Logging data in Nekonote Framework is multiprocess-safety and thread-safety. When you trun on the logger in logger.yml, logwrite method is callable in top level.

Configuring The Logger

It’s possible to configure the logger through logger.yml.

Turning On The Logger

You need to turn on the logger by logger.yml if the logger is disabled. The logger will be enabled by default after lintalling Nekonote Framework. So there is no need to do for using the logger.

There is the directive in logger.yml to turn on/off the logger.

preference/your_environment/logger.yml

enabled: true

When true is set to enabled directive logwrite method will be defined as public instance method of Object class. When there is a need to turn off the logger, you need to set false to the ‘enabled’ directive.

Logging Data

When the logger is enabled, it’s possible to call logwrite method anywhere you want because logwrite is a public instance method of Object class.

Here is the reference to logwrite method.

logwrite(msg, level = nil)

When log file: log/%s.log is set in logger.yml, the log file will be created to /path/to/app/root/log/your_environment.log when logwrite is called.

Logging Data Example

logwrite 'log message'

The above example will write log message similar into the log file as Thu 04 May 06:09:19 [INFO] log message.

$LEVEL was replaced INFO because the default log severity level is set to info. You may change the default log severity level by default_log_level directive in logger.yml.

The second argument for logwrite sets the log severity level for to its value which for the given log message.

Let’s change the log severity level to fatal. The following example reports the log message as fatal level.

logwrite 'log message', Nekonote::Logger::FATAL

The above example will write Thu 04 May 06:09:19 [FATAL] log message similar into the log file.

When the second argument passed to logwrite method, the default log severity level is set to the passed value automatically.

These are the specifiable constant values for the second argument of logwrite (log severity levels).

Here is the list of constants which can be passed to the second argument to logwrite.

Ignoring Uninportant Logs

Each of log messages has a severity level.

The following lists all severity levels in order of precedence: fatal, error, warn, info, debug.

When you don’t need to be reported debug or info level log messages, you need to configure as:

preference/your_environment/logger.yml

threshold: warn

The above exmaple says log messages which log severity levels are warn or error or fatal will be written to the log file. But log messages which are debug or info level will not be written to the log file.

logwrite 'DEBUG', Nekonote::Logger::DEBUG
logwrite 'INFO',  Nekonote::Logger::INFO
logwrite 'WARN',  Nekonote::Logger::WARN
logwrite 'ERROR', Nekonote::Logger::ERROR
logwrite 'FATAL', Nekonote::Logger::FATAL

The following result will be written similar to the log file.

Thu 04 May 07:24:34 [WARN] WARN
Thu 04 May 07:24:34 [ERROR] ERROR
Thu 04 May 07:24:34 [FATAL] FATAL

The result says debug or info levels were not reported.

Formatting Log Messages

It’s possible to customize the log message format by log > format directive in logger.yml.

There are the some special variables for formatting log messages. They will be replaced as the following list:

To change the default log message format, you need to configure as following exmaple:

preference/your_environment/logger.yml

format:
    log: "$LEVEL $DATE $LOG $PID"
    datetime: "%a %d %b %H:%M:%S"

The default format for $DATE is "%a %d %b %H:%M:%S". Let’s change it to RFC 822 date-time.

format:
    log: "$LEVEL $DATE $LOG $PID"
    datetime: "%a, %d %b %Y %H:%M:%S %z"

After editing logger.yml, you need to restart the web server to reload the new configuration file.

logwrite 'log message'

The result will say similar:

INFO Fri, 05 May 2017 06:14:39 +0000 log message 10248

Managing Log Files

The logger helps to manage your log files.

It can prevent eating up available hard drive space by single log file.

The logger can rotate your log files by removing the oldest ones and creating new ones. That will be done automatically when the threshold reached which set in logger.yml.

Log Rotation By File-Size

If some value is set to limit directive in logger.yml, The logger rotates files when file size of the log file reached the threshold set in logger.yml.

The value of limit directive needs to be set as an integer, but it can be put K, M, G at the end of the value. K means ‘kilobytes’, M means ‘megabytes’, G means ‘gigabytes’.

No K, M, G at the end of the value is assumed that it’s just ‘byte’.

preference/your_environment/logger.yml

limit: 500M #=> log files are rotated only if their file-size is bigger than 500 megabytes

Log Rotation Periodically

When "DAILY" or "WEEKLY" or "MONTHLY" is set in the limit directive, log files are rotated daily or weekly or monthly.

limit: WEEKLY #=> log files are rotated weekly

Keeping Log Files

After rotation done, the log file which reached threshold is keeped as an old log file. You may configure that how many old log files are supposed to be keeped in your server.

When the logger rotated log files,

The logger creates a new file and log messages will be written to the new one after rotating. Typically The previous one is retained as old file.

You may configure how many old log files are supposed to be keeping.

The default value of keep directive is set 0. The meaning of 0 is keeping all old log files and any log files are not removed by rotation.

Example

When you need to be kept four of old log files, you need to set 4 to keep directive in logger.yml.

Old log files can be compressed as well. You need to set as following:

preference/your_environment/logger.yml

keep: 4

compress:
    enabled: true # set compression enabled
    level: 6

The above level directive means compression level for gzip. Old log files are compressed as this gzip compression level.

You may change it in the range from 1 to 9. Larger numbers mean higher compression but it will take more time to compress files than smaller number.