calc-import-packetloss.sh

#!/bin/bash

for host in cp104{1,2,3,4}; do
    first_seq=$(head -n 1 $host.udp2log.seq | awk '{print $1}') # && echo $first_seq
    last_seq=$(tail -n 1 $host.hdfs.seq | awk '{print $1}') #  && echo $last_seq

    should_be=$(echo "$last_seq - $first_seq + 1" | bc) #&& echo $should_be

    # count lines in blog.hdfs.seq file between $first_seq and $last_seq
    hdfs_count=$(sed -n -e "/^$first_seq/,/^$last_seq/p" $host.udp2log.seq | wc -l) #&& echo $hdfs_count
    udp2log_count=$(sed -n -e "/^$first_seq/,/^$last_seq/p" $host.udp2log.seq | wc -l) #&& echo $udp2log_count

    echo "----"
    echo "$host:"
    echo "  should_be:     $should_be"; echo "  udp2log_count: $udp2log_count"; echo "  hdfs_count:    $hdfs_count";
done

stats.yaml

cp1041:
  should_be:     2996340
  udp2log_count: 2558004
  hdfs_count:    2558004
cp1042:
  should_be:     2992324
  udp2log_count: 2554088
  hdfs_count:    2554088
cp1043:
  should_be:     3004241
  udp2log_count: 2565111
  hdfs_count:    2565111
cp1044:
  should_be:     3005422
  udp2log_count: 2566392
  hdfs_count:    2566392

stats_percent.txt

cp1044:
    total:      3005422
    udp2log:  2,566,392 (-14.608%, -439,030)
    hdfs:     2,566,392 (-14.608%, -439,030)

cp1041:
    total:      2996340
    udp2log:  2,558,004 (-14.629%, -438,336)
    hdfs:     2,558,004 (-14.629%, -438,336)

cp1042:
    total:      2992324
    udp2log:  2,554,088 (-14.645%, -438,236)
    hdfs:     2,554,088 (-14.645%, -438,236)

cp1043:
    total:      3004241
    udp2log:  2,565,111 (-14.617%, -439,130)
    hdfs:     2,565,111 (-14.617%, -439,130)