Mailinglist Archive: opensuse (4634 mails)

< Previous Next >
Re: [opensuse] openSUSE-10.2-GM-DVD-x86_64.iso md5sum's do not match
  • From: "Mark Goldstein" <goldstein.mark@xxxxxxxxx>
  • Date: Mon, 18 Dec 2006 17:30:19 +0200
  • Message-id: <1d8633230612180730u79035bc0te10a8f709fb72714@xxxxxxxxxxxxxx>
On 12/18/06, Carlos E. R. <robin.listas@xxxxxxxxxxxxxx> wrote:
Hash: SHA1

The Monday 2006-12-18 at 16:11 +0200, Mark Goldstein wrote:

> In my case it was slightly different. ISO image had the correct md5sum
> b4d727d2fb62c701233ea537f220e9ee.
> But when I checked burned DVD, md5sum returned b750fb7280788860a002f58bf902c7.
> (I burned it using Nero under WinXP and verification passed).
> I then used installed x86 version of OpenSuSE 10.2 to check the media
> (YaST2->Software->Media Check) and was surprised to see Result OK.


Known bug/feature - to me, at least.

I have a correct image md5sum, but the burned image is different. If I
burn it five times, all five copies have the same md5, but different from
the downloaded image. I then use Yast, and it is correct - so I conclude
it is correct.

I have no idea why. Perhaps it is image padding in the media. Perhaps a
bug in the md5sum implementation. Dunno.

Which md5sum program did you use to check it, the one in Linux, or another
one in windows?

For this DVD stuff I checked image using md5sum in Windows, but
checked DVD itself in Linux.

Before that I checked CD images using Linux md5sum and got correct results.
I'll re-check using same program for all tests.


> How could it be? I remember somebody mentioned that YaST media check
> uses different test (SHA1 signature for each packet perhaps?).

No, it is an md5 but zeroing the section where the expected md5sum is

Thanks for explanation.

Mark Goldstein
To unsubscribe, e-mail: opensuse+unsubscribe@xxxxxxxxxxxx
For additional commands, e-mail: opensuse+help@xxxxxxxxxxxx

< Previous Next >
Follow Ups