It looks like what happened was IE didn't grab the last few bytes of
data... When I did a continue on the iso from Redhat's ftp site, the
checksum matches... Looks like I'll be able to do this after all...
BTW, IE sucks my sack.
--Tim
Jacob Meuser wrote:
>
> On Sat, Jul 21, 2001 at 03:48:44PM -0400, Tim Howe wrote:
> > I ran a checksum on the iso I have and it
> > was completely different!
> ^^^^^^^^^^
> funk:~/tmp% echo 1 > test1
> funk:~/tmp% echo 2 > test2
> funk:~/tmp% md5 test1
> MD5 (test1) = b026324c6904b2a9cb4b88d6d61c81d1
> funk:~/tmp% md5 test2
> MD5 (test2) = 26ab0db90d72e28ad0ba1e22ee510510
>
> Only one overlapping character (the 2 near the middle) out of 32.
>
> > I think the download was corrupted or
> > something. DO NOT TRUST IE TO DOWNLOAD!
>
> Pull that dunce cap down tight, Tim!
>
> This is a subroutine (modified to be standalone) from a package fetching
> script I wrote. Since it is theoretically possible (has anyone done it yet?)
> to modify a file and "pad" it in such a way that it produces the same md5
> checksum as another file (which was thought to be impossible), it's a good
> idea to check the file's size also.
>
> Save it as something like 'md5check', and use it like so -
>
> % md5check <file> <size> <md5 checksum>
>
> where <size> and <checksum> are what they should be.
>
> ----------------------------------------------------------
> #!/usr/bin/perl -w
>
> use strict;
>
> my ($file, $check_size, $check_md5) = @ARGV;
> if (! (-f "$file")) {
> die "Could not check_size_n_md5 $file because $file does not exist!\n";
> }
> my $size = (stat ($file))[7];
> my $md5 = `md5 $file`;
> chomp $md5;
> $md5 =~ s/MD5 \($file\) = //;
> if ($check_size ne $size) {
> print "$file is the wrong size!\n";
> exit (1);
> } elsif ($check_md5 ne $md5) {
> print "$file has the wrong md5 checksum!\n";
> exit (2);
> } else {
> print "$file checks out OK :)\n";
> exit (0);
> }
> ______________________________________________________________
>
> Written for OpenBSD, YMMV.
>
> <[EMAIL PROTECTED]>
>
