tomcat高并發的配置

以下内容來源于網際網路，具體出處不詳

據說伺服器運作TOMCAT+JDK環境能負載到動态1W的并發，貼上他的配置，以後有機會在測試！

java 環境配置：

export JAVA_OPTS="-server -Xms8g -Xmx8g -Xss128k -XX:ParallelGCThreads=20 -XX:+UseConcMarkSweepGC -XX:+UseParNewGC -XX:SurvivorRatio=8 -XX:TargetSurvivorRatio=90 -XX:MaxTenuringThreshold=31"

sysctl.conf的配置：

net.ipv4.tcp_syncookies = 1

net.ipv4.tcp_max_tw_buckets = 180000

net.ipv4.tcp_sack = 1

net.ipv4.tcp_window_scaling = 1

net.ipv4.tcp_rmem = 4096 87380 4194304

net.ipv4.tcp_wmem = 4096 16384 4194304

net.ipv4.tcp_max_syn_backlog = 65536

net.core.netdev_max_backlog = 32768

net.core.somaxconn = 327680

net.core.wmem_default = 8388608

net.core.rmem_default = 8388608

net.core.rmem_max = 16777216

net.core.wmem_max = 16777216

net.ipv4.tcp_timestamps = 0

net.ipv4.tcp_synack_retries = 2

net.ipv4.tcp_syn_retries = 2

net.ipv4.tcp_fin_timeout = 30

net.ipv4.tcp_tw_recycle = 1

#net.ipv4.tcp_tw_len = 1

net.ipv4.ip_local_port_range = 1024 650000

net.ipv4.tcp_keepalive_time = 5

net.ipv4.tcp_keepalive_probes=2

net.ipv4.tcp_keepalive_intvl=2

net.ipv4.tcp_tw_reuse = 1

net.ipv4.tcp_tw_recyle = 1

net.ipv4.tcp_max_syn_backlog=8192

net.ipv4.tcp_no_metrics_save = 0

net.ipv4.tcp_mem = 94500000 915000000 927000000

net.ipv4.tcp_max_orphans = 3276800

fs.file-max = 9553600

net.ipv4.netfilter.ip_conntrack_max = 655360

net.ipv4.netfilter.ip_conntrack_tcp_timeout_established = 60

net.ipv4.tcp_max_tw_buckets = 60000

net.ipv4.conf.lo.arp_ignore = 1

net.ipv4.conf.lo.arp_announce = 2

net.ipv4.conf.all.arp_ignore = 1

net.ipv4.conf.all.arp_announce = 2

tomcat的SERVER.XML配置：

server.conf 配置

<Connector port="8188" protocol="HTTP/1.1" 

maxThreads="30000" 

minSpareThreads="512" 

maxSpareThreads="2048"

enableLookups="false" 

redirectPort="8443" acceptCount="35000"

debug="0" 

connectionTimeout="40000" 

disableUploadTimeout="true"

URIEncoding="UTF-8" />

參數說明： 

connectionTimeout - 網絡連接配接逾時，機關：毫秒。設定為0表示永不逾時，這樣設定有隐患的。通常可設定為30000毫秒。 

keepAliveTimeout - 長連接配接最大保持時間（毫秒）。此處為15秒。 

maxKeepAliveRequests - 最大長連接配接個數（1表示禁用，-1表示不限制個數，預設100個。一般設定在100~200之間） 

maxHttpHeaderSize - http請求頭資訊的最大程度，超過此長度的部分不予處理。一般8K。 

URIEncoding - 指定Tomcat容器的URL編碼格式。 

acceptCount - 指定當所有可以使用的處理請求的線程數都被使用時，可以放到處理隊列中的請求數，超過這個數的請求将不予處理，預設為10個。

disableUploadTimeout - 上傳時是否使用逾時機制 

enableLookups - 是否反查域名，取值為：true或false。為了提高處理能力，應設定為false 

bufferSize - defines the size (in bytes) of the buffer to be provided for input streams created by this connector. By default, buffers of 2048 bytes are provided. 

maxSpareThreads - 最大空閑連接配接數，一旦建立的線程超過這個值，Tomcat就會關閉不再需要的socket線程 The default value is 50. 

maxThreads - 最多同時處理的連接配接數，Tomcat使用線程來處理接收的每個請求。這個值表示Tomcat可建立的最大的線程數。 minSpareThreads - 最小空閑線程數，Tomcat初始化時建立的線程數 . 

minProcessors - 最小空閑連接配接線程數，用于提高系統處理性能，預設值為10。

maxProcessors - 最大連接配接線程數，即：并發處理的最大請求數，預設值為75。

修改啟動時記憶體參數

window下， 在catalina.bat最前面： 

set JAVA_OPTS=-XX:PermSize=64M -XX:MaxPermSize=128m -Xms512m -Xmx1024m 

一定加在catalina.bat最前面。

linux下，在catalina.sh最前面增加： 

JAVA_OPTS="-XX:PermSize=64M -XX:MaxPermSize=128m -Xms512m -Xmx1024m -Duser.timezone=Asia/Shanghai"

注意：前後二者差別，有無set，有無雙引号。

寫Socket做測試

以下是Groovy代碼

content ='''GET /index.jsp HTTP/1.1

Accept-Language: zh-cn

Connection: Keep-Alive

Host: 192.168.0.53

Content-Length: 36

'''

def sleepTime=1000

Socket socket = new Socket('localhost',8080)

println 'is keep alive? ' + socket.getKeepAlive()

println 'sleep ' + sleepTime + ' ms.'

Thread.sleep(sleepTime)

println 'is closed? ' + socket.isClosed()

//Thread.sleep(sleepTime)

println 'write socket begin======'

writeStream(content, socket.getOutputStream())

println 'read socket begin======'

println readStream(socket.getInputStream())[0..300]

void writeStream(content, stream) {

OutputStream buf = new BufferedOutputStream(stream);

OutputStreamWriter out = new OutputStreamWriter(buf, "UTF-8");

out.write(content)

out.flush();

print content

//out.close();

}

String readStream(stream){

String sResult=''

byte[] buffer = new byte[1024];

int readCount = stream.read(buffer);

while (readCount != -1) {

sResult += new String(buffer, 0,

readCount, "utf-8");

readCount = stream.read(buffer);

stream.close()

return sResult