Allen Briggs wrote:

>> There are some serious problems with the sysinst code I submitted with 
>> that PR.  You might want the new sysinst code too.
>
>I definitely do.  I just corrupted my disk nicely on my first sysinst
>attempt... :-)
>
Sorry about that.  No doubt the result of a serious mismatch between the 
in-core and on-disk labels, plus not getting the size of the disk 
correct. (Been there myself.)  My new version of sysinst has a 
verification step to compare the labels.  If they don't match the 
installation is stopped. There's also some diagnostic code in there to 
provide details of where the labels failed to match to aid in finding any 
remaining problems with the code.

The sysinst documentation changes to the doc source files are filed with 
a PR but they were done against the 1.4.2 sources.  I've kept copies of 
the resulting documentation that people can download and review.  If 
there aren't any major complaints I'd be happy to try updaing them to 
work against the -current doc sources.

The latest version of sysinst sources are in the same place as the 
documentaiton files:

   ftp://murphy.dyndns.org/pub/sysinst

>> Or if Allen or Nigel update the Booter it may not be necessary.  Sure 
>> would like to get rid of that Mkfs program! ;-)
>
>Most definitely.  Hacks live much longer than you would ever imagine...

So true!  People are still relaying some of my misinfomation about Mkfs 
that was the result of a programming error I made years ago!  I finally 
decided the only way to eliminate it was to remove the need for Mkfs, 
thus sysinst.  ;-)

-bob