NAME¶
Log::Dispatch::FileRotate - Log to files that archive/rotate themselves
SYNOPSIS¶
use Log::Dispatch::FileRotate;
my $file = Log::Dispatch::FileRotate->new( name => 'file1',
min_level => 'info',
filename => 'Somefile.log',
mode => 'append' ,
size => 10,
max => 6,
);
# or for a time based rotation
my $file = Log::Dispatch::FileRotate->new( name => 'file1',
min_level => 'info',
filename => 'Somefile.log',
mode => 'append' ,
TZ => 'AEDT',
DatePattern => 'yyyy-dd-HH',
);
$file->log( level => 'info', message => "your comment\n" );
DESCRIPTION¶
This module provides a simple object for logging to files under the
Log::Dispatch::* system, and automatically rotating them according to
different constraints. This is basically a Log::Dispatch::File wrapper with
additions. To that end the arguments
name, min_level, filename and mode
behave the same as Log::Dispatch::File. So see its man page (perldoc
Log::Dispatch::File)
The arguments size and max specify the maximum size and maximum number of log
files created. The size defaults to 10M and the max number of files defaults
to 1. If DatePattern is not defined then we default to working in size mode.
That is, use size values for deciding when to rotate.
Once DatePattern is defined FileRotate will move into time mode. Once this
happens file rotation ignores size constraints and uses the defined date
pattern constraints.
If you setup a config file using
Log::Log4perl::init_and_watch() or the
like, you can switch between modes just by commenting out the DatePattern
line.
When using DatePattern make sure TZ is defined correctly and that the TZ you use
is understood by Date::Manip. We use Date::Manip to generate our recurrences.
Bad TZ equals bad recurrences equals surprises! Read the Date::Manip man page
for more details on TZ.
DatePattern will default to a daily rotate if your entered pattern is incorrect.
You will also get a warning message.
If you have multiple writers that were started at different times you will find
each writer will try to rotate the log file at a recurrence calculated from
its start time. To sync all the writers just use a config file and update it
after starting your last writer. This will cause
Log::Dispatch::FileRotate->
new() to be called by each of the
writers close to the same time, and if your recurrences aren't too close
together all should sync up just nicely.
I initially aasumed a long runinng process but it seems people are using this
module as part of short running CGI programs. So, now we look at the last
modified time stamp of the log file and compare it to a previous occurance of
a DatePattern, on startup only. If the file stat shows the mtime to be earlier
than the previous recurrance then I rotate the log file.
We handle multiple writers using
flock().
DatePattern¶
As I said earlier we use Date::Manip for generating our recurrence events. This
means we can understand Date::Manip's recurrence patterns and the normal log4j
DatePatterns. We don't use DatePattern to define the extension of the log file
though.
DatePattern can therefore take forms like:
Date::Manip style
0:0:0:0:5:30:0 every 5 hours and 30 minutes
0:0:0:2*12:30:0 every 2 days at 12:30 (each day)
3*1:0:2:12:0:0 every 3 years on Jan 2 at noon
DailyRollingFileAppender log4j style
yyyy-MM every month
yyyy-ww every week
yyyy-MM-dd every day
yyyy-MM-dd-a every day at noon
yyyy-MM-dd-HH every hour
yyyy-MM-dd-HH-MM every minute
To specify multiple recurrences in a single string separate them with a
semicolon:
yyyy-MM-dd; 0:0:0:2*12:30:0
This says we want to rotate every day AND every 2 days at 12:30. Put in as many
as you like.
A complete description of Date::Manip recurrences is beyond us here except to
quote (from the man page):
A recur description is a string of the format
Y:M:W:D:H:MN:S . Exactly one of the colons may
optionally be replaced by an asterisk, or an asterisk
may be prepended to the string.
Any value "N" to the left of the asterisk refers to
the "Nth" one. Any value to the right of the asterisk
refers to a value as it appears on a calendar/clock.
Values to the right can be listed a single values,
ranges (2 numbers separated by a dash "-"), or a comma
separated list of values or ranges. In a few cases,
negative values are appropriate.
This is best illustrated by example.
0:0:2:1:0:0:0 every 2 weeks and 1 day
0:0:0:0:5:30:0 every 5 hours and 30 minutes
0:0:0:2*12:30:0 every 2 days at 12:30 (each day)
3*1:0:2:12:0:0 every 3 years on Jan 2 at noon
0:1*0:2:12,14:0:0 2nd of every month at 12:00 and 14:00
1:0:0*45:0:0:0 45th day of every year
0:1*4:2:0:0:0 4th tuesday (day 2) of every month
0:1*-1:2:0:0:0 last tuesday of every month
0:1:0*-2:0:0:0 2nd to last day of every month
METHODS¶
- •
- new(%p)
This method takes a hash of parameters. The following options are
valid:
- -- name ($)
- The name of the object (not the filename!). Required.
- -- size ($)
- The maxium (or close to) size the log file can grow too.
- -- max ($)
- The maxium number of log files to create.
- -- TZ ($)
- The TimeZone time based calculations should be done in. This should match
Date::Manip's concept of timezones and of course your machines
timezone.
- -- DatePattern ($)
- The DatePattern as defined above.
- -- min_level ($)
- The minimum logging level this object will accept. See the Log::Dispatch
documentation for more information. Required.
- -- max_level ($)
- The maximum logging level this obejct will accept. See the Log::Dispatch
documentation for more information. This is not required. By default the
maximum is the highest possible level (which means functionally that the
object has no maximum).
- -- filename ($)
- The filename to be opened for writing. This is the base name. Rotated log
files will be renamed filename.1 thru to filename."max". Where
max is the paramater defined above.
- -- mode ($)
- The mode the file should be opened with. Valid options are 'write',
'>', 'append', '>>', or the relevant constants from Fcntl. The
default is 'write'.
- -- autoflush ($)
- Whether or not the file should be autoflushed. This defaults to true.
- -- callbacks( \& or [ \&, \&, ... ] )
- This parameter may be a single subroutine reference or an array reference
of subroutine references. These callbacks will be called in the order they
are given and passed a hash containing the following keys:
( message => $log_message, level => $log_level )
The callbacks are expected to modify the message and then return a single
scalar containing that modified message. These callbacks will be called
when either the "log" or "log_to" methods are called
and will only be applied to a given message once.
- -- DEBUG ($)
- Turn on lots of warning messages to STDERR about what this module is doing
if set to 1. Really only useful to me.
- •
- log_message( message => $ )
Sends a message to the appropriate output. Generally this shouldn't be
called directly but should be called through the "log()" method
(in Log::Dispatch::Output).
- •
- setDatePattern( $ or [ $, $, ... ] )
Set a new suite of recurrances for file rotation. You can pass in a single
string or a reference to an array of strings. Multiple recurrences can
also be define within a single string by seperating them with a semi-colon
(;)
See the discussion above regarding the setDatePattern paramater for more
details.
TODO¶
compression, signal based rotates, proper test suite
Could possibly use Logfile::Rotate as well/instead.
AUTHOR¶
Mark Pfeiffer, <markpf at mlp-consulting dot com dot au> inspired by Dave
Rolsky's, <autarch at urth dot org>, code :-)
Kevin Goess <cpan at goess dot org> suggested multiple writers should be
supported. He also conned me into doing the time based stuff. Thanks Kevin!
:-)
Thanks also to Dan Waldheim for helping with some of the locking issues in a
forked environment.
And thanks to Stephen Gordon for his more portable code on lockfile naming.
Copyright¶
Copyright 2005-2006, Mark Pfeiffer
This code may be copied only under the terms of the Artistic License, or GPL
License which may be found in the Perl 5 source kit.
Use 'perldoc perlartistic' to see the Artistic License. Use 'perldoc perlgpl' to
see the GNU General Public License.
Complete documentation for Perl, including FAQ lists, should be found on this
system using `man perl' or `perldoc perl'. If you have access to the Internet,
point your browser at
http://www.perl.org/, the Perl Home Page.