in reply to Re^5: Generic Data Collection
in thread Generic Data Collection

Excellent, POJ. This works well. I need to piece through your code to understand exactly what is doing, but I wondered if you could expand on this approach to help me in my next step. I need to identify the ten oldest PIDS and then them through another command. For simplicity, I'll just select the 2 oldest for this example. Once I identify those PIDS, I need to run the command: 

lmutil lmremove -h $PRODUCT $flexlmserver 27000 $PID[x]

[download]

Where $PIDx would be the two oldest (or whatever # I want to select) PIDS. I'm sure this can be done using your approach, but I have no idea of the syntax. Thanks so much for your time.

Replies are listed 'Best First'.
Re^7: Generic Data Collection
by poj (Abbot) on Mar 31, 2013 at 19:23 UTC
    This simple example assumes you don't want to capture the output of the command or check for success 
    while ( <DATA> ) {
  my @wanted = ( split )[ 0, 5, 9 ];
  $wanted[1] =~ s/\D//g;
  $wanted[2] = sprintf "%05s",$wanted[2]; 
  push @parsed, \@wanted;
}

my @sorted_by_time = sort { $a->[2] cmp $b->[2] } @parsed;

# oldest
my $LIMIT = 2;
my $count = 0;
for (@sorted_by_time){
  my $PID = $_->[1];
  system("lmutil lmremove -h $PRODUCT $flexlmserver 27000 $PID");
  last if ++$count >= $LIMIT 
}

    [download]
    poj
[reply]
[d/l]

      I'm not worthy! I'm not worthy! Thanks, POJ. This is exactly what I need and it is working. I would like to capture output and check for success. I will work on ways to do that today, but if you have any suggestions on that, I would greatly appreciate it. You've been a great help and I have learned a lot. Thanks so much.

[reply]
[reply]

      I have another question about this code. Today was I was able to test this live and I have discovered there are cases where my data is not aligned correctly. I have been able to use a regex to address the problem, but the data set I am correcting is not getting loaded into the array. Here's the code: 

      my $license_cmd = "lmutil lmstat -f $PRODUCT";
my @parsed;
      
open LICENSES, "$license_cmd |" || die "Can't execute ($license_cmd) \
+n$!\n";
    
while (<LICENSES>) 
    {
    my $data_line = $_;

    # Post FlexLM Version 7.0 - Find max number of licenses available
    if ($data_line =~ /Total of (\d+) licenses issued/)
        { $MAX_LIC = $1; }

    # Identify the lines of output that contain user license info
    elsif ($data_line  =~ /\, start /) 
        {
        # Remove leading spaces from the lines
        $data_line =~ s/^\s+//g;
        # Remove duplicate data from the lines (web logons)
        $data_line =~ tr/-//d; # Remove hyphens to remove the duplicat
+e server name
        $data_line =~ s/(\b\w+\b)(\s*\1)+/$1/g; # Remove duplicate ser
+ver name
        print "\n$data_line"; # <-- test results of data changes
        # Load user name, PID and Start Time into array USAGE_INFO
        my @USAGE_INFO = ( split)[ 0, 5, 9 ];
        $USAGE_INFO[1] =~ s/\D//g;
        $USAGE_INFO[2] = sprintf "%05s",$USAGE_INFO[2]; 
        push @parsed, \@USAGE_INFO;

        }
    }
close LICENSES;

      [download]

      The problem appears to be the line: 

      my @USAGE_INFO = ( split )[ 0, 5, 9 ];

      [download]

      ...is getting it's data directly from the raw data produced by the license command in the open file LICENSES. This makes sense, since I am massaging the data in $data_line. I have read up on SPLIT and array creations and have tried the following to build the @USAGE_INFO array with $data_line, but I can't seem to get it to work: 

      my @USAGE_INFO = ( split )[ 0, 5, 9 ], $data_line;
my @USAGE_INFO = ( split ($data_line) )[ 0, 5, 9 ];

      [download]

      Any further help would be greatly appreciated. Thanks!

[reply]
[d/l]
[select]
        try 
        my @USAGE_INFO = ( split /\s+/,$data_line )[ 0, 5, 9 ];

        [download]
        poj
[reply]
[d/l]

In Section Seekers of Perl Wisdom