#! /bin/bash

###
# Nagios Process monitoring plugin.
#
# This plugin check the CPU and memory usage of a process.
#
# @AUTHOR: Patrik Dufresne (http://patrikdufresne.com)
# Copyright 2015 Patrik Dufresne
# Last modified 2015-05-25
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, version 2 of the License.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program.  If not, see <http://www.gnu.org/licenses/>.
###

# You should provide a meaningful VERSION
VERSION=0.1
# Who can be contacted about this?
AUTHOR="Patrik Dufresne"
# Name what is being checked to be printed out next to OK/WARNING/CRITICAL/UNKNOWN
SERVICE="PROC"

# Replacement for the exit function, will cleanup any tempfiles or such
# before exiting.
function cleanup {
  exit $1
}

declare -rx PROGNAME=${0##*/}
declare -rx PROGPATH=${0%/*}/

if [ -r "${PROGPATH}utils.sh" ] ; then
  source "${PROGPATH}utils.sh"
else
  echo "Can't find utils.sh. This plugin needs to be run from the same directory as utils.sh which is most likely something like /usr/lib/nagios/plugins or /usr/lib64/nagios/plugins"
  printf "Currently being run from %s\n" "$PROGPATH"
  # Since we couldn't define STATE_UNKNOWN since reading utils.sh
  # failed, we use 3 here but everywhere else after this use cleanup $STATE
  cleanup 3
fi

# Set STATE to UNKNOWN as soon as we can (right after reading in util.sh
# where the STATES are defined)
STATE=$STATE_UNKNOWN

# provide a quick one liner of how to use the program
function usage {
  printf " %s %s for Nagios - Usage %s \
-p pidfile
[-c <warning cpu % usage>] [-C <critical cpu % usage>] \
[-m <warning memmory threshold>] [-M <critical memory threshold>] \
[-t timeout] [-v [-v [-v]]]\n" "$PROGNAME" "$VERSION" "$PROGNAME"
  cleanup $STATE_UNKNOWN
}

# provide detailed explanations of the command line syntax
function longhelp {
  # put your long help here
  printf "%s plugin version %s for Nagios by %s
  -h, --help                   Display this message.
  -c, --warning-cpu=val        Set the warning cpu usage threshold (in %).
  -C, --critical-cpu=val       Set the critical cpu usage threshold (in %).
  -m, --warning-mem=val        Set the warning memmory usage threshold (in MB).
  -M, --critical-mem=val       Set the critical memory usage threshold (in MB).
  -p, --pidfile=file           Set the pidfile to monitor.
  -t, --timeout=sec            Set script timeout in seconds.
  -v, --verbose                Up the verbosity level by one.
  --verbosity=val              Set the verbosity level to val.
  -V, --version                Print version information.
  --range_help                 Explain threshold ranges.
" "$PROGNAME" "$VERSION" "$AUTHOR"
  cleanup $STATE_UNKNOWN
}

# explanatory function you probably want to keep
function range_help {
  printf "
The format for ranges in Nagios can be confusing and it isn't always followed.

[@]start[:[end]]

Here are some example ranges:

Range   |  Generate an alert if value is    |  In English
--------+-----------------------------------+---------------------------------
10      |  outside the range of {0 .. 10}   |  Greater than 10
@10     |  inside the range of {0 .. 10}    |  Less than or equal to 10
10:     |  outside {10 .. ∞}                |  Greater than 10
~:10    |  outside the range of {-∞ .. 10}  |  Less than 10 including negative
10:20   |  outside the range of {10 .. 20}  |  Between 10 and 20
@10:20  |  inside the range of {10 .. 20}   |  Anything from 10 to 20
10      |  outside the range of {0 .. 10}   |  Greater than 10 or less than 0

Formal Rules:
1. start ≤ end
2. start and ":" is not required if start=0
3. if range is of format \"start:\" and end is not specified, end is infinity
4. to specify negative infinity, use "~"
5. alert is raised if metric is outside start and end range (inclusive)
6. if range starts with "@", then alert if inside this range (inclusive)
    10      < 0 or > 10, (outside the range of {0 .. 10})
    10:     < 10, (outside {10 .. ∞})
    ~:10    > 10, (outside the range of {-∞ .. 10})
    10:20   < 10 or > 20, (outside the range of {10 .. 20})
    @10:20  ≥ 10 and ≤ 20, (inside the range of {10 .. 20})
    10      < 0 or > 10, (outside the range of {0 .. 10})

More help at http://nagiosplug.sourceforge.net/developer-guidelines.html
"
  cleanup $STATE_UNKNOWN
}

# use getopt, trust me on this one. It's the easiest way
getopt -T
if [ $? -ne 4 ] ; then
  printf "%s: getopt is in compatibility mode.\n" "$SCRIPT"
  cleanup $STATE_UNKNOWN
fi

# Tell it which switches and longswitches you'll take and place a trailing
# colon (:) on the ones take arguments. Nagios guidelines require you to
# use all the ones specified below with the exception of --verbosity which I've
# added to circumvent the awkward -v -v -v syntax. Getopt takes care of
# positional parameters and errors for missing expected arguments so we can
# shift later without checking
RESULT=`getopt --name "$SCRIPT" --options "-h,-V,-v,-t:,-c:,-C:,-m:,-M:,-f:,-F:,-p:" \
--longoptions "help,version,verbose,verbosity:,warning-cpu:,critical-cpu:,warning-memory:,critical-memory:,warning-fileopen:,critical-fileopen:,timeout:,pidfile:" -- "$@"`

# make the result of getopt your new argument list ($@)
eval set -- "$RESULT"

# all scripts should have a mechanism to terminate themselves if they are
# running for too long. Scripts you might think of as innocuous could end
# up waiting forever on I/O, especially if a disk is failing
declare -i TIMELIMIT=15
# Nagios defines behavior for VERBOSITY 0 (default) through 3
declare -i VERBOSITY=0

while [ $# -gt 0 ] ; do
  case "$1" in
    -h | --help)
      longhelp;;
    -V | --version)
      print_revision "$PROGNAME" "$VERSION"
      cleanup $STATE;;
    -v | --verbose)
      VERBOSITY=$(($VERBOSITY + 1));;
    --verbosity)
      shift
      VERBOSITY=$1;;
    -c | --warning-cpu)
      shift
      CPU_WARNING=$1;;
    -C | --critical-cpu)
      shift
      CPU_CRITICAL=$1;;
    -m | --warning-memory)
      shift
      MEM_WARNING=$1;;
    -M | --critical-memory)
      shift
      MEM_CRITICAL=$1;;
    -p | --pidfile)
      shift
      PIDFILE=$1;;
    -t | --timeout)
      shift
      TIMELIMIT=$1;;
    --)
      shift
      break;;
    *)
      echo "Option $1 not supported. Ignored." >&2;;
  esac
  shift
done

#Verbosity level  Type of output
#0      Single line, minimal output. Summary
#1      Single line, additional information (eg list processes that fail)
#2      Multi line, configuration debug output (eg ps command used)
#3      Lots of detail for plugin problem diagnosis
if [ $VERBOSITY -gt 2 ] ; then
  shopt -o -s xtrace
fi

# what needs to happen in the event of a timeout
function timeout {
  echo "UNKNOWN - script timed out after $TIMELIMIT seconds."
  cleanup $STATE_UNKNOWN
}

function check_range {
  # # positive values only
  if [ ! -z "$1" ] ; then
    WARNFORMAT=`echo "$1" | grep -c '^@\?\([0-9]\+:[0-9]*\|[0-9]\+\)$'`
    if [ $WARNFORMAT -lt 1 ] ; then
      echo "Please check the format of your warning and critical thresholds."
      range_help
    fi
  fi
  if [ ! -z "$2" ] ; then
    CRITFORMAT=`echo "$2" | grep -c '^@\?\([0-9]\+:[0-9]*\|[0-9]\+\)$'`
    if [ $CRITFORMAT -lt 1 ] ; then
      echo "Please check the format of your warning and critical thresholds."
      range_help
    fi
  fi
}

check_range "$CPU_WARNING" "$CPU_CRITICAL"
check_range "$MEM_WARNING" "$MEM_CRITICAL"

# since we've processed the options which potentially set the timeout limit,
# we can setup a timeout trap now
trap timeout USR1
  # what we're doing here sending a USR1 signal back to this process which
  # we just set a trap to catch and run the timeout function the syntax of
  # this is important and very odd - if you know of a better way to do this, 
  # please email me what we're doing is starting another process in the
  # background that sleeps for TIMELIMIT seconds and then uses pgrep when
  # it 'wakes up' to see if a process with our number, name and user exists,
  # only then will the USR1 signal be sent we have to use pgrep so that we
  # don't sent a USR1 signal to just any program. The only risk we run with
  # this is sending USR1 to another instance of this script that just happens
  # to get assigned the same process ID it should be reasonable to assume
  # that your Nagios check interval is greater than the specified timeout
  # still, if you havea better idea...
  ( sleep $TIMELIMIT; if [ `pgrep -U $USER -f "$SCRIPT" | grep -c ^$$$` -gt 0 ] ; then kill -USR1 $$ ; fi; ) </dev/null &>/dev/null &

  # Get PID to monitor.
  PID=""
  if [ -e "$PIDFILE" ]; then
    PID=$(cat "$PIDFILE")
  fi
  # If process doesn't exists return Unknown.
  if [ ! -e "/proc/$PID" ]; then
    echo "UNKNOWN - process not running"
    cleanup $STATE_UNKNOWN
  fi

  # Get the sensor's value
  DATA=$(ps -p "$PID" efho vsize,rss,size,%cpu,nlwp)
  if [ $VERBOSITY -gt 2 ] ; then
    echo "$DATA"
  fi
  VSIZE_KB=$(echo "$DATA" | awk '{print $1;}')
  VSIZE_MB=$(expr $VSIZE_KB / 1024)
  RSS_KB=$(echo "$DATA" | awk '{print $2;}')
  RSS_MB=$(expr $RSS_KB / 1024)
  CPU=$(echo "$DATA" | awk '{print $4;}')
  THREAD=$(echo "$DATA" | awk '{print $5;}')

trap - USR1

function check_value {
  if [ -z "$1" ]; then
    return 0
  fi

  # If the range starts with an @, alert if value is inside the range,
    # otherwise alert if value is outside of range.
  INSIDE=`echo "$1" | grep -c '^@'`
  RANGE=`echo "$1" | sed 's/^@//'`

  # Start is anything left of the colon or 0.
    # End is anything right of the colon or the whole string if there's no
    # colon or infinity if there is a colon and nothing to the right of it

  # is there a colon?
  PARTS=`echo "$RANGE" | awk -F : '{ print NF }'`
  if [ $PARTS -gt 1 ] ; then
    START=${RANGE%%:*}
    END=${RANGE##*:}
  else
    START=0
    END=$RANGE
  fi

  # 4. to specify negative infinity, use "~"
  if [ "$START" == "~" ] ; then
    START=-999999999
  fi

  if [ -z "$END" ] ; then
    END=999999999
  fi

  if [ $START -gt $END ] ; then
    echo "In threshold START:END, START must be less than or equal to END"
    range_help
  fi

  # if the range starts with an @, alert if value is inside the range,
  # otherwise alert if value is outside of range all ranges are inclusive
  # of endpoints so we use less than or equal on the inside and just less
  # than on the outside
  if [ "$INSIDE" -gt 0 ] ; then
    if [ $(echo "$START <= $2 && $2 <= $END" | bc) -eq 1 ] ; then
      return 1
    fi
  else
    if [ $(echo "$2 < $START || $END < $2" | bc) -eq 1 ] ; then
      return 1
    fi
  fi

  return 0
}

# Check critical threshold
STATE=$STATE_OK
check_value "$CPU_CRITICAL" "$CPU"
if [ $? -gt 0 ] ; then
  STATE=$STATE_CRITICAL
fi
check_value "$MEM_CRITICAL" "$RSS_MB"
if [ $? -gt 0 ] ; then
  STATE=$STATE_CRITICAL
fi

# Check warning treshold
if [ $STATE -ne $STATE_CRITICAL ]; then
  if [ $? -gt 0 ] ; then
    STATE=$STATE_WARNING
  fi
  check_value "$MEM_WARNING" "$RSS_MB"
  if [ $? -gt 0 ] ; then
    STATE=$STATE_WARNING
  fi
fi

# STATE - Message | 'label'=value[unit of measure];[warn];[crit];[min];[max]
OUT="vsize: ${VSIZE_MB} MiB, rss: ${RSS_MB} MiB, cpu: $CPU%, threads: $THREAD | vsize=${VSIZE_KB}KiB rss=${RSS_KB}KiB cpu=${CPU}% threads=$THREAD"

case $STATE in
  $STATE_OK)
    printf "%s OK - %s\n" "$SERVICE" "$OUT";;
  $STATE_WARNING)
    printf "%s WARNING - %s\n" "$SERVICE" "$OUT";;
  $STATE_CRITICAL)
    printf "%s CRITICAL - %s\n" "$SERVICE" "$OUT";;
  $STATE_UNKNOWN)
    printf "%s UNKNOWN - %s\n" "$SERVICE" "$OUT";;
esac

cleanup $STATE