From: Eef Hartman on
Grant <omg(a)grrr.id.au> wrote:
> On Thu, 15 Jul 2010 19:29:29 +0000 (UTC), Sylvain Robitaille <syl(a)alcor.concordia.ca> wrote:
> I'm trying to remember too, which part of the installer requires that
> old, particular version of tar? Or was that issue fixed when the new
> compression method was added?

No, from installpkg itself:
TAR=tar-1.13
(skipped)
( cd $ROOT/ ; $packagecompression -dc | $TAR -xlUpvf - ) < $package >> $TMP/$shortname 2> /dev/null

so actually it is doing the de-compression first and then piping it
into tar 1.13
--
******************************************************************
** Eef Hartman, Delft University of Technology, dept. SSC/ICT **
** e-mail: E.J.M.Hartman(a)tudelft.nl - phone: +31-15-27 82525 **
******************************************************************
From: Grant on
On Fri, 16 Jul 2010 22:22:04 +0200, Eef Hartman <E.J.M.Hartman(a)tudelft.nl> wrote:

>Grant <omg(a)grrr.id.au> wrote:
>> On Thu, 15 Jul 2010 19:29:29 +0000 (UTC), Sylvain Robitaille <syl(a)alcor.concordia.ca> wrote:
>> I'm trying to remember too, which part of the installer requires that
>> old, particular version of tar? Or was that issue fixed when the new
>> compression method was added?
>
>No, from installpkg itself:
>TAR=tar-1.13
>(skipped)
>( cd $ROOT/ ; $packagecompression -dc | $TAR -xlUpvf - ) < $package >> $TMP/$shortname 2> /dev/null
>
>so actually it is doing the de-compression first and then piping it
>into tar 1.13

Thanks, similar to what Sylvain wrote, more detail :) I do know tar
can do some amazing things while transporting files and perms.

Grant.
From: Eef Hartman on
Grant <omg(a)grrr.id.au> wrote:
> That's what I mean by the observation that one can expect a known root
> environment only during the install. So adding accessories should do
> the additional checks or changes you suggested early on?

Pat's updates do expect the normal "root" environment too (the ones
in slackware-<version>/patches/packages).
So essentially, do not mess up Pat's normal root umask/path because
otherwise those will result into the same kind of problems:
drwxr-xr-x root/root 0 2010-06-30 06:46:47 ./
drwxr-xr-x root/root 0 2010-06-30 06:46:47 install/
-rw-r--r-- root/root 368 2010-06-30 06:46:47 install/doinst.sh
-rw-r--r-- root/root 852 2010-06-30 06:46:47 install/slack-desc
drwxr-xr-x root/root 0 2010-06-30 06:46:47 usr/
drwxr-xr-x root/root 0 2010-06-30 06:46:47 usr/lib/
(the beginning of 13.1's libtiff-3.9.4 update).

As you see with a 077 umask that will f'up the whole root filesystem.
--
******************************************************************
** Eef Hartman, Delft University of Technology, dept. SSC/ICT **
** e-mail: E.J.M.Hartman(a)tudelft.nl - phone: +31-15-27 82525 **
******************************************************************
From: Lew Pitcher on
On July 19, 2010 12:52, in alt.os.linux.slackware, E.J.M.Hartman(a)tudelft.nl
wrote:

[snip]
> Pat's updates do expect the normal "root" environment too (the ones
> in slackware-<version>/patches/packages).
> So essentially, do not mess up Pat's normal root umask/path because
> otherwise those will result into the same kind of problems:
[snip]

One of the hallmarks of Slackware Linux has been the unspoken attitude that
the system administrator is the "final word" on how the system should be
administered. Certainly, that attitude carries over into how Slackware
Package Management works, in that the sysadm is /expected/ and
almost /required/ to understand the implications of each package if only to
resolve package dependencies.

But, this attitude has other implications, primarily being that the sysadm
is always /right/, at least as far as how s/he wants to administer his/her
system is concerned.

I find it unreasonable to also assume that the sysadm will /not/ change the
system as far as root's environment and setup is concerned. If the sysadm
is assumed intelligent enough to perform dependency checking, then the
sysadm is also intelligent enough to set up the system the way it works for
him and his users.

The (unspoken) assumption in the SlackBuild environment seems to be that no
sysadm will vary root's configuration from how Pat delivered it. This seems
dangerous to me; we know that it's not true, and we know that Slackware (as
a distro) /expects/ sysadm's to make changes, so /why/ should SlackBuild
now expect that those self-same sysadms /have not/ made changes?

To me, there are (at least) two alternatives here:
1) Change the SlackBuild process so that it enforces a known-good
environment for implementation *within it's own environment*. Make it use
the "sane" environment that it currently assumes, and make sure that this
sane environment does not conflict with or alter the environment that the
sysadm has configured as his site standard.

-Or-

2) Document the environment requirements for SlackBuild, and ensure that any
deviation is reported (to the sysadm). Let the sysadm decide whether to
change root's environment back to the PV-default (as needed for sane
installation via SlackBuild) or to override SlackBuild's "concerns" and
build with the site-standard environment.

My /preference/ would be to have SlackBuild report my site's deviation from
the "norm", and ask if I wish to continue with the build using my (possibly
risky) site-standard environment, or override and build with a one-time
environment that matches the (sane) PV-delivered default environment.

Just my 2 cents worth for this topic
--
Lew Pitcher
Master Codewright & JOAT-in-training | Registered Linux User #112576
Me: http://pitcher.digitalfreehold.ca/ | Just Linux: http://justlinux.ca/
---------- Slackware - Because I know what I'm doing. ------


From: realto margarino on
On Jul 19, 9:28 pm, Lew Pitcher <lpitc...(a)teksavvy.com> wrote:

Stand clear of Brampton's Lew Pitcher, ex-employee for TDBank (aka
TDCanadaTrust).

Lew Pitcher is a domain thief and an internet stalker.

Check out http://lewpitcher.ca for all the facts.

Nice picture, too!

This notice is made in the public's interest.