Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 5 Feb 96 22:36 WET
From:      uhclem@nemesis.lonestar.org (Frank Durda IV)
To:        hackers@freebsd.org, current@freebsd.org
Subject:   Re: sup%d is broken - I agree
Message-ID:  <m0tjf8T-000COKC@nemesis.lonestar.org>

next in thread | raw e-mail | index | archive | help
[0]Jordan wrote:
[0]PLEASE DON'T USE SUP.FREEBSD.ORG TO UPDATE YOURSELF!

[1]John Polstra  replied:
[1]When I try an experiment, and that experiment attempts to delete
[1]220 MB of my files, I lose my thirst for adventure very quickly.
[1]Nevertheless, if you or somebody else can give me some clue about
[1]how to switch to a different sup server without losing all my
[1]perfectly good files, I'll try again.

[3]Richard Wackerbarth <rkw@dataplex.net> says:
[3]As I have pointed out, the cvs tree that I am getting from sup2 also seems
[3]to have problems. As a result, my distribution in the name of -stable,
[3]reflects the same condition.
[3]
[3]We MUST get these problems fixed or we all will be forced to "look to the
[3]source" for correct data.

I must also report similar problems:
Sup from sup2 either hangs for hours on end for no apparent reason
(no Busy message) and when it does wake up, it starts deleting entire
-current trees like mad.

Sup from sup3 delivers a corrupted tree where make world fails within
ten to forty minutes (sometimes it doesn't get finished cleaning
stuff before something goes wrong!) and the kernel will not compile
because files are missing, zero-length or truncated. 

Sup from Sup3 also repeatedly reports that the KerberosIV collection was
"invalid" and aborts the sup.

I was unwilling to see what sup4 would do to my system.

Following these unpleasant experiences, I accessed the "forbidden"
sup.freebsd.org and now have a -current kernel that actually links 
and a make world that has run far longer than in recent memory.  (If
it makes it beyond building libc it will be a record.)

An associate at work has run into the same sort of problems and he
was using sup3.  Part of his problems were caused by running an outdated
config, but with that fixed the kernel would still refuse to compile or
link for strange reasons.  Make world also failed for him, but in different
places than where it dies for me.

As the folks in Tuna Texas would say: "It's broke."  "Yep, it's broke
allright." "No question.  More broke than the Piggly Wiggly after that
there tornado."


I know the updates for recently changed files are present on sup3
as I verified some based on recent -cvs mail, although I don't know
if their contents are valid.  I also know that files that should be static
with relatively old dates seem to be botched so that compiles fail. 

I have no idea what is wrong with sup2 unless the entire tree is missing
or something isn't looking in the right place.  I wasn't willing to
see if sup2 would delete everything locally before I stopped the carnage.

Frank Durda IV <uhclem@nemesis.lonestar.org>|"The Knights who say "LETNi"
or uhclem%nemesis@rwsystr.nkn.net           | demand...  A SEGMENT REGISTER!!!"
	  ^------(this is the fastest route)|"A what?"
or ...letni!rwsys!nemesis!uhclem	    |"LETNi! LETNi! LETNi!"  - 1983




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?m0tjf8T-000COKC>