Logster - generate metrics from logfiles

Logster is a utility for reading log files and generating metrics to configurable outputs. It is ideal for visualizing trends of events that are occurring in your application/system/error logs. For example, you might use logster to graph the number of occurrences of HTTP response code that appears in your web server logs.

Logster maintains a cursor, via a tailer, on each log file that it reads so that each successive execution only inspects new log entries. In other words, a 1 minute crontab entry for logster would allow you to generate near real-time trends in the configured output for anything you want to measure from your logs.

This tool is made up of a framework script, logster, and parsing classes that are written to accommodate your specific log format. Sample parsers are included in this distribution. The parser classes essentially read a log file line by line, apply a regular expression to extract useful data from the lines you are interested in, and then aggregate that data into metrics that will be submitted to the configured output. The sample parsers should give you some idea of how to get started writing your own. A list of available parsers can be found on the Parsers page.

Graphite, Ganglia, Amazon CloudWatch, Nagios, StatsD and stdout outputs are provided, and Logster also supports the use of third-party output classes. A list of available output classes can be found on the Outputs page.

History

The logster project was created at Etsy as a fork of ganglia-logtailer (https://bitbucket.org/maplebed/ganglia-logtailer). We made the decision to fork ganglia-logtailer because we were removing daemon-mode from the original framework. We only make use of cron-mode, and supporting both cron- and daemon-modes makes for more work when creating parsing scripts. We care strongly about simplicity in writing parsing scripts -- which enables more of our engineers to write log parsers quickly.

Installation

Logster supports two methods for gathering data from a logfile:

1. By default, Logster uses the "logtail" utility that can be obtained from the logcheck package, either from a Debian package manager or from source:

   http://packages.debian.org/source/sid/logcheck
   

   RPMs for logcheck can be found here:

   http://rpmfind.net/linux/rpm2html/search.php?query=logcheck
   

2. Optionally, Logster can use the "Pygtail" Python module instead of logtail. You can install Pygtail using pip

   $ pip install pygtail
   

   To use Pygtail, supply the --tailer=pygtail option on the Logster commandline.

Also, Logster supports two methods for locking files (which it has to do):

1. By default, Logster uses fcntl.flock.

2. Optionally, Logster can use the "Portalocker" Python module instead of fcntl (which is not available on Windows). You can install Portalocker using pip, similar to Pygtail above.

   To use Portalocker, supply the --locker=portalocker option on the Logster commandline.

Once you have logtail or Pygtail installed, install Logster using the setup.py file:

$ sudo python setup.py install

Usage

You can test logster from the command line. The --dry-run option will allow you to see the metrics being generated on stdout rather than sending them to your configured output.

$ sudo /usr/bin/logster --dry-run --output=ganglia SampleLogster /var/log/httpd/access_log

$ sudo /usr/bin/logster --dry-run --output=graphite --graphite-host=graphite.example.com:2003 SampleLogster /var/log/httpd/access_log

You can use the provided parsers, or you can use your own parsers by passing the complete module and parser name. In this case, the name of the parser does not have to match the name of the module (you can have a logster.py file with a MyCustomParser parser). Just make sure the module is in your Python path - via a virtualenv, for example.

$ /env/my_org/bin/logster --dry-run --output=stdout my_org_package.logster.MyCustomParser /var/log/my_custom_log

Additional usage details can be found with the -h option:

$ logster -h
Usage: logster [options] parser logfile

Tail a log file and filter each line to generate metrics that can be sent to
common monitoring packages.

Options:
  -h, --help            show this help message and exit
  -t TAILER, --tailer=TAILER
                        Specify which tailer to use. Options are logtail and
                        pygtail. Default is "logtail".
  --logtail=LOGTAIL     Specify location of logtail. Default
                        "/usr/sbin/logtail2"
  -p METRIC_PREFIX, --metric-prefix=METRIC_PREFIX
                        Add prefix to all published metrics. This is for
                        people that may multiple instances of same service on
                        same host.
  -x METRIC_SUFFIX, --metric-suffix=METRIC_SUFFIX
                        Add suffix to all published metrics. This is for
                        people that may add suffix at the end of their
                        metrics.
  --parser-help         Print usage and options for the selected parser
  --parser-options=PARSER_OPTIONS
                        Options to pass to the logster parser such as "-o
                        VALUE --option2 VALUE". These are parser-specific and
                        passed directly to the parser.
  -s STATE_DIR, --state-dir=STATE_DIR
                        Where to store the tailer state file.  Default
                        location /var/run
  -l LOG_DIR, --log-dir=LOG_DIR
                        Where to store the logster logfile.  Default location
                        /var/log/logster
  --log-conf=LOG_CONF   Logging configuration file. None by default
  -o OUTPUT, --output=OUTPUT
                        Where to send metrics (can specify multiple times).
                        Choices are statsd, stdout, cloudwatch, graphite,
                        ganglia, nsca or a fully qualified Python class name
  -d, --dry-run         Parse the log file but send stats to standard output.
  -D, --debug           Provide more verbose logging for debugging.

Contributing

• Fork the project
• Add your feature
• If you are adding new functionality, document it in the README
• Verify your code by running the test suite, and adding additional tests if able.
• Push the branch up to GitHub (bonus points for topic branches)
• Send a pull request to the etsy/logster project.

If you have questions, you can find us on IRC in the #codeascraft channel on Freenode.