Re: [Cooker] Re: When??

2003-03-21 Thread Greg Meyer
On Friday 21 March 2003 07:56 am, Brian J. Murrell wrote:
 On Fri, Mar 21, 2003 at 02:42:33PM +1200, Jason Greenwood wrote:
  Will 9.1 Final ISO's hit the mirrors??

 I would like to know this too.

 Just what are we waiting for?  I don't recall any of this kind of wait
 with past releases.  ISO's came on the heels of the mandrake_release
 RPM finally being updated.

  It would be good to get an approximation from Mandrake on this.

 Completely agree!

 b.

Maybe there is finally going to be some sanity and the delay of the release 
until the boxes are ready.

-- 
Greg



Re: [Cooker] Re: When??

2003-03-21 Thread Duncan
On Fri 21 Mar 2003 05:56, Brian J. Murrell posted as excerpted below:
 On Fri, Mar 21, 2003 at 02:42:33PM +1200, Jason Greenwood wrote:
  Will 9.1 Final ISO's hit the mirrors??

 I would like to know this too.

 Just what are we waiting for?  I don't recall any of this kind of wait
 with past releases.  ISO's came on the heels of the mandrake_release
 RPM finally being updated.

This is just speculation on my part, but I'm guessing the lack of ISOs has to 
do with the kernel problem folks have mentioned here repeatedly.

 It seems a security vuln and kernel patch was released at the last minute, to 
late for the normal process of verification and  inclusion on the release 
ISOs.  Never-the-less, the patch was added, and the corrected Kernel packages 
created and delivered.

The problem appears to be that in the confusion of doing this at the last 
minute, without the normal verification and etc. someone made a mistake, and 
reused a previous kernel release number for the new kernel RPMs.  
Unfortunately, this means the various automated tools don't see the new 
kernel as a change from the previous one, and don't pick up that RPM.  At the 
same time, the official list along with sizes and probably the MD5 checksums 
include and are based on the new kernel version, that the automated tools 
don't pick up, because it looks to them the same as an old one.  Thus, there 
are all sorts of errors.

That brings us to the human side.  My guess is that in the work-up to the 
release, days off and other such things were skipped, and as soon as it was 
out the door, pretty much everyone left for the weekend, to catch up on the 
days off and such.  It's also possible that a number of folks were leaving 
Mdk to be effective at 9.1 release, due to the continual streamlining due to 
the financial problems.  These folks will now have cleared out their desk and 
be gone permanently.

Anyway, the problem is likely that the automated tools can't fix the problem 
as outlined above, and anybody that can do anything about it took off until 
next week.  Likewise, anyone in a position to make an official comment on it 
is off for the weekend, as it was supposed to be all out and all good, by 
now.

Maybe someone will check back in today (Saturday) to see how things are going, 
realize the problem, and come fix it.  More likely, it won't be fixed now 
until Monday, as someone else mentioned.  In either case, it shouldn't be a 
huge problem, it's just the matter of poor timing of that last minute vuln 
and patch, combined with the normally minor kernel RPM versioning mistake, 
has things a bit confused, while everyone's out for a well deserved rest 
after working real hard to get the release done, and no one's there now that 
can do much about it.

That's my admittedly gallery viewpoint opinion on what's likely going on, 
anyway, FWIW..

-- 
Duncan
They that can give up essential liberty to obtain a little
temporary safety, deserve neither liberty nor safety. --
Benjamin Franklin