Re: MD5SUMS

["J.A. Bezemer" <costar@panic.et.tudelft.nl>]

On Fri, 4 Aug 2000, Reuleaux, Andreas wrote:

> Yes, I am on this list and yes, I checked the MD5SUMS with the "-b"

Only to the list then, this time ;-)

> option (md5sum.exe -b potato-i386-2.raw) as I stated before.
> 
> I downloaded the images via http from
> http://ftp.gigabell.net/pub/debian-testimages/ - I have read the
> recommendations to use rsync from the pseudo-image-kit and I do want
> to respect them in the future, i. e. I don't want to slown down other
> people with my http/ftp downloads. - yet I used http because this was
> easier for me basically for two reasons (1) I downloaded from within
> my office (winNT) behind a firawall and we don't even have DNS but
> wins. ftp and http work fine for me and I am not sure yet if I can get
> rsync working for me. - I will try and see. (2) the transfer from the
> above mentioned gigabell mirror seemed fast to me.

Ok. I was about to rsync my (good) copy to that of gigabell, but gigabell
doesn't have rsync. Too bad.

> Therefore I haven't had a chance yet to just re-run rsync. I have
> done another http download with the same wrong MD5SUMS results.
> But since nobody else seems to have these MD5SUMS problems, I would
> conclude the MD5SUMS are OK, and I am sure I will figure out a way 
> to get the correct images.

The mirror in Finland should also be quite fast from .de if you must use
FTP/HTTP.

Another option might be taking the image home and running rsync over a "slow" 
modem connection to a "better" ISP (that allows rsync traffic). Since the
image is likely only having very few wrong bits, this should take less than
half an hour.

Also note that only the Binary-1 CD is "obligatory" for installs/upgrades.

And, if you're brave, just burn the wrong Binary-2 and use
"apt-cdrom add -f=no" which is supposed to scan all packages with individual
md5sums, and excludes the mismatching ones. Should take quite some time,
though.


Regards,
  Anne Bezemer


--  
To UNSUBSCRIBE, email to debian-testing-request@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org