#!/bin/bash
################################################
#  script for NC cPanel autobackup monitoring  #
#                                              #
#         Created by Bogdan Kukharskiy         #
#                   Namecheap                  #
################################################

PATH=/bin:/sbin:/usr/bin:/usr/sbin:/usr/local/bin:/usr/local/sbin

verbose=0
targetstr="/backup"
agetreshold=172800  # 48 hours
ratiotreshold=$(printf "%.2f\n" 0.20)
snapagetresholdoverride=""

PROGNAME=${0##*/}
fstype="nfs"
emrgstr="nfs/$(hostname -s)"

#green='\033[0;32m'
#yellow='\033[1;33m'
#red='\033[0;31m'
#NC='\033[0m' # No Color

print_usage() {
    echo ""
    echo "Usage: $PROGNAME [-v be Verbose] [-t Target mount string to check] [-a critical Age for emergency backup partition mount ] [-r failed/success Ratio] [-s Snapshot age treshold override]"
    echo "Usage: $PROGNAME --help"
}

print_help() {
        print_usage
        echo ""
        echo "This script runs different checks on NC cPanel Autobackup (see TO-17214)"
        echo ""
        echo "-v be Verbose (should be last argument)"
        echo "-t Target mount string to check, string. Default - /backup"
        echo "-a critical Age for emergency backup partition mount, seconds. Default - 172800 (48 hours)"
        echo "-r failed/success Ratio, float. Default - 0.20"
        echo "-s Snapshot age treshold override, seconds, Default, for regular and resellers - 302400 (84 hours)"
        echo "                                               for business and premium  - 129600 (36 hours)"
        echo "--help Print this help screen"
        echo ""
        exit 3
}

while [ $# -gt 0 ]; do
        case "$1" in
        --help)
                print_help
                exit 3
                ;;

        -h)
                print_help
                exit 3
                ;;

        -t)
                targetstr=$2; shift
                ;;

        -a)
                agetreshold=$2; shift
                ;;

        -r)
                r=$2; ratiotreshold=$(printf "%.2f\n" "$r"); shift
                ;;

        -s)
                snapagetresholdoverride=$2; shift
                ;;

        -v)
                verbose=1; shift;shift
                ;;

        *)
                echo >&2 "Unknown argument: $1"
                print_usage
                exit 3
                ;;
        esac
        shift
done

# lets check if emergency partition is mounted
if [[ ${verbose} == 1 ]]; then
    echo -e "1. Checking if the emergency partition is mounted"
fi
emrgchkres=$( findmnt -rno SOURCE "/backup" -t ${fstype} 2>/dev/null )
if [[ ${emrgchkres} != *"${emrgstr}"* ]]; then
    if [[ ${verbose} == 1 ]]; then
        echo "Emergency partition is mounted: ${emrgchkres}";
        # Then let's check the age of emergency backup partition
        age=$( grep -A2 /backup /proc/self/mountstats | grep "age" | cut -d":" -f2 | tr -d '[:space:]' )

        if [[ ${verbose} == 1 ]]; then
            echo -e "1.1. Checking the age of emergency backup partition"
            echo "   Age of ${emrgchkres} : ${age} sec ($((age / 3600)) hours)"
        fi
        if [[ ${age} -lt ${agetreshold} ]]; then 
            if [[ ${verbose} == 1 ]]; then echo -e "Emergency partition age is OK"; fi
        else echo -e "Emergency partition age > $((agetreshold / 3600)) hours ($((age / 3600)) hours)"; exit 2
        fi
    fi
else
    if [[ ${verbose} == 1 ]]; then echo -e "OK. Emergency partition is not mounted"; fi
fi


# find /backup and compare with fstab
sourcestr="$(grep "/backup" /etc/fstab | grep -v "#" | awk '{ print $1 }')"
mntchkres=$( findmnt -rno SOURCE "${targetstr}" -t ${fstype} 2>/dev/null )
if [ "${mntchkres: -1}" == "/" ]; then  # remove last / if presents
    mntchkres="${mntchkres%?}"
fi

if [[ ${verbose} == 1 ]]; then
    echo -e "2. Checking mountpoint ${targetstr} of fstype ${fstype} "
    echo "   Source string from /etc/fstab : ${sourcestr}"
    echo "   Findmnt results               : ${mntchkres}"
fi
if [[ "${mntchkres}" == "${sourcestr}" ]]; then
    if [[ ${verbose} == 1 ]]; then echo -e "OK. Mountpoint ${targetstr} is mounted according to fstab"; fi
else echo -e "Mountpoint ${targetstr} is not mounted"; exit 2
fi


# testing connection to HAPI and rsync
if [[ ${verbose} == 1 ]]; then
    echo -e "3. Testing connection to HAPI and rsync (with /root/bin/nc_cp_backup_with_snapshot.sh --test)"
fi
tstchkres=$( /root/bin/nc_cp_backup_with_snapshot.sh --test )
if [[ ${tstchkres} == *"Connection to HAPI: test OK"* ]] && [[ ${tstchkres} == *"Connection to rsync: test OK"* ]] ; then
    if [[ ${verbose} == 1 ]]; then echo -e "OK"; echo "${tstchkres}"; fi
else echo -e "Connection to HAPI or rsync failed"; echo "${tstchkres}"; exit 2
fi


# Checking snapshot freshness
if [[ ${verbose} == 1 ]]; then
    echo -e "4. Checking snapshot freshness (via log at /var/log/nc_audit/nc_cp_backup.log)"
fi

if [ -z "${snapagetresholdoverride}" ]; then
    if [[ $(hostname -s) == *"business"* ]] || [[ $(hostname -s) == *"premium"* ]] ; then
        snapagetreshold=129600 # 36 hours
    else snapagetreshold=302400 # 84 hours
    fi
else snapagetreshold=${snapagetresholdoverride}
fi

rawsnapshotage=$( grep -B 1 "snapshot creation api has been triggered" /var/log/nc_audit/nc_cp_backup.log | tail -1 | awk '{print$1" "$2" "$3}' )
snapshotage=$( date -d"${rawsnapshotage}" +%s)
snapdiff=$(($(date +%s) - snapshotage))

if [[ ${snapdiff} -lt ${snapagetreshold} ]]; then 
    if [[ ${verbose} == 1 ]]; then echo -e "OK. Snapshot is fresh (created at ${rawsnapshotage})"; fi
else echo -e "Snapshot was created more than $((snapagetreshold / 3600)) hours ago ($((snapdiff / 3600)))"; exit 2
fi


# Checking success/failed ratio (should be less than 0.2%)
if [[ ${verbose} == 1 ]]; then
    echo -e "5. Checking the success/failed ratio of backed up accounts"
fi

ratiores=$( grep -B 1 "snapshot creation api has been triggered" /var/log/nc_audit/nc_cp_backup.log | tail -2 | grep "failed:" )

if [[ ${#ratiores} -gt 0 ]]; then
    if [[ ${verbose} == 1 ]]; then echo "Failed accounts found. Calculating the ratio"; fi
    totalacc=$( echo "${ratiores}" | awk '{print$11}' )
    failedacc=$( echo "${ratiores}" | awk '{print$16}' )
    ratio=$( awk -v f="${failedacc}" -v t="${totalacc}" 'BEGIN {print 100*f/t}' )
    if (( $( echo "${ratio} > ${ratiotreshold}" | bc -l ) )); then
        echo -e "Failed ratio is more than treshold (${ratio} > ${ratiotreshold}, failed ${failedacc} from ${totalacc}) "; exit 2
    else 
        if [[ ${verbose} == 1 ]]; then echo "Mmm..OK. Failed ratio is not more than treshold (${ratio} < ${ratiotreshold}, failed ${failedacc} from ${totalacc})"; fi
    fi
else 
    if [[ ${verbose} == 1 ]]; then echo -e "OK. No failed accounts. Success ratio - 100% "; fi
fi

echo "All OK"