12 / 2 页下一页

论坛徽章:: 14

电梯直达

1楼 [收藏(0)] [报告]

发表于 2016-03-18 14:35 |显示全部楼层 |倒序浏览

本帖最后由 royzs 于 2016-03-18 14:35 编辑

看阿里云后台流量图形也很怪异，断断续续的，我怀疑就是程序假死，并发支撑不了超过1000，nginx并发不是4万么，求助求助

nginx_status内容如下

Active connections: 951
server accepts handled requests
428729 423504 796115
Reading: 0 Writing: 40 Waiting: 911

复制代码

Active connections这个数值从来没超过1000过

netstat -n | awk '/^tcp/ {++S[$NF]} END {for(a in S) print a, S[a]}'
的内容如下

TIME_WAIT 18216
CLOSE_WAIT 3
SYN_SENT 2
FIN_WAIT1 122
ESTABLISHED 1065
FIN_WAIT2 666
SYN_RECV 7
CLOSING 2
LAST_ACK 48

复制代码

文库|博客

royzs

富足长乐

论坛徽章:: 14

2楼 [报告]

发表于 2016-03-18 14:39 |显示全部楼层

网站运行运行着再看nginx_status就变成这样了，900多变成几百了

Active connections: 510
server accepts handled requests
528566 519322 966128
Reading: 0 Writing: 510 Waiting: 0

复制代码

实战分享：从技术角度谈机器学习入门| 【大话IT】RadonDB低门槛向MySQL集群下战书 | ChinaUnix打赏功能已上线！ | 新一代分布式关系型数据库RadonDB知多少？

royzs

富足长乐

论坛徽章:: 14

3楼 [报告]

发表于 2016-03-18 15:39 |显示全部楼层

网站速度也特别慢，我的nginx配置是这样的

server {
listen 80;
server_name www.sc.com;
index index.html index.htm index.php;
root /mnt/www/fanwe;
location / {
root /mnt/www/fanwe;
index index.html index.htm index.php;
}
location /ngx_status
{
stub_status on;
access_log off;
}
location ~ \.php$ {
root html;
fastcgi_pass 127.0.0.1:9000;
fastcgi_index index.php;
fastcgi_param SCRIPT_FILENAME /mnt/www/fanwe$fastcgi_script_name;
include fastcgi_params;
}
location ~ .*\.(php|php5)?$ {
fastcgi_pass unix:/tmp/php-cgi.sock;
fastcgi_index index.php;
#include fcgi.conf;
}
location ~ .*\.(ico|jpg|jpeg|png|gif)$ {
expires 1y;
valid_referers none blocked www.sc.com *.google.com *.baidu.com;
if ($invalid_referer) {
return 404;
}
}
}

复制代码

实战分享：从技术角度谈机器学习入门| 【大话IT】RadonDB低门槛向MySQL集群下战书 | ChinaUnix打赏功能已上线！ | 新一代分布式关系型数据库RadonDB知多少？

royzs

富足长乐

论坛徽章:: 14

4楼 [报告]

发表于 2016-03-18 22:24 |显示全部楼层

回复 4# lyhabc

tw：time_wait
# ehc0 1> /proc/sys/net/ipv4/tcp_tw_reuse
# ehc0 1>/proc/sys/net/ipv4/tcp_tw_recycle
# ech0 6000>/proc/sys/net/ipv4/tcp_max_tw_buckets

前两个参数都改过的，最下面的那个参数最开始是5000，不够用，系统一直报错kernel: TCP: time wait bucket table overflow
我给调成了2万，3万都不行，最后改成5万不再报错了，但是现在网站还是不稳定，刚启动nginx的时候看并发连接从一两百一直涨到900多1000多点，过几分钟或者半小时nginx就假死了，出现502，重启之后又是循环，所以流量图形里面显示断断续续的

实战分享：从技术角度谈机器学习入门| 【大话IT】RadonDB低门槛向MySQL集群下战书 | ChinaUnix打赏功能已上线！ | 新一代分布式关系型数据库RadonDB知多少？

royzs

富足长乐

论坛徽章:: 14

5楼 [报告]

发表于 2016-03-19 19:40 |显示全部楼层

本帖最后由 royzs 于 2016-03-19 20:22 编辑

回复 7# lyhabc

nginx 1.8.0

阿里云的sysctl.conf我改过，贴一下出来

# sysctl -p
net.ipv4.ip_forward = 0
net.ipv4.conf.default.accept_source_route = 0
kernel.sysrq = 0
kernel.core_uses_pid = 1
kernel.msgmnb = 65536
kernel.msgmax = 65536
kernel.shmmax = 68719476736
kernel.shmall = 4294967296
vm.swappiness = 0
net.ipv4.neigh.default.gc_stale_time = 120
net.ipv4.conf.all.rp_filter = 0
net.ipv4.conf.default.rp_filter = 1
net.ipv4.conf.default.arp_announce = 2
net.ipv4.conf.all.arp_announce = 2
net.ipv4.tcp_max_tw_buckets = 6000
net.ipv4.tcp_sack = 1
net.ipv4.tcp_window_scaling = 1
net.core.wmem_default = 8388608
net.core.rmem_default = 8388608
net.core.rmem_max = 16777216
net.core.wmem_max = 16777216
net.core.somaxconn = 262144
net.ipv4.tcp_max_orphans = 3276800
net.ipv4.tcp_max_syn_backlog = 262144
net.ipv4.tcp_timestamps = 0
net.ipv4.tcp_synack_retries = 1
net.ipv4.tcp_syn_retries = 1
net.ipv4.tcp_tw_recycle = 1
net.ipv4.tcp_mem = 94500000 915000000 927000000
net.ipv4.tcp_fin_timeout = 1
net.ipv4.tcp_keepalive_time = 30
net.ipv4.ip_local_port_range = 1024 65000
net.ipv4.tcp_syncookies = 1

复制代码

net.ipv4.tcp_max_tw_buckets = 6000
的时候并发1000就开始假死，我改成50000后勉强4000多并发就不行了

实战分享：从技术角度谈机器学习入门| 【大话IT】RadonDB低门槛向MySQL集群下战书 | ChinaUnix打赏功能已上线！ | 新一代分布式关系型数据库RadonDB知多少？

royzs

富足长乐

论坛徽章:: 14

6楼 [报告]

发表于 2016-03-19 19:42 |显示全部楼层

回复 8# lyhabc

nginx.conf我的配置是这样的

worker_processes 16;
worker_cpu_affinity 0000000000000001 0000000000000010 0000000000000100 0000000000001000 0000000000010000 0000000000100000 0000000001000000 0000000010000000 0000000100000000 0000001000000000 0000010000000000 0000100000000000 0001000000000000 0010000000000000 0100000000000000 1000000000000000;
error_log /var/log/nginx/error.log crit;
pid /var/run/nginx.pid;
worker_rlimit_nofile 65535;
events {
use epoll;
worker_connections 65535;
}