List all apache log files greater than 1GB in /var/log/[httpd|apache2] about apache2buddy

[richardforth]

Knocked out some test code that works:

#!/usr/bin/perl
use strict;

my $log_dir;
my $size;
$log_dir = "/var/log/";  # would need logic to set $log_dir based on distribution
$size = 1024; # this is bytes, I know, for testing only!


print "Scanning $log_dir for files larger than $size bytes....\n";
opendir(my $dh, $log_dir) || die "Can't open $log_dir: $!";
while (readdir $dh) {
  if ( -s >= $size ) {
    print "$_\n";
  }
}

[richardforth]

Test code committed to a test repository for ease of testing on different platforms, to develop the logic further

https://github.com/richardforth/perl_large_file_scanner/

[richardforth]

Working, next step is to merge into the main codebase of apache2buddy under staging.

[richardforth]

There is also a working function now in checkbox4 which is ready to port across (uses File::Find):

sub systemcheck_large_logs {
	my ($logdir) = @_;
	our $ymdhms;
        my $logfile = $logdir . "/" . $ymdhms . "_large_logs.log";
        unless (open(LOG_FILE, ">$logfile")) {
                die "Can't open log file $logfile for writing: $!\n"
        }
	info_print("Scanning /var/log for files larger than 1GB:");
	my $dir = '/var/log'; # whatever you want the starting directory to be
	my @logs;
	my $logfiles_raw = find(sub {push @logs, $File::Find::name  if -s >= 1024000000;},  $dir);
	foreach my $log (@logs) {
		chomp($log);
		my $size = -s $log;
		my $humansize = sprintf "%.2f", $size/1024/1024/1024;
		bad_print_item($log . " --> " . $humansize . "GB");
		print LOG_FILE $log . " --> " . $humansize . "GB\n";
	}
	if (@logs == 0) {
		good_print_item("No large logs files were found.");
	} else {
		bad_print("Consider setting up a log rotation policy.");
	}
}