Logstash Progress

#!/bin/bash

FILES_TO_BE_PARSED="/var/log/remotelogs/*-*.log*"
SINCE_DB_FILES="/opt/logstash/data/plugins/inputs/file/.sincedb_[a-z]*"

function posInFile() {
    inode=$1
    pid=$2
    fdnr=$3
    filename=$4
    filesize=$5

    fd=$(echo "$fdnr" | grep -o '[0-9]\+')
    readpos=$(grep 'pos' /proc/${pid}/fdinfo/${fd}| grep -o '[0-9]\+')

    echo "${inode} ${readpos} ${filesize} ${filename}" | awk '{ printf "%-30s %.1f%% %.0f %.0f %ld\n", $4, 100 * $2 / $3, $3, $2, $1 }'
}

function readPositions() {
    while read -r line; do
        posInFile ${line}
    done < ${tmpPositionsOfFilesToBeParsed}
}

function addIfFileNotOpen() {
    inode=$1
    filename=$2
    filesize=$3

    inode_exists=$(grep "^${inode} " ${tmpPositionsOfFilesToBeParsed})
    if [[ -z "${inode_exists}" ]]; then
        # Gets the biggest file position from sincedb files
        readpos=$(grep -h "^${inode} " ${SINCE_DB_FILES} | sort -rn -k 4,4 | head -n1 | awk '{print $4}')

        if [[ -z "${readpos}" ]]; then
            # File position not saved to sincedb
            echo "${filename} 0.0% ${filesize} 0 ${inode}"
        else
            echo "${inode} ${readpos} ${filesize} ${filename}" | awk '{ printf "%-30s %.1f%% %.0f %.0f %ld\n", $4, 100 * $2 / $3, $3, $2, $1 }'
        fi
    fi
}

function addUnopenedFiles() {
    while read -r line; do
        addIfFileNotOpen ${line}
    done < ${tmpFilesToBeParsedInfo}
}

LOGSTASH_PID=$(ps ax | grep logstash | grep java | awk '{print $1}')

if [[ -n "${LOGSTASH_PID}" ]]; then
    # Creates temporary files
    tmpLogstashReadHandles=$(mktemp); tmpFilesToBeParsedInfo=$(mktemp); tmpPositionsOfFilesToBeParsed=$(mktemp)

    # Gets the INODE, PID and FD
    su logstash -s /bin/bash -c "lsof -n -p ${LOGSTASH_PID}" | grep ' [0-9]\+r' | awk '{print $8" "$2" "$4}' | sort > ${tmpLogstashReadHandles}

    # Gets the INODE, File name and File size in bytes
    stat -c "%i %n %s" ${FILES_TO_BE_PARSED} | sort > ${tmpFilesToBeParsedInfo}

    # Appends the Logstash PID and FD to the lines that contains the INODE, File name and File Size. Remove lines with non-matching inode numbers
    join ${tmpLogstashReadHandles} ${tmpFilesToBeParsedInfo} > ${tmpPositionsOfFilesToBeParsed}

    # Obtain the Filename, %complete, file size, scanned size and inode nr from the files that are currently read by Logstash
    results=$(readPositions)

    # Obtain the Filename, N/A, file size, N/A and inode nr from the files that are not currently being read by Logstash
    # These files may include files that were previously scanned by Logstash and written to the sincedb file
    unopenedFiles=$(addUnopenedFiles)

    totalsize=$(cat ${tmpFilesToBeParsedInfo} | awk '{ sum += $3; }
         END { printf "%.0f", sum; }' "$@")
    scannedsize=$(echo "$results
$unopenedFiles" | awk '{ sum += $4; }
         END { printf "%.0f", sum; }' "$@")
    totalsummary=$(echo "$totalsize $scannedsize" | awk '{printf "%.1f%%", 100 * $2 / $1 }')

    # Cleanup temporary files
    rm -f ${tmpLogstashReadHandles} ${tmpFilesToBeParsedInfo} ${tmpPositionsOfFilesToBeParsed}

    # Output positions
    echo "<Filename> <% Complete> <Total Size> <Scanned Size> <inode>"
    echo "$results"
    echo "$unopenedFiles"
    echo "Total Summary: $scannedsize/$totalsize = $totalsummary"

else
    >&2 echo "Error: Unable to obtain Logstash PID"
fi