Re: potential mayhem with trial libc6 package and kernel-headers

[john]

Manoj Srivastava writes:
> Error configuring $package: /usr/src/linux-$version does not seem to
>   be a symbolic link, possibly from an old package or a left-over
>   from a user installed kernel. Please remove
>   /usr/src/linux-$version and try to reconfigure $package.

I suggest you change the message to:

Please remove or rename /usr/src/linux-$version 
and try to reconfigure $package.

Telling the user that she must remove it may be confusing.
-- 
John Hasler
[EMAIL PROTECTED] (John Hasler)
Dancing Horse Hill
Elmwood, WI


--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
[EMAIL PROTECTED] . 
Trouble?  e-mail to [EMAIL PROTECTED] .

Re: potential mayhem with trial libc6 package and kernel-headers

[Bart Schuller]

On Dec 20, Manoj Srivastava wrote
> > "Bart" == Bart Schuller <[EMAIL PROTECTED]> writes:
> Bart> Particularly if they happened to have a fully unpacked and
> Bart> configured kernel source tree in /usr/src/linux .
> 
> Bart> * Poof *
> 
>   No it does not! Would you please look at what the postinst
>  does before spreading FUD?

I'm sorry, I can't explain it, but that is exactly what happened to my
system. But you're perfectly right, looking at the postinst, it's
paranoid enough. So I'll just pretend it never happened :-)

-- 
Bart Schuller  [EMAIL PROTECTED] At Lunalabs, where the
Lunatech Research  http://www.lunatech.com/  future is made today..
Partner of The Perl Institute  http://www.perl.org/Linux http://www.li.org/


--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
[EMAIL PROTECTED] . 
Trouble?  e-mail to [EMAIL PROTECTED] .

Re: potential mayhem with trial libc6 package and kernel-headers

[Manoj Srivastava]

Hi,
>>"Karl" == Karl M Hegbloom <[EMAIL PROTECTED]> writes:

> "Bart" == Bart Schuller <[EMAIL PROTECTED]> writes:
Bart> Particularly if they happened to have a fully unpacked and
Bart> configured kernel source tree in /usr/src/linux .

Bart> * Poof *

No it does not! Would you please look at what the postinst
 does before spreading FUD?

Karl> Perhaps the `kernel-headers' install script ought to look for an
Karl> existing /usr/src/linux directory, do an lstat, and if it's
Karl> really a directory, not a symlink, rename it to linux-version.
Karl> It can get the version from the version file in the kernel
Karl> source can't it?

Instead, it aborts. Maybe I should just move the directory out
 of the way, but a directory existing where noen should exist means
 that my invariants are shot. When ones invariants are shot, the best
 thing to do is change as little as possible, and let the humans take
 care of the cruft. (Why don't people look at the behaviour before
 asking for a change in the operation?)

__
if (-e 'linux' && ! -l 'linux') {
  my $answer;
  print STDERR <<"EOF";
 Error configuring $package: /usr/src/linux does not seem to be a 
   symbolic link, possibly from an old package or a left-over from a 
   user installed kernel. Please remove /usr/src/linux and try to
   reconfigure $package.  

 Please Hit return to continue.
EOF
  $answer = ;
  exit 1;
}


if (-e "linux-$version" && ! -l "linux-$version" ) {
  my $answer;
  print STDERR <<"EOF";
 Error configuring $package: /usr/src/linux-$version does not seem to
   be a symbolic link, possibly from an old package or a left-over
   from a user installed kernel. Please remove
   /usr/src/linux-$version and try to reconfigure $package.

 Please Hit return to continue.
EOF
  $answer = ;
  exit 1;
}
__

Karl> I've installed kernel headers because I had to so libc would
Karl> install. I'd rather use the headers in the actual kernel source
Karl> though.

Please read my often posted FAQ as to why this is a really bad
 idea. 

manoj
 tired of repeating the same thing over and over again.
-- 
 "Make no little plans. They have no Magic to stir Men's blood."
 Hudson
Manoj Srivastava  <[EMAIL PROTECTED]> 
Key C7261095 fingerprint = CB D9 F4 12 68 07 E4 05  CC 2D 27 12 1D F5 E8 6E


--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
[EMAIL PROTECTED] . 
Trouble?  e-mail to [EMAIL PROTECTED] .

Re: potential mayhem with trial libc6 package and kernel-headers

[Karl M. Hegbloom]

> "Bart" == Bart Schuller <[EMAIL PROTECTED]> writes:

Bart> Particularly if they happened to have a fully unpacked and
Bart> configured kernel source tree in /usr/src/linux .

Bart> * Poof *

 Perhaps the `kernel-headers' install script ought to look for an
 existing /usr/src/linux directory, do an lstat, and if it's really a
 directory, not a symlink, rename it to linux-version.  It can get the
 version from the version file in the kernel source can't it?

 I've installed kernel headers because I had to so libc would install.
 I'd rather use the headers in the actual kernel source though.


--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
[EMAIL PROTECTED] . 
Trouble?  e-mail to [EMAIL PROTECTED] .

Re: potential mayhem with trial libc6 package and kernel-headers

[Bart Schuller]

On Dec 19, Rob Browning wrote
> People who weren't using kernel-headers before (because they never
> needed it), may be in for a shock.

Particularly if they happened to have a fully unpacked and configured
kernel source tree in /usr/src/linux .

* Poof *

It's only a slight inconvenience, but I'd rather have known beforehand,
so that I might have saved my .config

-- 
Bart Schuller  [EMAIL PROTECTED] At Lunalabs, where the
Lunatech Research  http://www.lunatech.com/  future is made today..
Partner of The Perl Institute  http://www.perl.org/Linux http://www.li.org/


--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
[EMAIL PROTECTED] . 
Trouble?  e-mail to [EMAIL PROTECTED] .

Re: potential mayhem with trial libc6 package and kernel-headers

[Rob Browning]

Manoj Srivastava <[EMAIL PROTECTED]> writes:

>   I genrally unpack into /usr/src/local and mv things one level
>  up, personally. 

That seems reasonable.  I guess I had just sort of always treated
/usr/src/ as if it was local, even though I probably shouldn't have.

> Rob> Now I'm happy to just change my behavior, and unpack the kernels
> Rob> I download somewhere else, but I think you're going to see some
> Rob> mayhem when this pair of packages is released and other people
> Rob> doing something similar suddenly have to treat /usr/src/linux as
> Rob> read only without being warned.
> 
>   What do you sugggest, modulo maintaining backward
>  compatibility to people who have old kernel-source packages
>  installed? 

I don't really have a good suggestion, but I think that this should
be somehow *widely* advertised with the new libc6 arrangement.  I
don't know if I'd go so far as an actual pause in the postinst, but
this could be a fairly serious problem. 

People who weren't using kernel-headers before (because they never
needed it), may be in for a shock.

-- 
Rob Browning <[EMAIL PROTECTED]>
PGP fingerprint = E8 0E 0D 04 F5 21 A0 94  53 2B 97 F5 D6 4E 39 30


--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
[EMAIL PROTECTED] . 
Trouble?  e-mail to [EMAIL PROTECTED] .

Re: potential mayhem with trial libc6 package and kernel-headers

[Manoj Srivastava]

Hi,
>>"Rob" == Rob Browning <[EMAIL PROTECTED]> writes:

Rob> I recently installed the new libc6 experiental pacakges which
Rob> also wants you to install kernel-headers.  The problem is that
Rob> kernel-headers thinks it "owns" /usr/src/linux.  For users using
Rob> the kernel-package (or whatvever) to build their own kernels,
Rob> this may be a problem.  It was at least surprising.

All the products of kernel-package (and that includes
 kernel-source-xxx and kernel-headers-xxx) have always behaved as if
 they own /usr/src/linux ;-(. So this ain't new behaviour; things have
 always behaved that way. Just as /usr/local is not under vendor
 control, /usr/src/should be assumed to be, right?

I genrally unpack into /usr/src/local and mv things one level
 up, personally. 

Rob> In my case, I didn't even notice that kernel-headers added the
Rob> /usr/src/linux link, and since I normally don't ever have a
Rob> /usr/src/linux on my machine, I can just assume that it's safe to
Rob> untar a new kernel in /usr/src and then "mv linux
Rob> linux-" [1].  However, when the kernel-headers package
Rob> is installed, this means I'm unpacking my new kernel source into
Rob> the kernel-header package's directory -- not good.

Hmm, right.

Rob> Now I'm happy to just change my behavior, and unpack the kernels
Rob> I download somewhere else, but I think you're going to see some
Rob> mayhem when this pair of packages is released and other people
Rob> doing something similar suddenly have to treat /usr/src/linux as
Rob> read only without being warned.

What do you sugggest, modulo maintaining backward
 compatibility to people who have old kernel-source packages
 installed? 

manoj
-- 
 "No man steps in the same river twice, for it's not the same river,
 and he's not the same man." Heraclitus
Manoj Srivastava  <[EMAIL PROTECTED]> 
Key C7261095 fingerprint = CB D9 F4 12 68 07 E4 05  CC 2D 27 12 1D F5 E8 6E


--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
[EMAIL PROTECTED] . 
Trouble?  e-mail to [EMAIL PROTECTED] .