NGINX Documentation

Configuring Logging

This article describes how to configure logging of errors and processed requests in NGINX Open Source and NGINX Plus.

Setting Up the Error Log

NGINX writes information about encountered issues of different severity levels to the error log. The error_log directive sets up logging to a particular file, stderr, or syslog and specifies the minimal severity level of messages to log. By default, the error log is located at logs/error.log (the absolute path depends on the operating system and installation), and messages from all severity levels above the one specified are logged.

The configuration below changes the minimal severity level of error messages to log from error to warn:

error_log logs/error.log warn;

In this case, messages of warn, error crit, alert, and emerg levels are logged.

The default setting of the error log works globally. To override it, place the error_log directive in the main (top-level) configuration context. Settings in the main context are always inherited by other configuration levels (http, server, location). The error_log directive can be also specified at the http, stream, server and location levels and overrides the setting inherited from the higher levels. In case of an error, the message is written to only one error log, the one closest to the level where the error has occurred. However, if several error_log directives are specified on the same level, the message are written to all specified logs.

Note: The ability to specify multiple error_log directives on the same configuration level was added in NGINX Open Source version 1.5.2.

Setting Up the Access Log

NGINX writes information about client requests in the access log right after the request is processed. By default, the access log is located at logs/access.log, and the information is written to the log in the predefined combined format. To override the default setting, use the log_format directive to change the format of logged messages, as well as the access_log directive to specify the location of the log and its format. The log format is defined using variables.

The following examples define the log format that extends the predefined combined format with the value indicating the ratio of gzip compression of the response. The format is then applied to a virtual server that enables compression.

http {
    log_format compression '$remote_addr - $remote_user [$time_local] '
                           '"$request" $status $body_bytes_sent '
                           '"$http_referer" "$http_user_agent" "$gzip_ratio"';

    server {
        gzip on;
        access_log /spool/logs/nginx-access.log compression;
        ...
    }
}

Another example of the log format enables tracking different time values between NGINX and an upstream server that may help to diagnose a problem if your website experience slowdowns. You can use the following variables to log the indicated time values:

  • $upstream_connect_time – The time spent on establishing a connection with an upstream server
  • $upstream_header_time – The time between establishing a connection and receiving the first byte of the response header from the upstream server
  • $upstream_response_time – The time between establishing a connection and receiving the last byte of the response body from the upstream server
  • $request_time – The total time spent processing a request

All time values are measured in seconds with millisecond resolution.

http {
    log_format upstream_time '$remote_addr - $remote_user [$time_local] '
                             '"$request" $status $body_bytes_sent '
                             '"$http_referer" "$http_user_agent"'
                             'rt=$request_time uct="$upstream_connect_time" uht="$upstream_header_time" urt="$upstream_response_time"';

    server {
        access_log /spool/logs/nginx-access.log upstream_time;
        ...
    }
}

Here are some rules how to read the resulting time values:

  • When a request is processed through several servers, the variable contains several values separated by commas
  • When there is an internal redirect from one upstream group to another, the values are separated by semicolons
  • When a request is unable to reach an upstream server or a full header cannot be received, the variable contains “0” (zero)
  • In case of internal error while connecting to an upstream or when a reply is taken from the cache, the variable contains “-” (hyphen)

Logging can be optimized by enabling the buffer for log messages and the cache of descriptors of frequently used log files whose names contain variables. To enable buffering use the buffer parameter of the access_log directive to specify the size of the buffer. The buffered messages are then written to the log file when the next log message does not fit into the buffer as well as in some other cases.

To enable caching of log file descriptors, use the open_log_file_cache directive.

Similar to the error_log directive, the access_log directive defined on a particular configuration level overrides the settings from the previous levels. When processing of a request is completed, the message is written to the log that is configured on the current level, or inherited from the previous levels. If one level defines multiple access logs, the message is written to all of them.

Enabling Conditional Logging

Conditional logging allows excluding trivial or unimportant log entries from the access log. In NGINX, conditional logging is enabled by the if parameter to the access_log directive.

This example excludes requests with HTTP status codes 2xx (Success) and 3xx (Redirection):

map $status $loggable {
    ~^[23]  0;
    default 1;
}

access_log /path/to/access.log combined if=$loggable;

Usecase: Sampling TLS Parameters

Many clients use TLS versions older than TLS 1.3. Though many ciphers are declared insecure, older implementations still use them; ECC certificates offer greater performance than RSA, but not all clients can accept ECC. Many TLS attacks rely on a “man in the middle” who intercepts the cipher negotiation handshake and forces the client and server to select a less secure cipher. Therefore, it’s important to configure NGINX Plus to not support weak or legacy ciphers, but doing so may exclude legacy clients.

You can evaluate the SSL data obtained from the client and determine what proportion of clients get excluded if support for older SSL protocols and ciphers is removed.

The following configuration example logs the SSL protocol, cipher, and User-Agent header of any connected TLS client, assuming that each client selects the most recent protocol and most secure ciphers it supports.

In this example, each client is identified by its unique combination of IP address and User-Agent.

  1. Define the custom log format sslparams that includes the version of the SSL protocol ($ssl_protocol), ciphers used in the connection ($ssl_cipher), the client IP address ($remote_addr), and the value of standard User Agent HTTP request field ($http_user_agent):

    log_format sslparams '$ssl_protocol $ssl_cipher '
                      '$remote_addr "$http_user_agent"';
    
  2. Define a key-value storage that will keep the IP address of the client and its User Agent, for example, clients:

    keyval_zone zone=clients:80m timeout=3600s;
    
  3. Create a variable, for example, $seen for each unique combination of $remote_addr and User-Agent header:

    keyval $remote_addr:$http_user_agent $seen zone=clients;
    
    server {
        listen 443 ssl;
    
        ssl_protocols TLSv1 TLSv1.1 TLSv1.2;
        ssl_ciphers   HIGH:!aNULL:!MD5;
    
        if ($seen = "") {
            set $seen  1;
            set $logme 1;
        }
        access_log  /tmp/sslparams.log sslparams if=$logme;
    
        # ...
    }
    
  4. View the log file generated with this configuration:

    TLSv1.2 AES128-SHA 1.1.1.1 "Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Firefox/45.0"
    TLSv1.2 ECDHE-RSA-AES128-GCM-SHA256 2.2.2.2 "Mozilla/5.0 (iPhone; CPU iPhone OS 9_1 like Mac OS X) AppleWebKit/601.1.46 (KHTML, like Gecko) Version/9.0 Mobile/13B143 Safari/601.1"
    TLSv1.2 ECDHE-RSA-AES128-GCM-SHA256 3.3.3.3 "Mozilla/5.0 (Windows NT 6.1; WOW64; rv:58.0) Gecko/20100101 Firefox/58.0"
    TLSv1.2 ECDHE-RSA-AES128-GCM-SHA256 4.4.4.4 "Mozilla/5.0 (Android 4.4.2; Tablet; rv:65.0) Gecko/65.0 Firefox/65.0"
    TLSv1 AES128-SHA 5.5.5.5 "Mozilla/5.0 (Android 4.4.2; Tablet; rv:65.0) Gecko/65.0 Firefox/65.0"
    TLSv1.2 ECDHE-RSA-CHACHA20-POLY1305 6.6.6.6 "Mozilla/5.0 (Linux; U; Android 5.0.2; en-US; XT1068 Build/LXB22.46-28) AppleWebKit/537.36 (KHTML, like Gecko) Version/4.0 Chrome/57.0.2987.108 UCBrowser/12.10.2.1164 Mobile Safari/537.36"
    
  5. Process the log file to determine the spread of data:

    cat /tmp/sslparams.log | cut -d ' ' -f 2,2 | sort | uniq -c | sort -rn | perl -ane 'printf "%30s %s\n", $F[1], "="x$F[0];'
    

    In this output, low‑volume, less secure ciphers are identified:

    ECDHE-RSA-AES128-GCM-SHA256 =========================
    ECDHE-RSA-AES256-GCM-SHA384 ========
                     AES128-SHA ====
    ECDHE-RSA-CHACHA20-POLY1305 ==
        ECDHE-RSA-AES256-SHA384 ==
    

    Then you can check the logs to determine which clients are using these ciphers and then make a decision about removing these ciphers from the NGINX Plus configuration.

    For more information about sampling requests with NGINX conditional logging see the blog post.

Logging to Syslog

The syslog utility is a standard for computer message logging and allows collecting log messages from different devices on a single syslog server. In NGINX, logging to syslog is configured with the syslog: prefix in error_log and access_log directives.

Syslog messages can be sent to a server= which can be a domain name, an IP address, or a UNIX-domain socket path. A domain name or IP address can be specified with a port to override the default port, 514. A UNIX-domain socket path can be specified after the unix: prefix:

error_log  server=unix:/var/log/nginx.sock debug;
access_log syslog:server=[2001:db8::1]:1234,facility=local7,tag=nginx,severity=info;

In the example, NGINX error log messages are written to a UNIX domain socket at the debug logging level, and the access log is written to a syslog server with an IPv6 address and port 1234.

The facility= parameter specifies the type of program that is logging the message. The default value is local7. Other possible values are: auth, authpriv, daemon, cron, ftp, lpr, kern, mail, news, syslog, user, uucp, local0 ... local7.

The tag= parameter applies a custom tag to syslog messages (nginx in our example).

The severity= parameter sets the severity level of syslog messages for access log. Possible values in order of increasing severity are: debug, info, notice, warn, error (default), crit, alert, and emerg. Messages are logged at the specified level and all more severe levels. In our example, the severity level error also enables crit, alert, and emerg levels to be logged.

Live Activity Monitoring

NGINX Plus provides a real-time live activity monitoring interface that shows key load and performance metrics of your HTTP and TCP upstream servers. See the Live Activity Monitoring article for more information.

To learn more about NGINX Plus, please visit the Products page.