现象:web网站配置负载均衡,使用nginx,有些页面可以,但是登陆后报502 bad gateway
原因:登陆后会有向客户端写cookie的操作导致header比普通访问要大
(可以根据nginx错误日志/usr/local/nginx/logs中upstream sent too big header while reading response header from upstream确定)
解决方案:
这是由于nginx缓冲区有一个bug造成的,我们网站的页面消耗占用缓冲区可能过大,我们是nginx反向代理
proxy是nginx作为client转发时使用的,如果header过大,超出了默认的1k,就会引发上述的upstream sent too big header (说白了就是nginx把外部请求给后端apache ,apache返回的header 太大nginx处理不过来就导致了。
server {
listen 80;
server_name www.server110.com ;
large_client_header_buffers 4 16k;
#charset koi8-r;
# access_log off;
location / {
#添加这3行 ,
proxy_buffer_size 64k;
proxy_buffers 32 32k;
proxy_busy_buffers_size 128k;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
set $baiduspider '';
if ( $http_user_agent ~ Baiduspider) {
set $baiduspider Baidu;
}
............
如果是 nginx+PHPcgi 就该
fastcgi_connect_timeout 60;
fastcgi_send_timeout 180;
fastcgi_read_timeout 180;
fastcgi_buffer_size 128k;
fastcgi_buffers 4 256k;
fastcgi_busy_buffers_size 256k;
fastcgi_temp_file_write_size 256k;
fastcgi_intercept_errors on
011/01/07 11:12:57 [error] 10770#0: *38585340 upstream timed out (110: Connection timed out) while reading response header from upstream,
后来原来那错误没了出了新错误了 upstream timed out 超时?
server {
listen 80;
server_name www.server110.com ;
large_client_header_buffers 4 16k;
client_max_body_size 300m;
client_body_buffer_size 128k;
proxy_connect_timeout 600;
proxy_read_timeout 600;
proxy_send_timeout 600;
proxy_buffer_size 64k;
proxy_buffers 4 32k;
proxy_busy_buffers_size 64k;
proxy_temp_file_write_size 64k;
#charset koi8-r;
# access_log off;