Nginx错误日志说明

ゝ一世哀愁。 2022-02-23 11:48 1510阅读 0赞

## 错误日志类型 ##

*  类型1: upstream timed out
 *  类型2: connect() failed
 *  类型3: no live upstreams
 *  类型4: upstream prematurely closed connection
 *  类型5: 104: Connection reset by peer
 *  类型6: client intended to send too large body
 *  类型7: upstream sent no valid HTTP/1.0 header

## 详细说明 ##

<table> 
 <thead> 
 <tr> 
 <th> 类型 </th> 
 <th> 错误日志 </th> 
 <th> 原因 </th> 
 <th> 解决办法 </th> 
 </tr> 
 </thead> 
 <tbody> 
 <tr> 
 <th> 类型 </th> 
 <th> 错误日志 </th> 
 <th> 原因 </th> 
 <th> 解决办法 </th> 
 </tr> 
 </tbody> 
 <tbody> 
 <tr> 
 <td>1</td> 
 <td>upstream timed out (110: Connection timed out) while connecting to upstream</td> 
 <td>nginx与upstream建立tcp连接超时，nginx默认连接建立超时为200ms</td> 
 <td>排查upstream是否能正常建立tcp连接</td> 
 </tr> 
 <tr> 
 <td colspan="1">1</td> 
 <td colspan="1">upstream timed out (110: Connection timed out) while reading response header from upstream</td> 
 <td colspan="1">nginx从upstream读取响应时超时，nginx默认的读超时为20s，读超时不是整体读的时间超时，而是指两次读操作之间的超时，整体读耗时有可能超过20s</td> 
 <td colspan="1">排查upstream响应请求为什么过于缓慢</td> 
 </tr> 
 <tr> 
 <td>2</td> 
 <td>connect() failed (104: Connection reset by peer) while connecting to upstream</td> 
 <td>nginx与upstream建立tcp连接时被reset</td> 
 <td>排查upstream是否能正常建立tcp连接</td> 
 </tr> 
 <tr> 
 <td>2</td> 
 <td>connect() failed (111: Connection refused) while connecting to upstream</td> 
 <td>nginx与upstream建立tcp连接时被拒</td> 
 <td>排查upstream是否能正常建立tcp连接</td> 
 </tr> 
 <tr> 
 <td>3</td> 
 <td>no live upstreams while connecting to upstream</td> 
 <td>nginx向upstream转发请求时发现upstream状态全都为down</td> 
 <td>排查nginx的upstream的健康检查为什么失败</td> 
 </tr> 
 <tr> 
 <td>4</td> 
 <td>upstream prematurely closed connection</td> 
 <td>nginx在与upstream建立完tcp连接之后，试图发送请求或者读取响应时，连接被upstream强制关闭</td> 
 <td>排查upstream程序是否异常，是否能正常处理http请求</td> 
 </tr> 
 <tr> 
 <td>5</td> 
 <td>recv() failed (104: Connection reset by peer) while reading response header from upstream</td> 
 <td>nginx从upstream读取响应时连接被对方reset</td> 
 <td>排查upstream应用已经tcp连接状态是否异常</td> 
 </tr> 
 <tr> 
 <td>6</td> 
 <td>client intended to send too large body</td> 
 <td>客户端试图发送过大的请求body，nginx默认最大允许的大小为1m，超过此大小，客户端会受到http 413错误码</td> 
 <td> 
 <ol> 
 <li> 调整请求客户端的请求body大小； </li> 
 <li> 调大相关域名的nginx配置：client_max_body_size； </li> 
 </ol></td> 
 </tr> 
 <tr> 
 <td>7</td> 
 <td>upstream sent no valid HTTP/1.0 header</td> 
 <td>nginx不能正常解析从upstream返回来的请求行</td> 
 <td>排查upstream http响应异常</td> 
 </tr> 
 </tbody> 
</table>

--------------------

nginx日志错误日志说明

<table> 
 <thead> 
 <tr> 
 <th> 错误信息 </th> 
 <th> 错误说明 </th> 
 </tr> 
 </thead> 
 <tbody> 
 <tr> 
 <th> 错误信息 </th> 
 <th> 错误说明 </th> 
 </tr> 
 </tbody> 
 <tbody> 
 <tr> 
 <td> "upstream prematurely（过早的）&nbsp;closed connection" </td> 
 <td> 请求uri的时候出现的异常，是由于upstream还未返回应答给用户时用户断掉连接造成的，对系统没有影响，可以忽略 </td> 
 </tr> 
 <tr> 
 <td> "recv() failed (104: Connection reset by peer)" </td> 
 <td> （1）服务器的并发连接数超过了其承载量，服务器会将其中一些连接Down掉；&nbsp; （2）客户关掉了浏览器，而服务器还在给客户端发送数据；&nbsp; （3）浏览器端按了Stop </td> 
 </tr> 
 <tr> 
 <td> "(111: Connection refused) while connecting to upstream" </td> 
 <td> 用户在连接时，若遇到后端upstream挂掉或者不通，会收到该错误 </td> 
 </tr> 
 <tr> 
 <td> "(111: Connection refused) while reading response header from upstream" </td> 
 <td> 用户在连接成功后读取数据时，若遇到后端upstream挂掉或者不通，会收到该错误 </td> 
 </tr> 
 <tr> 
 <td> "(111: Connection refused) while sending request to upstream" </td> 
 <td> Nginx和upstream连接成功后发送数据时，若遇到后端upstream挂掉或者不通，会收到该错误 </td> 
 </tr> 
 <tr> 
 <td> "(110: Connection timed out) while connecting to upstream" </td> 
 <td> nginx连接后面的upstream时超时 </td> 
 </tr> 
 <tr> 
 <td> "(110: Connection timed out) while reading upstream" </td> 
 <td> nginx读取来自upstream的响应时超时 &nbsp; </td> 
 </tr> 
 <tr> 
 <td> "(110: Connection timed out) while reading response header from upstream" </td> 
 <td> nginx读取来自upstream的响应头时超时 </td> 
 </tr> 
 <tr> 
 <td> "(110: Connection timed out) while reading upstream" </td> 
 <td> nginx读取来自upstream的响应时超时 </td> 
 </tr> 
 <tr> 
 <td> "(104: Connection reset by peer) while connecting to upstream" </td> 
 <td> upstream发送了RST，将连接重置 </td> 
 </tr> 
 <tr> 
 <td> "upstream sent invalid header while reading response header from upstream" </td> 
 <td> upstream发送的响应头无效 </td> 
 </tr> 
 <tr> 
 <td> "upstream sent no valid HTTP/1.0 header while reading response header from upstream" </td> 
 <td> upstream发送的响应头无效 </td> 
 </tr> 
 <tr> 
 <td> "client intended to send too large body" </td> 
 <td> 用于设置允许接受的客户端请求内容的最大值，默认值是1M，client发送的body超过了设置值 </td> 
 </tr> 
 <tr> 
 <td> "reopening logs" </td> 
 <td> 用户发送kill&nbsp;&nbsp;-USR1命令 </td> 
 </tr> 
 <tr> 
 <td> "gracefully shutting down", </td> 
 <td> 用户发送kill&nbsp;&nbsp;-WINCH命令 </td> 
 </tr> 
 <tr> 
 <td> "no servers are inside upstream" </td> 
 <td> upstream下未配置server </td> 
 </tr> 
 <tr> 
 <td> "no live upstreams while connecting to upstream" </td> 
 <td> upstream下的server全都挂了 </td> 
 </tr> 
 <tr> 
 <td> "SSL_do_handshake() failed" </td> 
 <td> SSL握手失败 </td> 
 </tr> 
 <tr> 
 <td> "SSL_write() failed (SSL:) while sending to client" </td> 
 <td>&nbsp;</td> 
 </tr> 
 <tr> 
 <td> "(13: Permission denied) while reading upstream" </td> 
 <td>&nbsp;</td> 
 </tr> 
 <tr> 
 <td> "(98: Address already in use) while connecting to upstream" </td> 
 <td>&nbsp;</td> 
 </tr> 
 <tr> 
 <td> "(99: Cannot assign requested address) while connecting to upstream" </td> 
 <td>&nbsp;</td> 
 </tr> 
 <tr> 
 <td> "ngx_slab_alloc() failed: no memory in SSL session shared cache" </td> 
 <td> ssl_session_cache大小不够等原因造成 </td> 
 </tr> 
 <tr> 
 <td> "could not add new SSL session to the session cache while SSL handshaking" </td> 
 <td> ssl_session_cache大小不够等原因造成 </td> 
 </tr> 
 <tr> 
 <td> "send() failed (111: Connection refused)" </td> 
 <td>&nbsp;</td> 
 </tr> 
 </tbody> 
</table>

ref：[http://blog.sina.com.cn/s/blog\_6e07f1eb010185hp.html][http_blog.sina.com.cn_s_blog_6e07f1eb010185hp.html]

[http_blog.sina.com.cn_s_blog_6e07f1eb010185hp.html]: http://blog.sina.com.cn/s/blog_6e07f1eb010185hp.html