Thanks for the update. I had discovered the same thing. Chris posted an older version and also had a correction. Your change is rather nice in that it only goes through the twpol.txt file once. Neat. ---------- Original Message ----------- From: "Turner, (Resolute Address) Jeff" <jeff.turner@resolute-ltd.com.au> Sent: Fri, 24 Jan 2003 13:23:09 +0800 Subject: RE: Tripwire > I have a redhat server which runs tripwire but was giving me the long list of missing files in every report. After reading this thread I thought it would be a good idea to implement the script written by Chris, BUT I found that as posted it only updated the policy with the LAST line of missing files. > > The section > > for line in `cat tmp.fix2` > do > sed "s/.*$line/# &/" "$2" > twpol.txt.fixed > done > > replaces twpol.txt.fixed each time through the loop. > > I replaced this with > > for line in `cat tmp.fix2` > do > echo "s/.*$line/# &/" >> tmp.fix3 > # create a list of substitutions for sed to perform > done > > sed -f tmp.fix3 "$2" > twpol.txt.fixed > # Run sed commenting out ALL missing files from the list > rm -rf tmp.fix3 > > I know this could have been done when te tmp.fix2 file was created but this was easier for me to write and as it is a script that will only be run once per machine it didnt seem worth worrying too much about efficiency. > > I would have used the Perl script also submitted but as it was sent as an attachment and I get the list in digest format I couldn't (easily) get it. > > I ran the shell script and now my tripwire reports 0 errors ! > > > > > Subject: Re: Tripwire > > From: Chris Cuevas <fclcac@nersp.nerdc.ufl.edu> > > To: "RedHat 8.0 list" <psyche-list@redhat.com> > > Date: 16 Jan 2003 10:38:11 -0500 > > Reply-To: psyche-list@redhat.com > > Sent this yesterday not sure why it didn't get posted. Here it is > > again. > > Mike, > > I agree there should be a simpler method. After doing the same > > for about the first ten systems I set up I decided to write a > > script to > > do all the leg work for me. Here is my twhelp.sh script that should > > make life much simpler for you. Hope this helps out. Any comments or > > ideas for improvement are welcome. > > #! /bin/bash > > # > > # twhelp.sh > > # > > ############################################################# > > # # > > # Christopher Cuevas # > > # fclcac@nersp.nerdc.ufl.edu # > > # Nov. 13th 2002 # > > # Florida Center for Library Automation # > > # http://www.fcla.edu # > > # # > > # twhelp will comment out lines from a twpol.txt file when # > > # supplied with a twreport_file and the path to twpol.txt # > > # and create a twpol.txt.fixed file # > > # # > > # usage: twhelp twreport_file path_to_twpol.txt # > > # # > > ############################################################# > > E_NOARGS=65 > > E_ARGERROR=66 > > if [ $# -eq 0 ] > > then > > echo "Usage: `basename $0` twreport_file path/to/twpol.txt" >&2 > > # Error message to stderr > > exit $E_ARGERROR > > fi > > # Test for correct file type > > type=`eval file $1 | awk '{ print $2 }'` > > # "file $1" echos file type... > > # then awk removes all but the second field > > # the result is fed into the variable "type" and compared to > > "correct_type" > > correct_type="ASCII" > > if [ "$type" != "$correct_type" ] > > then > > echo > > echo "This script only works on non executable ascii files." > > echo > > fi > > cat "$1" | grep Filename: | awk -F: '{ print $2 }' > tmp.fix1 > > # awk through the twreport file and create a tmp.fix1 file > > # with all paths to files that are not on the system > > sed 's/\//\\\//g' tmp.fix1 > tmp.fix2 > > # add a \ in front of the path so sed will comment it out correctly > > # output this to tmp.fix2 > > for line in `cat tmp.fix2` > > do > > sed "s/.*$line/# &/" "$2" > twpol.txt.fixed > > done > > # comment out lines from twpol.txt and create twpol.txt.fixed > > rm -rf tmp.fix1 > > rm -rf tmp.fix2 > > # clean up the tmp.fix files > > exit 0 ------- End of Original Message ------- -- Psyche-list mailing list Psyche-list@redhat.com https://listman.redhat.com/mailman/listinfo/psyche-list
