Metadata-Version: 1.0
Name: nagios.responsetime
Version: 1.0
Summary: A Nagios plugin that collects response times from logs.
Home-page: http://pypi.python.org/pypi/nagios.responsetime
Author: gocept
Author-email: mail@gocept.com
License: ZPL 2.1
Description: A Nagios plugin that collects response times from logs.
        
        Usage: ``check_responsetime --logfile /path/to/logfile``
        
        The plugin remembers the timestamp of the latest log entry it has seen on each
        run, and on the next run only considers log entries that are more recent.
        
        Log format
        ==========
        
        The supported log format is as follows::
        
            <timestamp> "<http-verb> <path> <http-version>" <responsetime>
        
        - timestamp: %d/%b/%Y:%H:%M:%S %z
        - http-verb: GET, POST
        - path: the requested URL path
        - http-version: e.g. HTTP/1.1
        - responsetime in seconds, e. g. 0.5
        
        Example::
        
            17/May/2011:14:11:18 +0200 "GET /index.html HTTP/1.1" 0.289
        
        An example configuration for the nginx server looks like this::
        
            http {
                log_format timing '$time_local "$request" $upstream_response_time';
        
                server {
                    access_log /path/to/timing.log timing;
                }
            }
        
        
        Future directions
        =================
        
        One obvious additional feature is the ability to define thresholds (e.g. "a
        mean responsetime above 3 seconds is WARN, above 5 CRITICAL")
        
        .. contents::
        
        =======
        Changes
        =======
        
        
        1.0 (2011-07-01)
        ----------------
        
        - Fixed bug with incomplete log file lines ("invalid literal for float(): -").
        
        
        1.0rc1 (2011-06-10)
        -------------------
        
        * First release.
        
        
        
Platform: UNKNOWN
