md5sum problems
in [Slackware]
|
Prev: dhcpd angst...
Next: Firefox 3.6.6 crashing on print
From: root on 27 Jun 2010 10:21 Recently I downloaded the dvd versions of Slack 13.0 and 13.1. After downloading 13.1 I ran an md5sum on the downloaded file and compared that to the published value. I then burned the dvd with a burner on /dev/hdb. I followed up by doing: md5sum /dev/hdb and got the same value as the original file. Then I tried the same thing with the 13.0 image. As before, the downloaded file matched the published md5sum, but this time, after burning and using the md5sum on /dev/hdb the checksum did not match. Do I tried burning a different brand of DVD. Same result. Then I tried both brands of dvd blanks in a second burner on /dev/hda. All the md5sums agreed with each other, but all were different from the file and different from the published value. Any ideas of what might explain the differences? TIA
From: Chris Sorenson on 27 Jun 2010 11:49 root wrote: > Recently I downloaded the dvd versions of Slack > 13.0 and 13.1. After downloading 13.1 I ran > an md5sum on the downloaded file and compared > that to the published value. I then burned the > dvd with a burner on /dev/hdb. I followed up > by doing: > > md5sum /dev/hdb > > and got the same value as the original file. > > Then I tried the same thing with the 13.0 image. > As before, the downloaded file matched the > published md5sum, but this time, after burning > and using the md5sum on /dev/hdb the checksum > did not match. > > Do I tried burning a different brand of DVD. > Same result. Then I tried both brands of dvd > blanks in a second burner on /dev/hda. All > the md5sums agreed with each other, but all > were different from the file and different > from the published value. > > Any ideas of what might explain the differences? > DVD burning is _not_ an exact science... ;)
From: Grant on 27 Jun 2010 18:44 On Sun, 27 Jun 2010 10:49:40 -0500, Chris Sorenson <csoren(a)isd.net> wrote: >root wrote: >> Recently I downloaded the dvd versions of Slack >> 13.0 and 13.1. After downloading 13.1 I ran >> an md5sum on the downloaded file and compared >> that to the published value. I then burned the >> dvd with a burner on /dev/hdb. I followed up >> by doing: >> >> md5sum /dev/hdb >> >> and got the same value as the original file. >> >> Then I tried the same thing with the 13.0 image. >> As before, the downloaded file matched the >> published md5sum, but this time, after burning >> and using the md5sum on /dev/hdb the checksum >> did not match. >> >> Do I tried burning a different brand of DVD. >> Same result. Then I tried both brands of dvd >> blanks in a second burner on /dev/hda. All >> the md5sums agreed with each other, but all >> were different from the file and different >> from the published value. >> >> Any ideas of what might explain the differences? >> > >DVD burning is _not_ an exact science... ;) There's information outside the file image too? Number of padding blocks might change the returned md5sum? I've never expected burned DVD to match md5sum, instead I trust the burner app to correctly verify the image after writing the CD/DVD. Grant. -- http://bugs.id.au/
From: Baho Utot on 27 Jun 2010 18:48 root wrote: > Recently I downloaded the dvd versions of Slack > 13.0 and 13.1. After downloading 13.1 I ran > an md5sum on the downloaded file and compared > that to the published value. I then burned the > dvd with a burner on /dev/hdb. I followed up > by doing: > > md5sum /dev/hdb > > and got the same value as the original file. > > Then I tried the same thing with the 13.0 image. > As before, the downloaded file matched the > published md5sum, but this time, after burning > and using the md5sum on /dev/hdb the checksum > did not match. > > Do I tried burning a different brand of DVD. > Same result. Then I tried both brands of dvd > blanks in a second burner on /dev/hda. All > the md5sums agreed with each other, but all > were different from the file and different > from the published value. > > Any ideas of what might explain the differences? > > TIA check-burn.sh #!/bin/bash if [ $1 ]; then isoFile=$1 else echo "Usage: $0 <iso-image> <cd-drive>" echo "E.g. $0 /tmp/slackware-12.0.iso /dev/dvd" exit 1 fi if [ $2 ]; then cdDrive=$2 else echo "Usage: $0 <iso-image> <cd-drive>" echo "E.g. $0 /tmp/slackware-12.0.iso /dev/dvd" exit 1 fi if [ ! -b $cdDrive ]; then echo "ERROR. '$cdDrive' is not a block device." exit 1 fi if [ ! -r $isoFile ]; then echo "ERROR. ISO image '$isoFile' does not exist." exit 1 else echo "** Verifying md5sums between $isoFile <-> $cdDrive" dd if=$cdDrive | head -c $(stat --format=%s $isoFile) | md5sum \ && md5sum $isoFile fi
From: root on 28 Jun 2010 04:52
Baho Utot <baho-utot(a)invalid.com> wrote: > > check-burn.sh > > #!/bin/bash > if [ $1 ]; then > isoFile=$1 > else > echo "Usage: $0 <iso-image> <cd-drive>" > echo "E.g. $0 /tmp/slackware-12.0.iso /dev/dvd" > exit 1 > fi > > if [ $2 ]; then > cdDrive=$2 > else > echo "Usage: $0 <iso-image> <cd-drive>" > echo "E.g. $0 /tmp/slackware-12.0.iso /dev/dvd" > exit 1 > fi > > if [ ! -b $cdDrive ]; then > echo "ERROR. '$cdDrive' is not a block device." > exit 1 > fi > > if [ ! -r $isoFile ]; then > echo "ERROR. ISO image '$isoFile' does not exist." > exit 1 > else > echo "** Verifying md5sums between $isoFile <-> $cdDrive" > dd if=$cdDrive | head -c $(stat --format=%s $isoFile) | md5sum \ > && md5sum $isoFile > fi Thanks for the script. Unfortunately the md5sum generated by the script differed from both the iso and the one formed by simply md5sum /dev/hdb. I guess the real question is why the checksum worked on the 13.1 disk I burned first. |