參考：Nginx 作爲反向代理優化要點proxy_buffering
利用nginx做反向代理時，配置文件可以參考下面的配置：
user nginx;
worker_processes 2;
error_log logs/error.log warn;
pid logs/nginx.pid;
events {
worker_connections 10240;
multi_accept on;
use epoll;
}
http {
include mime.types;
default_type application/octet-stream;
log_format main '$remote_addr - $remote_user [$time_local] "$request" '
'$status $body_bytes_sent "$http_referer" '
'"$http_user_agent" "$http_x_forwarded_for"';
access_log logs/access.log main;
sendfile on;
tcp_nopush on;
tcp_nodelay on;
underscores_in_headers on;#off 表示如果header name中包含下劃線,則忽略掉,部署後就獲取不到 on則相反
keepalive_timeout 240;
server_tokens off; #隱藏版本號
send_timeout 10m;
client_header_buffer_size 256k;
client_max_body_size 1000m;
client_body_timeout 10m;
client_header_timeout 10m;
large_client_header_buffers 8 64k;
client_body_buffer_size 20m;
fastcgi_buffers 6 256k;
fastcgi_buffer_size 1024k;
fastcgi_busy_buffers_size 1024k;
proxy_request_buffering off;
proxy_buffering off; #開啓從後端被代理服務器的響應內容緩衝
proxy_buffer_size 128k; #設置緩衝區的大小和數量
proxy_buffers 100 128k;
proxy_busy_buffers_size 128k;
upstream NAME {
server host:port;
}
upstream NAME_SSL {
server host:443;
}
server {
listen 80;
server_name a.example.com;
charset utf8;
ignore_invalid_headers off;
error_page 404 403 =https://a.example.com/404.html;
location / {
include conf.d/deny_appid.conf; #location自定義配置
proxy_pass http://NAME/;
proxy_redirect off;
proxy_set_header Host $http_host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header Cookie $http_cookie;
chunked_transfer_encoding off;
proxy_http_version 1.1;
proxy_set_header Upgrade $http_upgrade;
}
}
server {
listen 443 ssl;
server_name a.example.com;
ignore_invalid_headers off;
charset utf8;
ssl_certificate example.com.crt;
ssl_certificate_key example.com.key;
ssl_session_cache shared:SSL:1m;
ssl_session_timeout 5m;
ssl_ciphers HIGH:!aNULL:!MD5;
ssl_prefer_server_ciphers on;
error_page 404 403 =https://a.example.com/404.html;
location / {
include conf.d/deny_appid.conf;
proxy_pass https://NAME_SSL/;
proxy_ssl_session_reuse off;
proxy_redirect off;
proxy_set_header Host $http_host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header Cookie $http_cookie;
chunked_transfer_encoding off;
proxy_http_version 1.1;
proxy_set_header Upgrade $http_upgrade;
}
}
}nginx+tomcat 報錯：『an upstream response is buffered to a temporary file 』
參考鏈接：https://blog.51cto.com/11413817/1950467
nginx的an upstream response is buffered to a temporary file報錯
client_header_buffer_size 256k; 把原來的32k改爲256k
nginx fastcgi_buffers to an upstream response is buffered to a temporary file
fastcgi_buffers 16 16k; 指定本地需要用多少和多大的緩衝區來緩衝FastCGI的應答，如上所示，如果一個php腳本所產生的頁面大小爲256k，則會爲其分配16個16k的緩衝區來緩存，如果大於256k，增大於256k的部分會緩存到fastcgi_temp指定的路徑中，當然這對服務器負載來說是不明智的方案，因爲內存中處理數據速度要快於硬盤，通常這個值的設置應該選擇一個你的站點中的php腳本所產生的頁面大小的中間值，比如你的站點大部分腳本所產生的頁面大小爲256k就可以把這個值設置爲16 16k，或者4 64k 或者64 4k，但很顯然，後兩種並不是好的設置方法，因爲如果產生的頁面只有32k，如果用4 64k它會分配1個64k的緩衝區去緩存，而如果使用64 4k它會分配8個4k的緩衝區去緩存，而如果使用16 16k則它會分配2個16k去緩存頁面，這樣看起來似乎更加合理。 //+++++++++++++++++++++++++++++++++++++++++ client_max_body_size 100m; #允許客戶端請求的最大單文件字節數 client_body_buffer_size 2048k; #緩衝區代理緩衝用戶端請求的最大字節數， fastcgi_buffer_size 1024k; fastcgi_buffers 6 256k; fastcgi_busy_buffers_size 1024k; fastcgi_buffer等於：fastcgi_buffer_size + the_number * is_size fastcgi_buffers 256 4k; #設置buffer大小爲：4k + 256 * 4k = 1028k 上面配置可以解決 an upstream response is buffered to a temporary file 或nginx+php-fpm慢問題
php502問題解決：recv() failed (104: Connection reset by peer) while reading response header from upstream
參考：https://blog.csdn.net/donglynn/article/details/76759548
nginx error: upstream prematurely closed connection while reading response header from upstream
參考：https://www.cnblogs.com/xinpureZhu/p/6142183.html
nginx access log 關閉
access_log off;
nginx error_log 日誌調整級別：
error_log logs/error.log info|notice|warn|error;