.6666 > localhost.44195: Flags [.],ack 6,TS val 10249525 ecr 10249525],length 0
02:57:55.309532 IP localhost.44195 > localhost.6666: Flags [P.],seq 6:11,TS val 10249686 ecr 10249525],length 5
02:57:55.309576 IP localhost.6666 > localhost.44195: Flags [.],ack 11,TS val 10249686 ecr 10249686],length 0
02:57:58.395206 IP localhost.44195 > localhost.6666: Flags [P.],seq 11:16,TS val 10249994 ecr 10249686],length 5
02:57:58.395239 IP localhost.6666 > localhost.44195: Flags [.],ack 16,TS val 10249994 ecr 10249994],length 0
02:58:01.858094 IP localhost.44195 > localhost.6666: Flags [P.],seq 16:21,TS val 10250341 ecr 10249994],length 5
02:58:01.858126 IP localhost.6666 > localhost.44195: Flags [.],ack 21,TS val 10250341 ecr 10250341],length 0
02:58:04.132584 IP localhost.44195 > localhost.6666: Flags [P.],seq 21:26,TS val 10250568 ecr 10250341],length 5
02:58:04.132609 IP localhost.6666 > localhost.44195: Flags [.],ack 26,TS val 10250568 ecr 10250568],length 0
02:58:05.895704 IP localhost.44195 > localhost.6666: Flags [P.],seq 26:31,TS val 10250744 ecr 10250568],length 5
02:58:05.895728 IP localhost.6666 > localhost.44195: Flags [.],ack 31,TS val 10250744 ecr 10250744],length 0
02:58:07.150265 IP localhost.44195 > localhost.6666: Flags [P.],seq 31:36,TS val 10250870 ecr 10250744],length 5
02:58:07.150288 IP localhost.6666 > localhost.44195: Flags [.],ack 36,TS val 10250870 ecr 10250870],length 0
02:58:08.349124 IP localhost.44195 > localhost.6666: Flags [P.],seq 36:41,TS val 10250990 ecr 10250870],length 5
02:58:08.349156 IP localhost.6666 > localhost.44195: Flags [.],ack 41,TS val 10250990 ecr 10250990],length 0
02:58:09.906223 IP localhost.44195 > localhost.6666: Flags [P.],seq 41:46,TS val 10251145 ecr 10250990],length 5
02:58:09.906247 IP localhost.6666 > localhost.44195: Flags [.],ack 46,TS val 10251145 ecr 10251145],length 0
有9組資料包的傳送。(這裡的Flags [P.]代表Push的含義)
此時伺服器還沒有close掉連線,實現了客戶端保活連線的功能。然後我們停止傳送ping,過了5秒後tcpdump就會輸出一組:
02:58:14.811761 IP localhost.6666 > localhost.44195: Flags [F.],TS val 10251636 ecr 10251145],length 0
02:58:14.816420 IP localhost.44195 > localhost.6666: Flags [.],TS val 10251637 ecr 10251636],length 0
服務端那邊傳送了FIN包,說明服務端close掉了連線。服務端的輸出如下:
~/codeDir/phpCode/hyperf-skeleton # php server.php
string(17) "Client: Connect 1"
string(5) "ping
"
string(5) "ping
"
string(5) "ping
"
string(5) "ping
"
string(5) "ping
"
string(5) "ping
"
string(5) "ping
"
string(5) "ping
"
string(5) "ping
"
string(10) "close fd 1"
然後我們在客戶端那邊ctrl + c來關閉連線:
~/codeDir/phpCode/hyperf-skeleton # nc 127.0.0.1 6666
ping
ping
ping
ping
ping
ping
ping
ping
ping
^Cpunt!
~/codeDir/phpCode/hyperf-skeleton #
此時,tcpdump的輸出如下:
03:03:02.257667 IP localhost.44195 > localhost.6666: Flags [F.],seq 46,TS val 10280414 ecr 10251636],length 0
03:03:02.257734 IP localhost.6666 > localhost.44195: Flags [R],seq 2678621620,win 0,length 0
應用層心跳
1、制定ping/pong協議(mysql等自帶ping協議)
2、客戶端靈活的傳送ping心跳包
3、服務端OnRecive檢查可用性回覆pong
例如:
$server->on('receive',function (\Swoole\Server $server,$data) { if ($data == 'ping') { checkDB(); checkServiceA(); checkRedis(); $server->send('pong'); } });
結論
1、tcp的keepalive最簡單,但是有相容性問題,不夠靈活
2、swoole提供的keepalive最實用,但是需要客戶端配合,複雜度適中
3、應用層的keepalive最靈活但是最麻煩
以上就是詳解PHP Swoole長連線常見問題的詳細內容,更多關於PHP Swoole長連線常見問題的資料請關注我們其它相關文章!