<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> <html> <head> </head> <body bgcolor="#ffffff" text="#000099"> I'm using the "check_multiprocs" add on available from NagiosExchange. With it configured for a host, if I drill in on the test in the web interface, I see: <blockquote>Current Status: OK (for 0d 8h 51m 46s) Status Information: PROCS OK cron ok: 1 syslogd ok: 1 xntpd ok: 1 sshd ok: 7 sendmail ok: 2 rpcbind ok: 1 inetd ok: 1 lockd ok: 1 statd ok: 1 nfsd ok: 1 rpc.metad ok: 1 mountd ok: 2 </blockquote> However, last night one of the services was restarted (probably syslogd) due to cron. I received an alert via email about it, but all I saw was: <blockquote>***** Nagios ***** Notification Type: PROBLEM Service: PROCS Host: <server> [truncated by me] Address: <IP address> [truncated by me] State: WARNING Date/Time: Tue Mar 31 06:04:47 EDT 2009 Additional Info: PROCS WARNING </blockquote> What I need is to see which processes are down in the "Additional Info:" field when its emailed. Anyone know how I would go about this? Unfortunately, I'm not a perl guy... I've included the script below... # libexec> cat check_multiprocs #!/usr/bin/perl -w # Check Multiprocs use Getopt::Long; sub fail($); my ($opt_c,$opt_h); my $PROGNAME = "PROCS"; my $STATE = "OK"; # Initially, we're fine my $RETSTRING = ""; # SYSTEM COMMANDS #my $PS = "/bin/ps ax"; my $PS = "/bin/ps -aef"; my $GREP = "/bin/egrep"; my %RETVALUE = ( 'OK' => 0, 'WARNING' => 1, 'CRITICAL' => 2 ); GetOptions ("c=s" => \$opt_c, "config=s" => \$opt_c, "h" => \$opt_h, "help" => \$opt_h); if ($opt_h) { print_help(); exit 3; } ($opt_c) || fail("No config file specified"); my $inFile = $1 if ($opt_c =~ /([-.\/a-zA-Z0-9]+)/); open (INFILE,"<$inFile") || fail("Could not open file: $inFile"); while (<INFILE>) { my ($wmin, $wmax, $cmin, $cmax); next if (/^#|^$/); # No comments my @line = split(/;/); ($#line == 5) || fail("Error parsing $inFile:\n@line"); # We need 6 arguments per line my $host = $line[0]; my $proc = $line[1]; $wmin = ($line[2] =~ /[0-9]+/ ? $line[2] : 1); # Get an actual number for the mins/maxes $wmax = ($line[3] =~ /[0-9]+/ ? $line[3] : 1000); $cmin = ($line[4] =~ /[0-9]+/ ? $line[4] : 1); $cmax = ($line[5] =~ /[0-9]+/ ? $line[5] : 1000); if ($host eq "localhost" || $host eq $ENV{'HOSTNAME'}) { # Now, check each process my $procStr; if ($proc =~ /^".*"$/) { # How does our process string look? $procStr = $proc; # For some reason this works, but the other way doesn't } else { $procStr = "'$proc( |\$)+'"; } my @OUT = `$PS | $GREP -e $procStr | $GREP -v grep`; my $count = $#OUT + 1; if ($count < $cmin) { # Be more specific for crit procs $STATE = "CRITICAL"; $RETSTRING .= "\n$proc, too few processes: $count"; } elsif ($count > $cmax ) { $STATE = "CRITICAL"; $RETSTRING .= "\n$proc, too many processes: $count"; } elsif ($count < $wmin || $count > $wmax) { if ( $STATE ne "CRITICAL" ) { $STATE = "WARNING"; } $RETSTRING .= "\n$proc exceeded warn threhold: $count"; } else { $RETSTRING .= "\n$proc ok: $count"; } } } if ( $RETSTRING eq "" ) { $RETSTRING = "\nAll monitored processes OK" . $RETSTRING; } print "$PROGNAME $STATE $RETSTRING\n"; exit $RETVALUE{$STATE}; ############################################################ # # SUBROUTINE DEFINITIONS # ############################################################ sub fail($) { print "Error: @_\n"; print_usage(); exit 3; } sub print_usage { print "Usage: $PROGNAME -c <config>\n"; return; } sub print_help { print "$PROGNAME, Revision: 1.0\n"; print "\n"; print_usage(); print " -c, --config=config file Configuration file. Processes should be in the format: <process>;<min>;<max> "; print "\n"; return; } <pre class="moz-signature" cols="72">-- A. Davis Email: <a class="moz-txt-link-abbreviated" href="mailto:nccomp@gmail.com">nccomp@gmail.com</a> "There is no limit to what a man can accomplish if he doesn't care who gets the credit." - Ronald Reagan </pre> </body> </html>