#!/bin/bash

#
# entry example
#
# (1579051) 10/16/2009 0:02:13 AM - (not logged in) (192.168.1.11)> USER chris
#
# sort -k1n fzs-2009-10-16.log
# does not work properly because when the session id and timestamp are the same
# it goes on and tries to sort based on the message
# if the timestamp granularity was high enough it would work (and so much faster!)
#

[ $# -ne 1 ] && echo "number of args required is 1" && exit 1

[ ! -e $1 ] && echo "file $1 does not exist" && exit 2

FILEPATH=`dirname $1`
LOGNAME=`basename $1`

LOGNAME_SORTED=${FILEPATH}/${LOGNAME}".sorted"
SID_SORTED_FILENAME=${FILEPATH}/"fzlog_sid_sorted.$$"

rm -f $LOGNAME_SORTED

awk '{ split( $1, L, "\\(|\\)" ); print L[ 2 ]; }' $1 | sort | uniq > ${SID_SORTED_FILENAME}

cat ${SID_SORTED_FILENAME} | while read line
do
        grep "^($line)" $1 >> ${LOGNAME_SORTED}
done

rm -f ${SID_SORTED_FILENAME}

exit 0

bash

1

sort filezilla server log entries by session id

#!/bin/bash

#
# entry example
#
# (1579051) 10/16/2009 0:02:13 AM - (not logged in) (192.168.1.11)> USER chris
#
# sort -k1n fzs-2009-10-16.log
# does not work properly because when the session id and timestamp are the same
# it goes on and tries to sort based on the message
# if the timestamp granularity was high enough it would work (and so much faster!)
#

[ $# -ne 1 ] && echo "number of args required is 1" && exit 1

[ ! -e $1 ] && echo "file $1 does not exist" && exit 2

FILEPATH=`dirname $1`
LOGNAME=`basename $1`

LOGNAME_SORTED=${FILEPATH}/${LOGNAME}".sorted"
SID_SORTED_FILENAME=${FILEPATH}/"fzlog_sid_sorted.$$"

rm -f $LOGNAME_SORTED

awk '{ split( $1, L, "\\(|\\)" ); print L[ 2 ]; }' $1 | sort | uniq > ${SID_SORTED_FILENAME}

cat ${SID_SORTED_FILENAME} | while read line
do
        grep "^($line)" $1 >> ${LOGNAME_SORTED}
done

rm -f ${SID_SORTED_FILENAME}

exit 0

copy | embed

• awk, 
• bash, 
• filezilla, 
• filezilla server, 
• log, 
• sort