ClearCase Support: Automate Checking of ClearCase Server Log Files

On a daily basis, I automatically check all of the ClearCase log files for NEW error and warning messages on each of the ClearCase VOB and Views servers that I am responsible for. I accomplish this by using a simple UNIX Perl script that can be found down below. Here are the two primary features of this script:

Reports all new error messages in any ClearCase log file, since the last time this script was executed.

Reports all new warning messages in any ClearCase log file, since the last time this script was executed.

From my experience, I recommend running this script about once a day via cron on each VOB and View server that you have. I also recommend manually running this script when users are reporting server problems.

Here is how to execute this script.

$ check_clearcase_logs.pl

You can download this Perl script here, or you can copy-n-paste from here:

#!/usr/local/bin/perl
 
 
 # This script was written for www.philforhumanity.com.
 # If you have any questions, issues, or enhancement requests, please contact us via the website. 
 # Use this script at your own risk.
 
 
 # Local Variables
 my $CP = "/bin/cp";
 my $DIFF = "/bin/diff";
 my $each_remove;
 my $file;
 my $first;
 my $found;
 my $GREP = "/bin/grep";
 my @ignore    = ("Warning: Version mismatch, software:3.0, packet:1.2",   # Don't care and very common MultiSite warning
                  "lock manager already running",                          # Don't care, admin error
                  "Unable to contact albd_server on host",                 # Common issue
                  "unable to contact the albd on host",                    # Common issue
                  "Moving object to vob lost\\+found directory",           # Don't care, user error 
                  "is not a sync. packet",                                 # Common issue
                  "Warning: Object .* no longer referenced.",              # Don't care
                  "Warning: To avoid name conflict",                       # Don't care
                  "has been reformatted; please exit and re-enter",        # Don't care
                  "directory not selected in configuration specification", # Don't care, user error messge
                  "error in the config specification for view"             # Don't care, user error message
                  );
 my $issue;
 my @issues_found = ();
 my @issues_found2 = ();
 my @log_files = ("/var/adm/atria/log/*_log", "/var/adm/atria/log/sync_logs/*");
 my @look_for  = ("error", "warning");
 my $look_for_type;
 my $loop;
 my $save_file = "/tmp/check_clearcase_logs.txt";
 my $second;
 my $TOUCH = "/bin/touch";
 my @tmp;
 
 
 # Scan log files for any issues
 foreach $file (@log_files)
 {
   foreach $look_for_type (@look_for)
   {
     @issues_found = (@issues_found, `$GREP -i $look_for_type $file`);
   }
 }
 
 
 # Remove issues that have been defined to be ignored.
 foreach $issue (@issues_found)
 {
   $found = 0;
 
   
   # For each type to remove
   foreach $each_remove (@ignore)
   {
     # Search for ignored lines
     @tmp = split (/$each_remove/, $issue);
   
   
     # If found, then tag to remove
     if (scalar(@tmp) != 1)
     {
       $found = 1;
     }
   }
 
 
   # Keep only lines that were not tagged to be removed.
   if ($found == 0)
   {
     # Remove date and time from the line, so that sorting and unique-ing each line will produce less output to investigate
     @tmp = split (/[0-9][0-9]\/[0-9][0-9]\/[0-9][0-9] [0-9][0-9]:[0-9][0-9]:[0-9][0-9]/, $issue);
     if (scalar(@tmp) != 1)
     {
       $issue = @tmp[0] . @tmp[1];
     }
 
 
     #Remove process # from the line, so that unique'ing each line will produce less output to investigate
     @tmp = split (/$[0-9]+$/, $issue);
     if (scalar(@tmp) != 1)
     {
       $issue = @tmp[0] . @tmp[1];
     }
 
 
     # Save issue
     @issues_found2 = (@issues_found2, $issue);
   }
 }
 @issues_found = @issues_found2;
 
 
 # Sort output all lines
 @issues_found = sort @issues_found;
 
 
 # Remove duplicate lines, aka unique-ing 
 @issues_found2 = (@issues_found[0]);
 for ($loop=1; $loop <= scalar(@issues_found)-1; $loop++)
 {
   $first = @issues_found[$loop];
   $second = @issues_found2[scalar(@issues_found2)-1];
 
 
   # Store only unique drive names
   if ($first ne $second)
   {
     @issues_found2 = (@issues_found2, @issues_found[$loop]);
   }
 }
 @issues_found = @issues_found2;
 
 
 # Save output to a temp file
 unless (open (OUT, "> $save_file.tmp"))
 {
   die "Sorry, could not open temp file.\n";
 }
 foreach $issue (@issues_found)
 {
 	print OUT $issue;
 }
 close (OUT);
 
 
 # Test if output file exists, if not make it.
 if (! (-e "$save_file") )
 {
   print `$TOUCH $save_file`;
 }
 
 
 # Print the difference of the two files, but only the changes added to the tmp file
 print `$DIFF  $save_file  $save_file.tmp `;
 
 
 # Copy tmp file to final file
 print `$CP  $save_file.tmp  $save_file`;
 
 
 # Remove the tmp file
 unlink ("$save_file.tmp");
 
 
 # Exit
 exit 0;

This script was written for www.philforhumanity.com.

by Phil B.

A Guide for the Survival of Humankind, and Improving the World, Society, and Yourself!

ClearCase Support: Automate Checking of ClearCase Server Log Files

Related Articles
» ClearCase Support: Automate Checking of ClearCase VOBs
» ClearCase Support: Automate Checking of ClearCase Shipping Bays on VOB Servers
» ClearCase Support: Automate Checking of ClearCase Views

A Guide for the Survival of Humankind, and Improving the World, Society, and Yourself!

ClearCase Support: Automate Checking of ClearCase Server Log Files

Related Articles » ClearCase Support: Automate Checking of ClearCase VOBs » ClearCase Support: Automate Checking of ClearCase Shipping Bays on VOB Servers » ClearCase Support: Automate Checking of ClearCase Views

Related Articles
» ClearCase Support: Automate Checking of ClearCase VOBs
» ClearCase Support: Automate Checking of ClearCase Shipping Bays on VOB Servers
» ClearCase Support: Automate Checking of ClearCase Views