Re: A case study of a new user turned off debian

[Florian Weimer]

Billy Biggs wrote:

> > [...] Lot of new HW has a better chance to be (better) supported on
> > newer system (are new kernels available for stable?)
> 
>   Of particular interest to desktop users is XFree86's video card
> drivers.

Or, increasingly, GE NIC support.

(For X support, current unstable isn't really helpful either.)

Re: A case study of a new user turned off debian

[Cameron Patrick]

On Tue, Nov 04, 2003 at 04:35:13PM +0100, Josip Rodin wrote:
| On Tue, Nov 04, 2003 at 01:53:36AM -0600, Chris Cheney wrote:
| > It would be helpful if Debian could even be installed on machines newer
| > than about 2 years old.
| 
| It would be helpful if people wouldn't make sweeping generalizations all the
| time.

Yes.  Everyone knows that all generalisations are false anyway :-)

Cameron.

Re: A case study of a new user turned off debian

[Darren Salt]

I demand that Greg Stark may or may not have written...

> Darren Salt <[EMAIL PROTECTED]> writes:
>> BTW, no need to Cc: me - or did Gnus not notice the Mail-Followup-To
>> header?

> Uhm. What Mail-Followup-To header? I didn't receive one on this message,
> perhaps it's stripped by the mail server? Or perhaps you're mistaken about
> it being included?

Its absence in that particular message is because I'd started the reply
before noticing that you'd also sent your message to the mailing list.

It's there in this message. I have a macro which inserts it when I start
editing. It works perhaps too well; I normally forget that it's there ;-)

-- 
| Darren Salt   | linux (or ds) at | nr. Ashington,
| woody, sarge, | youmustbejoking  | Northumberland
| RISC OS   | demon co uk  | Toon Army
|   Let's keep the pound sterling

The million to one chance - nine times out of ten.

Re: A case study of a new user turned off debian

[Josip Rodin]

On Thu, Nov 06, 2003 at 01:51:06PM -0500, Branden Robinson wrote:
> > It would be helpful if people wouldn't make sweeping generalizations all
> > the time.
> 
> /me hands Josip an Excellence in Un-self-conscious Irony Award

I was merely responding to a sentence written with the same beginning, but
now that you mentioned it, I might as well claim that I originally intended
it to sound like that ;)

-- 
 2. That which causes joy or happiness.

Re: A case study of a new user turned off debian

[Josip Rodin]

On Wed, Nov 05, 2003 at 11:59:26PM -0500, Greg Stark wrote:
> > > > It would be helpful if people wouldn't make sweeping generalizations
> > > > all the time. 
> 
> "All the time"? ...

Someone has been saying that every once in a while for the last N years.

-- 
 2. That which causes joy or happiness.

Re: A case study of a new user turned off debian

[Branden Robinson]

On Tue, Nov 04, 2003 at 04:35:13PM +0100, Josip Rodin wrote:
> It would be helpful if people wouldn't make sweeping generalizations all the
> time.

/me hands Josip an Excellence in Un-self-conscious Irony Award

-- 
G. Branden Robinson|A committee is a life form with six
Debian GNU/Linux   |or more legs and no brain.
[EMAIL PROTECTED] |-- Robert Heinlein
http://people.debian.org/~branden/ |


signature.asc
Description: Digital signature

Re: A case study of a new user turned off debian

[era+debian]

On 06 Nov 2003 01:06:25 -0500, Greg Stark <[EMAIL PROTECTED]> posted to
debian-devel:
 > Personally I'm of the opinion that stable is useless. It certainly
 > has no use for me. Perhaps if I ran a production server on debian I
 > might think otherwise but I rather doubt it. When I had production
 > servers they all ran 2.4 and needed the latest stable releases of
 > anything important like database, mail, web server services.
 > If I ran production servers on debian today I would probably pick
 > an arbitrary date off snapshot.debian.org and declare that my
 > "stable". If I had security problems I would pick a date recent
 > enough to have the security fixes, test it, and declare it
 > "stable".
 > It wouldn't be guaranteed to be bug-free, but then nothing is.
 > Stable has tons of minor bugs that no upstream maintainer would
 > listen to because they were fixed aeons ago anyways, or more likely
 > are no longer relevant in current sources.

Sounds more like a case of "stable plus backports of the important
pieces". Now if only somebody were telling me where to find "stable"
backports for Woody of the packages I need ... (Probably I'm too much
of a skeptic for not believing that a random hit in the search engine
at apt-get.org is what I should be using.)

/* era */

-- 
formail -s procmail http://www.euro.cauce.org/
cat | more | cathttp://www.debian.org/

Re: A case study of a new user turned off debian

[Greg Stark]

Graham Wilson <[EMAIL PROTECTED]> writes:

> > Regardless. Having people install fresh machines with things like
> > Postgres 7.2 is just embarrassing.
> 
> I am not embarrassed.

Well perhaps you should be. Whenever they ask for support those users will be
told the version their running is hopelessly out of date and all the trouble
their having is because of their choice of version. (Actually the postgres
list does an admirable job of attempting to provide support for 7.2 and even
7.1 but inevitably the answer turns out to be "that problem was fixed in 7.3.
Or now, 7.4.)

Those users will also be struggling with major production issues like being
unable to run 24x7 because of required periodic maintenance (vacuum and
reindex) that require downtime.

Basically, given that 7.3 has been out for an entire release cycle (7.4 will
be released within days), giving 7.2 to new users is simply ridiculous. 

The same holds for having new users install 2.2 kernels or XFree86 4.1. I
mean, sure there are cases when these things are passable or even useful, but
by default telling a new user that these awful buggy releases are what he or
she should be installing on a fresh install is just, well, as I said,
embarrassing.

Personally I'm of the opinion that stable is useless. It certainly has no use
for me. Perhaps if I ran a production server on debian I might think otherwise
but I rather doubt it. When I had production servers they all ran 2.4 and
needed the latest stable releases of anything important like database, mail,
web server services.

If I ran production servers on debian today I would probably pick an arbitrary
date off snapshot.debian.org and declare that my "stable". If I had security
problems I would pick a date recent enough to have the security fixes, test
it, and declare it "stable".

It wouldn't be guaranteed to be bug-free, but then nothing is. Stable has tons
of minor bugs that no upstream maintainer would listen to because they were
fixed aeons ago anyways, or more likely are no longer relevant in current
sources.

-- 
greg

Re: A case study of a new user turned off debian

[Greg Stark]

> > Josip Rodin <[EMAIL PROTECTED]> writes:
> > > 
> > > It would be helpful if people wouldn't make sweeping generalizations
> > > all the time. 

"All the time"? ...

-- 
greg

Re: A case study of a new user turned off debian

[Greg Stark]

Darren Salt <[EMAIL PROTECTED]> writes:

> BTW, no need to Cc: me - or did Gnus not notice the Mail-Followup-To header?

Uhm. What Mail-Followup-To header? I didn't receive one on this message,
perhaps it's stripped by the mail server? Or perhaps you're mistaken about it
being included?

I've attached the original message including headers.

--- Begin Message ---
I demand that Greg Stark may or may not have written...

> Darren Salt <[EMAIL PROTECTED]> writes:
>> I demand that Greg Stark may or may not have written...

> What does that mean?

It's (more or less) from The Hitch-Hiker's Guide to the Galaxy. The bit in
question concerns two philosophers who are trying to stop the computer Deep
Thought from being programmed to find the Answer to the Question of Life, The
Universe and Everything.

If you want to know more, then I suggest that you read the first of the
books, watch the TV series or listen to the original radio series...

BTW, no need to Cc: me - or did Gnus not notice the Mail-Followup-To header?

-- 
| Darren Salt   | linux (or ds) at | nr. Ashington,
| woody, sarge, | youmustbejoking  | Northumberland
| RISC OS   | demon co uk  | Toon Army
|   Let's keep the pound sterling

Barney of Borg: "I assimilate you; you assimilate me..."


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]


--- End Message ---



-- 
greg

Re: A case study of a new user turned off debian

[Graham Wilson]

On Tue, Nov 04, 2003 at 11:56:37AM -0500, Greg Stark wrote:
> Josip Rodin <[EMAIL PROTECTED]> writes:
> > On Tue, Nov 04, 2003 at 01:53:36AM -0600, Chris Cheney wrote:
> > > It would be helpful if Debian could even be installed on machines newer
> > > than about 2 years old.
> > 
> > It would be helpful if people wouldn't make sweeping generalizations
> > all the time. Only a part of the new machines are made with hardware
> > that needs particularly special drivers, there's plenty of it that
> > works fine.
>
> [...]
> 
> Regardless. Having people install fresh machines with things like
> Postgres 7.2 is just embarrassing.

I am not embarrassed.

-- 
gram


signature.asc
Description: Digital signature

Re: A case study of a new user turned off debian

[Darren Salt]

I demand that Matt Zimmerman may or may not have written...

> On Wed, Nov 05, 2003 at 05:20:18PM +, Darren Salt wrote:
>> I keep some around. I'd prefer better management of this, though: ATM all
>> that I can do (with apt-get/aptitude) is remove all older versions or
>> purge the cache.

> I use a dead simple cron.daily script which prunes packages with an atime
> above a certain threshold.

Hmm. That sounds useful; perhaps it, or something similar, should be in apt.
Useful configuration options, AFAICS, would be atime or ctime (and, of
course, a threshold) and/or disk space used or remaining.

(Yes, I know. Something for me to do... now, if there's already a script
which can be put to this use...)

> It would be only slightly less trivial to move the packages into another
> directory and run apt-ftparchive.

Maybe... my use of apt-proxy on one of my machines gives me the option of
cleaning the cache after an upgrade. I don't plan to do this, though ;-)

-- 
| Darren Salt   | nr. Ashington, | linux (or ds) at
| woody, sarge, | Northumberland | youmustbejoking
| RISC OS   | Toon Army  | demon co uk
|   Oh, sarge too...

Smile... people will wonder what you've been up to.

Re: A case study of a new user turned off debian

[Darren Salt]

I demand that Greg Stark may or may not have written...

> Darren Salt <[EMAIL PROTECTED]> writes:
>> I demand that Greg Stark may or may not have written...

> What does that mean?

It's (more or less) from The Hitch-Hiker's Guide to the Galaxy. The bit in
question concerns two philosophers who are trying to stop the computer Deep
Thought from being programmed to find the Answer to the Question of Life, The
Universe and Everything.

If you want to know more, then I suggest that you read the first of the
books, watch the TV series or listen to the original radio series...

BTW, no need to Cc: me - or did Gnus not notice the Mail-Followup-To header?

-- 
| Darren Salt   | linux (or ds) at | nr. Ashington,
| woody, sarge, | youmustbejoking  | Northumberland
| RISC OS   | demon co uk  | Toon Army
|   Let's keep the pound sterling

Barney of Borg: "I assimilate you; you assimilate me..."

Re: A case study of a new user turned off debian

[Matt Zimmerman]

On Wed, Nov 05, 2003 at 05:20:18PM +, Darren Salt wrote:

> I keep some around. I'd prefer better management of this, though: ATM all
> that I can do (with apt-get/aptitude) is remove all older versions or
> purge the cache.

I use a dead simple cron.daily script which prunes packages with an atime
above a certain threshold.  It would be only slightly less trivial to move
the packages into another directory and run apt-ftparchive.

-- 
 - mdz

Re: A case study of a new user turned off debian

[Greg Stark]

Darren Salt <[EMAIL PROTECTED]> writes:

> > If apt kept even a single old revision in its cache then rolling back could
> > be as simple as
> >   apt-get install -t previous libc6
> 
> That would be good. (Similarly for aptitude, of course.)
> 
> One question occurs, however: should this also (try to) roll back packages on
> which $PREVIOUS depends?

Well that would be the main advantage. The user can do the above fairly easily
if he can find the .deb but he has to track down all the dependencies that APT
broke trying to install the broken package.

It should roll back anything to previous that depend on a newer version of
libc6 than being installed and recursively.

-- 
greg

Re: A case study of a new user turned off debian

[Greg Stark]

Darren Salt <[EMAIL PROTECTED]> writes:

> I demand that Greg Stark may or may not have written...

What does that mean?

-- 
greg

Re: A case study of a new user turned off debian

[Darren Salt]

I demand that Greg Stark may or may not have written...

[snip - package rollback?]
> all it would take to make the tools handle this would be to somehow make
> apt aware of more revisions of packages. They're all in the pool after all.
> Short of making some king of humongous mega-Packages file with every
> revision of every package -- which apt wouldn't scale up to anyways --
> they're currently unavailable to APT.

> The low hanging fruit here would be to have APT keep packages you had
> installed yourself in the cache rather than immediately discarding them as
> soon as they're upgraded.

I keep some around. I'd prefer better management of this, though: ATM all
that I can do (with apt-get/aptitude) is remove all older versions or purge
the cache.

OTOH I'm running apt-proxy on my dialup box.

> At a minimum keeping one extra revision would at least let you roll back.
> Something more flexible keeping old revisions for n days after being
> replaced would be even cooler.

Yes. It'd help to avoid /var becoming full - this happened yesterday on one
of my desktop boxes. One 'aptitude clean' later, and 2/3 of /var was free.

[snip]
> If apt kept even a single old revision in its cache then rolling back could
> be as simple as
>   apt-get install -t previous libc6

That would be good. (Similarly for aptitude, of course.)

One question occurs, however: should this also (try to) roll back packages on
which $PREVIOUS depends?

> or perhaps a little less automatic,
>   apt-cache show libc6
> to list the available revisions then explicitly
>   apt-get install libc6:2.3.2-8

(Noting reference to "=" syntax elsewhere...)

> Actually this wouldn't really have helped my friend at all because he was
> unlucky enough that the *first* version of libc6 from unstable that he saw
> happened to be the buggy one. That doesn't really happen that often to
> libc6 so he had particularly bad luck there.

Perhaps it should be the default that, when installing from unstable, the
package files for testing are also fetched...?

-- 
| Darren Salt   | linux (or ds) at | nr. Ashington,
| woody, sarge, | youmustbejoking  | Northumberland
| RISC OS   | demon co uk  | Toon Army
|   http://www.youmustbejoking.demon.co.uk/> (PGP 2.6, GPG keys)

Give me all your lupins!

Re: A case study of a new user turned off debian

[Adam Heath]

On 4 Nov 2003, Greg Stark wrote:

> So all it would take to make the tools handle this would be to somehow make
> apt aware of more revisions of packages. They're all in the pool after all.
> Short of making some king of humongous mega-Packages file with every revision
> of every package -- which apt wouldn't scale up to anyways -- they're
> currently unavailable to APT.

snapshot.debian.net

Re: A case study of a new user turned off debian

[Jonathan Dowland]

On Mon, Nov 03, 2003 at 08:12:38PM +, Steve Kemp wrote:
> On Mon, Nov 03, 2003 at 03:05:56PM -0500, Greg Stark wrote:
> 
> > All he had to do was install an older version of libc6 and every other 
> > package
> > would have been happy. All the infrastructure is there to do this, the old
> > packages are all on the ftp/http sites, the package may even be sitting in
> > apt's cache. But there's no interface for it.
> 
>   Sure there is, 'dpkg --install $foo.deb'.  Doesn't that do exactly the
>  correct thing, even at the expense of downgrading?

So its possible.. it isn't easy, intuitive or flexible. I suppose he
could grab the pristine tarball too.

-- 
Jon Dowland
http://jon.dowland.name/

Re: [debian-devel] Re: A case study of a new user turned off debian

[Jonathan Dowland]

On Tue, Nov 04, 2003 at 06:48:09AM +, Magos?nyi ?rp?d wrote:
> I guess having kernel-image-vanilla and/or
> kernel-image-onlybugfix in debian would not hurt.

Or kernel-image-hx ... what caught me out was believing
kernel-image-2.4.xx or whatever was relatively pristine. However naive
that makes me I know I am not alone.

> I think the same about the rest, however I understand that
> Herbert does the best kernel debs you can apt-get install,
> and he does a very importand job no one else is willing to
> do, and he does it well (but with a different approach I or
> you do not do).

I definitely appreciate Herbert's work, regardless of my differences of
opinion on how it should be done :-)

-- 
Jon Dowland
http://jon.dowland.name/

Re: A case study of a new user turned off debian

[Steve Greenland]

On 03-Nov-03, 16:22 (CST), Erik Steffl <[EMAIL PROTECTED]> wrote: 
>   Oh, not this crap again. Or perhaps you're contending that what is 
> usefull for you is usefull for everybody.

No, I didn't, My point was to object to YOUR contention that anything
over 3 months old is "Useless". Woody Emacs works just fine. So does
GNOME 1.2. Not as pretty, perhaps, as 2.4, but it seems to start up a
hell of lot quicker. And Galeon 1.2.9 works a lot better than what's in
stable - more features (that I find useful, anyway), fewer bugs.

And I object to people who insist on installing 'unstable" and then
bitch when it breaks. There's reasons for testing, and the only way it's
worse than unstable (as some have said) is that you have build your own
security fixes. Hey, baby, such is life.

And for the record, you can get the vast majority of the new desktop
stuff from one or more backport repositories (see www.apt-get.org), all
with the safety of a stable libc.

The woody install instructions describe how to use an alternate kernel
for installation. There's absolutely nothing preventing you from making
a CD image with a newer kernel. If there's such a huge fscking demand
for it, why isn't it out there?

And yes, I'm going to continue to object everytime someone makes stupid
statements like "Woody is so old it's useless". That's just noise,
unless you add "... for purposes a, b, and c".

Steve

-- 
Steve Greenland
The irony is that Bill Gates claims to be making a stable operating
system and Linus Torvalds claims to be trying to take over the
world.   -- seen on the net

Re: A case study of a new user turned off debian

[Brian May]

On Tue, Nov 04, 2003 at 12:01:10PM -0500, Greg Stark wrote:
> Hm, it appears to be true that not every single revision is here. But there
> are certainly more than just the unstable and testing revisions too:
> 
>  libc6_2.2.5-11.2_i386.deb   26-Sep-2002 11:32   3.2M  

stable

>  libc6_2.2.5-11.5_i386.deb   08-Jun-2003 01:32   3.2M  

proposed-updates

>  libc6_2.3.2-9_i386.deb  26-Oct-2003 21:47   3.6M  

testing

>  libc6_2.3.2.ds1-8_i386.deb  30-Oct-2003 12:17   4.6M  

guessing: will get deleted soon if hasn't been deleted
already.

Packages aren't removed the instant the are no longer referenced,
there is a delay of several days.

>  libc6_2.3.2.ds1-9_i386.deb  02-Nov-2003 11:18   4.6M  

unstable.

Source: madison on ftp-master:


[EMAIL PROTECTED]:~$ madison libc6
W: Archive maintenance is in progress; database inconsistencies are possible.
 libc6 | 2.2.5-11.2 |stable | arm, hppa, i386, m68k, mips, mipsel, 
powerpc, s390, sparc
 libc6 | 2.2.5-11.5 | proposed-updates | arm, hppa, i386, m68k, mips, 
mipsel, powerpc, s390, sparc
 libc6 |2.3.2-7 |   testing | mips, mipsel
 libc6 |2.3.2-7 |  unstable | mips
 libc6 |2.3.2-9 |   testing | arm, hppa, i386, m68k, powerpc, s390, 
sparc
 libc6 | 2.3.2.ds1-8 |  unstable | arm, m68k
 libc6 | 2.3.2.ds1-9 |  unstable | hppa, i386, mipsel, powerpc, s390, 
sparc
-- 
Brian May <[EMAIL PROTECTED]>

Re: A case study of a new user turned off debian

[Matt Zimmerman]

On Tue, Nov 04, 2003 at 01:11:43AM -0500, Joey Hess wrote:

> Suggested project: Create a package that, a-l-apt-move, pulls packages
> out of the apt cache and creates apt repositories from them. But make it
> create a new repository after every upgrade, by hooking into apt. And
> auto-add these repositories to sources.list, and remove old ones after a
> while, the whole nine yards.

This was more or less how I thought this should look, if it were worth
implementing.  But it should work just as well with a single repository,
pruned from time to time.  All that'd be required would be a tool to move
packages out of apt's cache and run apt-ftparchive on them.

> Eight reasons why you should be using aptitude instead of apt-get.
> 
> 1. aptitude can look just like apt-get
> 
>If you run 'aptitude update' or 'aptitude upgrade' or 'aptitude
>install', it looks and works just like apt-get, with a few enhancements.
>So there is no learning curve.

For the most common operations, there is no learning curve (and there is a
lot more power, e.g. the Y/n/? prompt).  However, aptitude doesn't quite
supersede apt-get yet, and there are two significant reasons why I'm not
quite yet able to tell users "use aptitude" as an unconditional replacement
for apt-get (which I otherwise would like to do).

1. Source package handling is not (yet?) implemented

2. aptitude in woody has some significant bugs (for example, it can't
install imp), so stable users could run into some unexpected problems

> 6. aptitude makes it easy to keep track of obsolete software
> 
>If Debian stops distributing a package, apt will leave it on your system
>indefinitly, with no warnings, and no upgrades. Aptitude lists such
>packages in its "Obsolete and Locally Created Packages" section, so you
>can be informed of the problem and do something about it.

This method of dealing with obsolete packages (dselect also does this,
right?) is much better than doing nothing, but still leaves something to be
desired.  The difference between a locally created package and an obselete
package from Debian is great, but they look the same in such a display.

-- 
 - mdz

Re: A case study of a new user turned off debian

[Andrew Suffield]

On Tue, Nov 04, 2003 at 05:53:04PM +0800, Cameron Patrick wrote:
> On Tue, Nov 04, 2003 at 03:22:10AM -0600, Chris Cheney wrote:
> | I refuse to use nvidia products, but I somehow doubt that boards based
> | on their nforce2 chipset work properly either.
> 
> I have a machine using the nforce2 chipset and the Woody installer
> doesn't recognise its IDE controller.  (Proper support is only in kernel
> versions >= 2.4.21.)

Actually it does, at least well enough to install. You need a kernel
argument to persuade it to work, and I can't remember what it was
offhand, but it's quite doable. I believe it is merely that the kernel
can't autodetect the presence of this controller in older versions.

-- 
  .''`.  ** Debian GNU/Linux ** | Andrew Suffield
 : :' :  http://www.debian.org/ |
 `. `'  |
   `- -><-  |


signature.asc
Description: Digital signature

Re: A case study of a new user turned off debian

[Matt Zimmerman]

On Tue, Nov 04, 2003 at 12:47:30AM -0500, Greg Stark wrote:

> So all it would take to make the tools handle this would be to somehow
> make apt aware of more revisions of packages. They're all in the pool
> after all.  Short of making some king of humongous mega-Packages file with
> every revision of every package -- which apt wouldn't scale up to anyways
> -- they're currently unavailable to APT.
> 
> The low hanging fruit here would be to have APT keep packages you had
> installed yourself in the cache rather than immediately discarding them as
> soon as they're upgraded. At a minimum keeping one extra revision would at
> least let you roll back. Something more flexible keeping old revisions for
> n days after being replaced would be even cooler.

So because it's not practical for apt to try to maintain a database of all
packages it has ever seen, in addition to the current (enormous) package
database, you're suggesting that apt try to maintain a database of all
packages it has ever installed.  This seems to have exactly the same
problem (unbounded growth) on a somewhat smaller scale.  What's more, it's
already been implemented at snapshot.debian.net.

Trying to roll back to an earlier version of a package is not a "beginner
friendly" sort of operation.  If you don't know what's going on behind the
scenes, you're in for a world of pain even if the tools made it easier for
you to do this.  Debian packages don't expect to be downgraded, and it's
rather unlikely to do the right thing for non-trivial packages.

> or perhaps a little less automatic,
> 
> apt-cache show libc6
> 
> to list the available revisions then explicitly
> 
> apt-get install libc6:2.3.2-8

Which, surprisingly enough, is already implemented by apt, using '=' where
you have written ':'.

> Actually this wouldn't really have helped my friend at all because he was
> unlucky enough that the *first* version of libc6 from unstable that he saw
> happened to be the buggy one. That doesn't really happen that often to
> libc6 so he had particularly bad luck there.

So, in other words, implementing the feature you have described wouldn't
have fixed the problem in the first place.

-- 
 - mdz

Re: A case study of a new user turned off debian

[Josip Rodin]

On Tue, Nov 04, 2003 at 11:56:37AM -0500, Greg Stark wrote:
> > > It would be helpful if Debian could even be installed on machines newer
> > > than about 2 years old.
> > 
> > It would be helpful if people wouldn't make sweeping generalizations all the
> > time. Only a part of the new machines are made with hardware that needs
> > particularly special drivers, there's plenty of it that works fine.
> 
> Sure, as long as I didn't need the network interface or access to all my hard
> drives 2.2 would have been fine.

Hey, I installed a machine recently that had a set of network cards that
didn't work with our bf24 kernel. Just as I was about to mumble something
about Debian, it turned out that the driver isn't in 2.4.22 either. I'm not
particularly bothered, workarounds are a fact of life.

-- 
 2. That which causes joy or happiness.

Re: A case study of a new user turned off debian

[Greg Stark]

Chris Cheney <[EMAIL PROTECTED]> writes:

> Er no they are not all in the pool. The only packages in the pool are
> the current versions for stable/testing/unstable/experimental. There are
> also the few packages that haven't been completely compiled on all archs
> yet and so are still left in archive while this is being done.
> snapshot.debian.net has nearly every deb since 2002/06/04, but it is not
> an official debian repo afaik.

Hm, it appears to be true that not every single revision is here. But there
are certainly more than just the unstable and testing revisions too:

 libc6_2.2.5-11.2_i386.deb   26-Sep-2002 11:32   3.2M  
 libc6_2.2.5-11.5_i386.deb   08-Jun-2003 01:32   3.2M  
 libc6_2.3.2-9_i386.deb  26-Oct-2003 21:47   3.6M  
 libc6_2.3.2.ds1-8_i386.deb  30-Oct-2003 12:17   4.6M  
 libc6_2.3.2.ds1-9_i386.deb  02-Nov-2003 11:18   4.6M  

As it turned out 2.3.2-9 was a perfectly reasonable revision to roll back to.

-- 
greg

Re: A case study of a new user turned off debian

[Greg Stark]

Josip Rodin <[EMAIL PROTECTED]> writes:

> On Tue, Nov 04, 2003 at 01:53:36AM -0600, Chris Cheney wrote:
> > It would be helpful if Debian could even be installed on machines newer
> > than about 2 years old.
> 
> It would be helpful if people wouldn't make sweeping generalizations all the
> time. Only a part of the new machines are made with hardware that needs
> particularly special drivers, there's plenty of it that works fine.

Sure, as long as I didn't need the network interface or access to all my hard
drives 2.2 would have been fine.

Hm...

Regardless. Having people install fresh machines with things like Postgres 7.2
is just embarrassing.

-- 
greg

Re: A case study of a new user turned off debian

[Josip Rodin]

On Tue, Nov 04, 2003 at 01:53:36AM -0600, Chris Cheney wrote:
> It would be helpful if Debian could even be installed on machines newer
> than about 2 years old.

It would be helpful if people wouldn't make sweeping generalizations all the
time. Only a part of the new machines are made with hardware that needs
particularly special drivers, there's plenty of it that works fine.

-- 
 2. That which causes joy or happiness.

Re: A case study of a new user turned off debian

[Marcelo E. Magallon]

On Tue, Nov 04, 2003 at 03:22:10AM -0600, Chris Cheney wrote:

 > Intel
 > -
 > i845 10 Sept 2001
 > i875 14 Apr  2003
 > i865 21 May  2003

 The last two don't have AGP support before 2.4.23-preX (not sure if X
 is 1 or not), and the on-board network cards which many of the the
 motherboards based on this chipset use aren't supported until
 2.4.23-preY (Y around 3, IIRC).  I remember having had _a lot_ of
 trouble with DMA, but IIRC the thing did boot with older kernels.

-- 
Marcelo

Re: A case study of a new user turned off debian

[Karsten M. Self]

on Mon, Nov 03, 2003 at 04:57:34PM -0500, Greg Stark ([EMAIL PROTECTED]) wrote:
> 
> "Julian Mehnle" <[EMAIL PROTECTED]> writes:
> 
> > First, I think what Daniel Jacobowitz said is entirely true. Why didn't you
> > start with "testing"?
> 
> Sure testing is less likely to trigger this. 

Frankly, I'd have started stable.  Moving up's tons easier than moving
down.  As you've noted.  And largely trivial.


> > Or, you could create a file /etc/apt/preferences and pin the
> > "testing" version of the package with a high enough priority. See
> > `man apt_preferences`. Then do a `apt-get dist-upgrade`.
> 
> That's about the last place I would send a new user. I read that man
> page about three times during this crisis before I decided it would be
> hopeless to try to explain this procedure online. 

Explain what?

Greg:  Grab this file, copy it to /etc/apt/preferences:
/preferences
Friend:  OK.

> This is what I meant about there not being an "interface". 

File copying, particularly for an experienced sysadmin, is a damned
useful "interface".


> If apt said "Hm, version 1.2 of libc failed to configure, would you
> like to install the previous version (1.1) from testing and hold back
> the following packages that depend on the new one (awk, grep, sed)
> [Yn]?" That would be an interface. 

Pinning can get you much of the way here.

Problem is that dependencies work forward:  you can't be sure of getting
a package in a prior release -- it may not be available.

libc is a particularly pathelogical case, for all the obvious reasons.


> Telling people, go edit this random file with to set "pin priorities"
> for things to arbitrary numbers, find out what package dependencies
> fail, add those to your list of pin priorities, etc. That's not a
> useful interface for this case.

Telling people to start off with unstable is about as useless.


> In any case having the granularity of "stable", "testing", "unstable"
> really doesn't help. All the package versions are in the pool. I want
> to be able to tell apt to try such and such version, or at least put
> back the version I had before and restore whatever other packages it
> must to satisfy dependencies.

Look:  Either KISS and stick with stable, accept the pain of testing, or
learn how to use pinning and come up with a perferences file that works.
Your friend can copy this from any given website you can access.


> I didn't say it was a good idea or that it was going to work. 
> My whole point is that that approach sucks and we should make
> something more effective rather than leave the admin stuck.

Assuming _you_ have experience with Debian, and are aware of limitations
and trade-offs, you've led your friend astray.

I spend a fair amount of time in IRC support for #debian.  Lots of n00bs
come on wanting to install Debian unstable.  The advice is consistent:
Don't.  Not if you don't know your way around tha packaging system yet.
I've had a number of friends convert to Debian -- most of them love it,
aggressively.  But to a one they all wanted to go straight to unstable.
After nearly five years, I still hold to testing with a few exceptions
handled through pins.


Peace.

-- 
Karsten M. Self http://kmself.home.netcom.com/
 What Part of "Gestalt" don't you understand?
Bush/Cheny '04:  BU__SH__!


signature.asc
Description: Digital signature

Re: A case study of a new user turned off debian

[Rob Weir]

On Tue, Nov 04, 2003 at 12:47:30AM -0500, Greg Stark said
> or perhaps a little less automatic,
> 
> apt-cache show libc6
> 
> to list the available revisions then explicitly
> 
> apt-get install libc6:2.3.2-8

With a s/:/=/ and a sources.list line pointing at sid, say, two days ago
on http://snapshot.debian.net/, this would have worked.  If the Debian
Reference doesn't mention this, it certainly should, but this is
something that your friend could have solved by asking another Debian
user IRL, on debian-user or on IRC.  When people say "newbies shouldn't
use unstable", they're not (generally) being elitist, it's shorthand for
"newbies won't know about the tools they need to fix their systems when
unstable goes bad".

-- 
Rob Weir <[EMAIL PROTECTED]> | [EMAIL PROTECTED]  |  Do I look like I want a CC?
Words of the day:   colonel encryption BATF Manfurov hackers CipherTAC-2000 ANC


signature.asc
Description: Digital signature

Re: A case study of a new user turned off debian

[Arnaud Vandyck]

On Tue, 4 Nov 2003 12:18:00 +
Colin Watson <[EMAIL PROTECTED]> wrote:

> On Tue, Nov 04, 2003 at 12:54:25PM +0100, Arnaud Vandyck wrote:
> > On Tue, 4 Nov 2003 10:04:11 +
> > Will Newton <[EMAIL PROTECTED]> wrote:
> > > Er, doesn't apt-get install libc6=2.3.2-8 do exactly this?
> > 
> > No because of a conflict with (file own by) libdb1-compat.
> 
> Uh, 2.3.2-8 is well after the introduction of libdb1-compat. Are you
> sure you don't mean an earlier version?

Sorry 2.2.5, the one in stable.

-- 
  .''`.   ** Debian GNU/Linux **
 : :' :  Arnaud   Vandyck
 `. `'   http://people.debian.org/~avdyk/
   `-


pgpzc2p8ojshh.pgp
Description: PGP signature

Re: A case study of a new user turned off debian

[Colin Watson]

On Tue, Nov 04, 2003 at 12:54:25PM +0100, Arnaud Vandyck wrote:
> On Tue, 4 Nov 2003 10:04:11 +
> Will Newton <[EMAIL PROTECTED]> wrote:
> > Er, doesn't apt-get install libc6=2.3.2-8 do exactly this?
> 
> No because of a conflict with (file own by) libdb1-compat.

Uh, 2.3.2-8 is well after the introduction of libdb1-compat. Are you
sure you don't mean an earlier version?

-- 
Colin Watson  [EMAIL PROTECTED]

Re: A case study of a new user turned off debian

[Arnaud Vandyck]

On Tue, 4 Nov 2003 10:04:11 +
Will Newton <[EMAIL PROTECTED]> wrote:

> Er, doesn't apt-get install libc6=2.3.2-8 do exactly this?

No because of a conflict with (file own by) libdb1-compat.

Well, it was the problem I faced.

Best regards,

-- 
  .''`.   ** Debian GNU/Linux **
 : :' :  Arnaud   Vandyck
 `. `'   http://people.debian.org/~avdyk/
   `-


pgpzlXkz7SAQH.pgp
Description: PGP signature

Re: A case study of a new user turned off debian

[Marco d'Itri]

On Nov 04, Andrew Suffield <[EMAIL PROTECTED]> wrote:

 >On Tue, Nov 04, 2003 at 01:53:36AM -0600, Chris Cheney wrote:
 >> It would be helpful if Debian could even be installed on machines newer
 >> than about 2 years old. Neither my KT400 based machine nor my i875P
 >> based machine could be installed using the standard Debian
 >> boot-floppies. I had to resort to using Knoppix with debootstrap.
 >> 
 >> So no Debian stable really is not an option for a large portion of
 >> users.
 >
 >I challenge the assertion that this affects a large portion of users.
I can report I had the same problem with two other KT400-based
workstations. The kernel (both plain and bf24) would panic at boot or
break shortly after.

00:00.0 Host bridge: VIA Technologies, Inc. VT8377 [KT400 AGP] Host Bridge

-- 
ciao, |
Marco | [2889 diFNHN2DNpWWo]


signature.asc
Description: Digital signature

Re: A case study of a new user turned off debian

[Andreas Metzler]

Chris Cheney <[EMAIL PROTECTED]> wrote:
[...]
> Regardless updating boot-floppies when new kernels come out would be a
> good idea so that newbies with recent machines can still use Debian.

Iirc Eduard Bloch (blade) provides inofficial updated boot-floppies on
people.debian.org.
  cu andreas

Re: A case study of a new user turned off debian

[Will Newton]

On Tuesday 04 Nov 2003 05:47, Greg Stark wrote:

> to list the available revisions then explicitly
>
> apt-get install libc6:2.3.2-8
>
> Actually this wouldn't really have helped my friend at all because he was
> unlucky enough that the *first* version of libc6 from unstable that he saw
> happened to be the buggy one. That doesn't really happen that often to
> libc6 so he had particularly bad luck there.

Er, doesn't apt-get install libc6=2.3.2-8 do exactly this?

man pages are a wonderful thing.

Re: A case study of a new user turned off debian

[Cameron Patrick]

On Tue, Nov 04, 2003 at 03:22:10AM -0600, Chris Cheney wrote:
| I refuse to use nvidia products, but I somehow doubt that boards based
| on their nforce2 chipset work properly either.

I have a machine using the nforce2 chipset and the Woody installer
doesn't recognise its IDE controller.  (Proper support is only in kernel
versions >= 2.4.21.)

Incidentally, wasn't there a woody 3.0r2 planned to be released a couple
of months ago with a newer kernel version and miscellaneous security
fixes?

Cameron.

Re: A case study of a new user turned off debian

[Chris Cheney]

On Tue, Nov 04, 2003 at 08:17:06AM +, Andrew Suffield wrote:
> I challenge the assertion that this affects a large portion of users.
> 
> In the past few months, I've installed woody on roughly 30-40
> different types of box, all aged 0-3 years, and only one was
> unsupported by woody (and that was freaky new apple kit).
> 
> I think you have unusual hardware.

Neither are that unusual. I suppose it may have been a little highend
for some when I first bought them, but now they are relatively cheap,
an AthlonXP Via KT400 based motherboard now is as cheap as $42 USD
and the P4 i865/i875 chipset series is as cheap as $62 USD. Perhaps all
of your 0-3 year old systems had older chipset logic in them, or perhaps
truly new systems just aren't common in your country. Via KT400 was
released 16 Aug 2002 and the i865/i875 line was released 14 Apr 2003.
Note that nearly all P4 systems with the 800MHz FSB use the i865/i875
chipsets. I refuse to use nvidia products, but I somehow doubt that
boards based on their nforce2 chipset work properly either.

Regardless updating boot-floppies when new kernels come out would be a
good idea so that newbies with recent machines can still use Debian.

Chris


Release Dates:

Intel
-
i84510 Sept 2001
i87514 Apr  2003
i86521 May  2003

Nvidia
--
nforce   4 June 2001
nforce2 16 July 2002

Via
---
KT266   20 Sept 2000
KT266A   3 Sept 2001
KT333   20 Feb  2002
KT400   16 Aug  2002
KT400A  10 Mar  2003
KT600   18 Jun  2003


signature.asc
Description: Digital signature

Re: A case study of a new user turned off debian

[Chris Cheney]

On Tue, Nov 04, 2003 at 12:47:30AM -0500, Greg Stark wrote:
> So all it would take to make the tools handle this would be to somehow make
> apt aware of more revisions of packages. They're all in the pool after all.
> Short of making some king of humongous mega-Packages file with every revision
> of every package -- which apt wouldn't scale up to anyways -- they're
> currently unavailable to APT.

Er no they are not all in the pool. The only packages in the pool are
the current versions for stable/testing/unstable/experimental. There are
also the few packages that haven't been completely compiled on all archs
yet and so are still left in archive while this is being done.
snapshot.debian.net has nearly every deb since 2002/06/04, but it is not
an official debian repo afaik.

Chris


signature.asc
Description: Digital signature

Re: A case study of a new user turned off debian

[Andrew Suffield]

On Mon, Nov 03, 2003 at 03:05:56PM -0500, Greg Stark wrote:
> I finally convinced a sysadmin friend of mine that Debian was the way and the
> light.

Ah, there's your mistake.

Don't do that. Anybody who uses Debian as a result of this sort of
evangelism is going to have a similar experience, so it's
self-defeating and wastes our time. Debian requires that users have a
desire to make it work.

> All he had to do was install an older version of libc6 and every other package
> would have been happy.

Heh, yaright. Downgrading libc6 tends to break stuff, that's why it
isn't supported.

> All the infrastructure is there to do this,

...except for the bit where the packages support downgrading, which
they don't. They never have, so supporting it in the frontends is
fairly futile.

-- 
  .''`.  ** Debian GNU/Linux ** | Andrew Suffield
 : :' :  http://www.debian.org/ |
 `. `'  |
   `- -><-  |


signature.asc
Description: Digital signature

Re: A case study of a new user turned off debian

[Andrew Suffield]

On Tue, Nov 04, 2003 at 01:53:36AM -0600, Chris Cheney wrote:
> It would be helpful if Debian could even be installed on machines newer
> than about 2 years old. Neither my KT400 based machine nor my i875P
> based machine could be installed using the standard Debian
> boot-floppies. I had to resort to using Knoppix with debootstrap.
> 
> So no Debian stable really is not an option for a large portion of
> users.

I challenge the assertion that this affects a large portion of users.

In the past few months, I've installed woody on roughly 30-40
different types of box, all aged 0-3 years, and only one was
unsupported by woody (and that was freaky new apple kit).

I think you have unusual hardware.

-- 
  .''`.  ** Debian GNU/Linux ** | Andrew Suffield
 : :' :  http://www.debian.org/ |
 `. `'  |
   `- -><-  |


signature.asc
Description: Digital signature

Re: A case study of a new user turned off debian

[Chris Cheney]

On Mon, Nov 03, 2003 at 03:43:29PM -0600, Steve Greenland wrote:
> On 03-Nov-03, 14:21 (CST), Erik Steffl <[EMAIL PROTECTED]> wrote: 
> >  > is even worse than unstable>
> 
> Oh, not this crap again. Or perhaps you're contending that I've not
> gotten anything done at work in the last two years using my "useless"
> Debian stable desktop.
> 
> Hint: there's more to "useful" than running the latest version of
> everything. Particularly for a sys admin, who I'd expect to be heavily
> command line oriented, and who doesn't need the latest 3-d games or dvd
> player[1].

It would be helpful if Debian could even be installed on machines newer
than about 2 years old. Neither my KT400 based machine nor my i875P
based machine could be installed using the standard Debian
boot-floppies. I had to resort to using Knoppix with debootstrap.

So no Debian stable really is not an option for a large portion of
users. At least anyone who has a machine newer than when the kernel
on b-f was last updated in Woody's case is kernel 2.4.18 from
Feb 25 2002.

Chris Cheney


signature.asc
Description: Digital signature

Re: [debian-devel] Re: A case study of a new user turned off debian

[Magosányi Árpád]

A levelezőm azt hiszi, hogy [EMAIL PROTECTED] a következőeket írta:
> > > but I wouldn't touch Herbert's kernels with a ten-feet pole.
[]
>   a) I can do better


Please do then.

I guess having kernel-image-vanilla and/or
kernel-image-onlybugfix in debian would not hurt.

>   b) I don't do huge monolitic patches
>   c) I don't like Herbert's taste
>   d) some boxen are used for 2.6 work

I think the same about the rest, however I understand that
Herbert does the best kernel debs you can apt-get install,
and he does a very importand job no one else is willing to
do, and he does it well (but with a different approach I or
you do not do).

-- 
GNU GPL: csak tiszta forrásból

Re: A case study of a new user turned off debian

[Goswin von Brederlow]

Greg Stark <[EMAIL PROTECTED]> writes:

> I finally convinced a sysadmin friend of mine that Debian was the way and the
> light. He started a new job and showed up on his first day to set up his
> machine by installing Debian. In short, things went horribly wrong and he
> started this new job by wasting two days picking up the pieces. He's now very
> leery of suggesting using Debian on other machines at work or of using it
> himself at home.
> 
> What started the chain of events was that a fairly routine minor bug bit the
> latest libc6 release. He's an experienced sysadmin though and wasn't the least
> bit fazed by that. What drove him batty was that it was so hard to recover
> from the mess and all the obvious avenues just made the problem worse.

So he starts out on his first day of work trying out Debian for the
very first time and uses unstbale? No pitty there then.

> All he had to do was install an older version of libc6 and every other package
> would have been happy. All the infrastructure is there to do this, the old
> packages are all on the ftp/http sites, the package may even be sitting in
> apt's cache. But there's no interface for it.

apt-get libc6=version

> The only interface for rolling back is switching the entire machine to an
> earlier distribution and telling apt to try to downgrade -- which is unlikely
> to work. And worse, every time you run apt it only downloads and unpacks
> *more* packages, all of which, of course, fail as well.

snapshot.debian.net just rocks for previous versions.

> What would be really neat would be if aptitude or perhaps even apt checked for
> earlier versions of the package in the pool and offered them as options if the
> current one fails to configure.

There never are any previous versions in the pool of ftp.debian.org.
The old version is removed when the new one enters.

And debs in the cache are easily installed with dpkg if need be.

MfG
Goswin

Re: A case study of a new user turned off debian

[Joey Hess]

Greg Stark wrote:
> The only interface for rolling back is switching the entire machine to an
> earlier distribution and telling apt to try to downgrade -- which is unlikely
> to work. And worse, every time you run apt it only downloads and unpacks
> *more* packages, all of which, of course, fail as well.

(I'm suprised that I need to include my attachment on this list; I
normally only post it to -user.)

Hardly true. Besides dpkg, if your friend had testing in sources.list in
addition to unstable, he would only need to open aptitude, hit enter on
libc6, and he would get a list like this (this is older unstable and
stable though):

  --\ Versions
  ih  2.3.2.ds1-8   
  
  ph  2.2.5-6

Press + on the desired version. Here it breaks 600 packages, but
presumably it would not with the libc6 in testing. And yes, aptitude
will really downgrade[1] and I'm sure this qualifies as an "interface".
Not the best possible one..

Suggested project: Create a package that, a-l-apt-move, pulls packages
out of the apt cache and creates apt repositories from them. But make it
create a new repository after every upgrade, by hooking into apt. And
auto-add these repositories to sources.list, and remove old ones after a
while, the whole nine yards.

-- 
see shy jo

[1] vis --

[EMAIL PROTECTED]:/home/joey>aptitude
Reading changelogs... Done
dpkg - warning: downgrading pdmenu from 1.2.82 to 1.2.69.
(Reading database ... 163455 files and directories currently installed.)
Preparing to replace pdmenu 1.2.82 (using .../pdmenu/pdmenu_1.2.69_i386.deb) ...
Unpacking replacement pdmenu ...
Setting up pdmenu (1.2.69) ...
Installing new version of config file /etc/pdmenurc ...
Installing new version of config file /etc/menu-methods/pdmenu ...

Press return to continue.

Reading changelogs... Done
(Reading database ... 163456 files and directories currently installed.)
Preparing to replace pdmenu 1.2.69 (using .../pdmenu_1.2.82_i386.deb) ...
Unpacking replacement pdmenu ...
Setting up pdmenu (1.2.82) ...
Installing new version of config file /etc/pdmenurc ...
Installing new version of config file /etc/menu-methods/pdmenu ...

Press return to continue.
Eight reasons why you should be using aptitude instead of apt-get.

1. aptitude can look just like apt-get

   If you run 'aptitude update' or 'aptitude upgrade' or 'aptitude
   install', it looks and works just like apt-get, with a few enhancements.
   So there is no learning curve.

2. aptitude tracks automatically installed packages

   Stop worrying about pruning unused libraries and support packages from
   your system. If you use aptitude to install everything, it will keep
   track of what packages are pulled in by dependencies alone, and remove
   those packages when they are no longer needed.

3. aptitude sanely handles recommends

   A long-standing failure of apt-get has been its lack of support for
   the Recommends relationship. Which is a problem because many packages
   in Debian rely on Recommends to pull in software that the average user
   generally uses with the package. This is a not uncommon cause of
   trouble, even though apt-get recently became able to at least mention
   recommended packages, it's easy to miss its warnings.

   Aptitude supports Recommends by default, and can be confgigured to
   support Suggests too. It even supports installing recommended packages
   when used in command-line mode.

4. use aptitude as a normal user and avoid hosing your system

   Maybe you didn't know that you can run aptitude in gui mode as a regular
   user. Make any changes you'd like to try out. If you get into a real
   mess, you can hit 'q' and exit, your changes will not be saved.
   (aptitude also lets you use ctrl-u to undo changes). Since it's running
   as a normal user, you cannot hose your system until you tell aptitude to
   do something, at which point it will prompt you for your root password.

5. aptitude has a powerful UI and searching capabilities

   Between aptitude's categorical browser and its great support for
   mutt-style filtering and searching of packages by name, description,
   maintainer, dependencies, etc, you should be able to find packages
   faster than ever before using aptitude.

6. aptitude makes it easy to keep track of obsolete software

   If Debian stops distributing a package, apt will leave it on your system
   indefinitly, with no warnings, and no upgrades. Aptitude lists such
   packages in its "Obsolete and Locally Created Packages" section, so you
   can be informed of the problem and do something about it.

7. aptitude has an interface to the Debian task system

   Aptitude lets you use Debian's task system as it was designed to be
   used. You can browse the available tasks, select a task for install, and
   then dig into it and de-select parts of the task that you don't want.
   apt-get has no support for tasks, and aptitude is better even than
   special purpose tools like 

RE: A case study of a new user turned off debian

[Nathanael Nerode]

Julian Mehnle <[EMAIL PROTECTED]> wrote:
>Greg Stark <[EMAIL PROTECTED]> wrote:
>> [...]
>
>First, I think what Daniel Jacobowitz said is entirely true.  Why didn't you 
>start with "testing"?
>
>> All he had to do was install an older version of libc6 and every other
>> package would have been happy. All the infrastructure is there to do
>> this, the old packages are all on the ftp/http sites, the package may
>> even be sitting in apt's cache. But there's no interface for it.
>
>Wrong.  If, on a "unstable" system, Apt sources for "testing" are also listed 
>in /etc/apt/sources.list, you can always do a `apt-get -t testing install 
>libc6` or `apt-get install libc6/testing`.
>
>Or, you could create a file /etc/apt/preferences and pin the "testing" 
>version of the package with a high enough priority.  See `man 
>apt_preferences`.  Then do a `apt-get dist-upgrade`.

It gets better.  ;-)  Look at 'man apt-get', in the 'install' section:

  A specific version of a package can be selected for installation
  by  following the package name with an equals and the version of
  the package to select.  This  will  cause  that  version  to  be
  located  and selected for install.

Find out what version you want, and if it's in the cache, or anywhere in 
sources.list, you can get it.  If necessary pass --force-yes (also 
documented).

If that doesn't work, download the appropriate debs, and do
dpkg --force=downgrade --install foo.deb bar.deb 
(I found that under 'man dpkg'.)

OK, so if you didn't know that 'apt-get' and 'dpkg' are the interesting 
programs, you'd have trouble finding this information.  But surely you knew 
of at least *one* of the two of them?

Re: A case study of a new user turned off debian

[Greg Stark]

Brian May <[EMAIL PROTECTED]> writes:

> On Mon, Nov 03, 2003 at 03:05:56PM -0500, Greg Stark wrote:
> > What started the chain of events was that a fairly routine minor bug bit the
> > latest libc6 release. He's an experienced sysadmin though and wasn't the 
> > least
> 
> What (probably; I am guessing a bit) continued the chain of events:
> ...

All wonderful guesses, but not really at all relevant to what Debian could do
better to handle the situation. All I'm trying to do is look at what I did as
an experienced Debian user and trying to figure out a) why a new Debian user's
instincts were all wrong, b) why the existing tools made the problem worse,
and c) why the tools can't just do what I did or at least make it easier to
reach the right approach.

The main difference between apt's error handling and my own was that I was
aware that I could simply roll back to a version other than the current
"unstable" or "testing". There are many versions in between and rolling back
to "testing" was overkill and would have caused tons more problems than it
would have solved. In the case of someone tracking "testing" there isn't even
any such option, (rolling back to "stable" being laughable).

So all it would take to make the tools handle this would be to somehow make
apt aware of more revisions of packages. They're all in the pool after all.
Short of making some king of humongous mega-Packages file with every revision
of every package -- which apt wouldn't scale up to anyways -- they're
currently unavailable to APT.

The low hanging fruit here would be to have APT keep packages you had
installed yourself in the cache rather than immediately discarding them as
soon as they're upgraded. At a minimum keeping one extra revision would at
least let you roll back. Something more flexible keeping old revisions for n
days after being replaced would be even cooler.

Currently recovering from a package failure means manually downloading a
single .deb and using dpkg to install it, and then tracking down the right
versions of the dependencies for that .deb, and trying to install those, and
... basically reverting to RedHat-style manual dependency resolution; 

If apt kept even a single old revision in its cache then rolling back could be
as simple as 

apt-get install -t previous libc6

or perhaps a little less automatic,

apt-cache show libc6

to list the available revisions then explicitly

apt-get install libc6:2.3.2-8

Actually this wouldn't really have helped my friend at all because he was
unlucky enough that the *first* version of libc6 from unstable that he saw
happened to be the buggy one. That doesn't really happen that often to libc6
so he had particularly bad luck there.

-- 
greg

Re: A case study of a new user turned off debian

[viro]

On Tue, Nov 04, 2003 at 01:48:25PM +1100, Martin Michlmayr wrote:
> * [EMAIL PROTECTED] [2003-11-03 22:52]:
> > but I wouldn't touch Herbert's kernels with a ten-feet pole.
> 
> Can you elaborate why?


a) I can do better
b) I don't do huge monolitic patches
c) I don't like Herbert's taste
d) some boxen are used for 2.6 work

as long as make-kpkg produces valid .deb...

Re: A case study of a new user turned off debian

[Martin Michlmayr]

* [EMAIL PROTECTED] [2003-11-03 22:52]:
> but I wouldn't touch Herbert's kernels with a ten-feet pole.

Can you elaborate why?
-- 
Martin Michlmayr
[EMAIL PROTECTED]

Re: A case study of a new user turned off debian

[Brian May]

On Mon, Nov 03, 2003 at 03:05:56PM -0500, Greg Stark wrote:
> What started the chain of events was that a fairly routine minor bug bit the
> latest libc6 release. He's an experienced sysadmin though and wasn't the least

What (probably; I am guessing a bit) continued the chain of events:

- no prior experience at using Debian.
- using an unstable distribution for the first time.
- no prior experience at fixing broken Linux systems.
- need for a working system important due to new job; sense or urgency
  may have affected decisions made.
- overconfidence.

As well as the ones you mentioned:

- libc6 bug.
- Problems using the debian package management utilities.

In order to eliminate some of these links from this chain, I would have:

- Played around with installing Debian before going to the job.
- Only used stable distribution, at least for first few days of new job.
- Maybe even asked around find out potential problems that could occur
and what to do if if this happened at new job.

There are probably other things he could have done, thats all I can
think of right now.
-- 
Brian May <[EMAIL PROTECTED]>

release cycle Was: [debian-devel] Re: A case study of a new user turned off debian

[Magosányi Árpád]

A levelezőm azt hiszi, hogy Daniel Jacobowitz a következőeket írta:
> 
>  what they get>
> 
> No, really.  This is what stable and testing releases are for.

I fully agree. But...

When I tell it to my friends, some say that stable is way too
old for them. In these cases I used to think about how the
release cycle could be shortened. (I am also thinking about
this thing when we talk about the kernel...)

In theory it would be possible to state only one or two
release goals, announcing feature freeze shortly after
stable is out, and some draconian rules for deep freeze
would help. If one knows that the next window to put new
features in is close, and even if the release goals are
clear for two or more next stable releases, it would be
easier to behave seriously both for the maintainer who
is eager to stick new nifty features in, and the user
who wants to use them.

This is IMHO. And what is your opinion?

-- 
GNU GPL: csak tiszta forrásból

Re: A case study of a new user turned off debian

[Billy Biggs]

Erik Steffl ([EMAIL PROTECTED]):

> [...] Lot of new HW has a better chance to be (better) supported on
> newer system (are new kernels available for stable?)

  Of particular interest to desktop users is XFree86's video card
drivers.

  -Billy

Re: A case study of a new user turned off debian

[viro]

On Mon, Nov 03, 2003 at 02:22:00PM -0800, Erik Steffl wrote:
>   Oh, not this crap again. Or perhaps you're contending that what is 
> usefull for you is usefull for everybody.
> 
>   Hint: there's more to "useful" than old version of software in early 
> stages of development. Lot of desktop oriented apps are changing fast 
> and older version are fairly poor (functionality, stability etc.) - open 
> office, mozilla (and other browsers), kde, gnome etc. Lot of new HW has 
> a better chance to be (better) supported on newer system (are new 
> kernels available for stable?)

man make-kpkg

Debian is *wonderful* for kernel development, but I wouldn't touch Herbert's
kernels with a ten-feet pole.  And yes, new kernels work fine on -stable:
$ grep '^deb ' /etc/apt/sources.list 
deb http://http.us.debian.org/debian woody main contrib non-free
deb http://http.us.debian.org/debian woody-proposed-updates main contrib 
non-free
deb http://non-us.debian.org/debian-non-US woody/non-US main contrib non-free
deb http://non-us.debian.org/debian-non-US woody-proposed-updates/non-US main 
contrib non-free
deb http://security.debian.org woody/updates main contrib non-free
$ uname -a
Linux miles 2.4.23-pre8 #1 Fri Oct 24 02:08:34 EDT 2003 alpha unknown
$

- no need to put -testing on the firewall box.  The rest of Linux boxen
run -testing here...

Re: A case study of a new user turned off debian

[Erik Steffl]

Steve Greenland wrote:
On 03-Nov-03, 14:21 (CST), Erik Steffl <[EMAIL PROTECTED]> wrote: 



Oh, not this crap again. Or perhaps you're contending that I've not
gotten anything done at work in the last two years using my "useless"
Debian stable desktop.
Hint: there's more to "useful" than running the latest version of
everything. Particularly for a sys admin, who I'd expect to be heavily
command line oriented, and who doesn't need the latest 3-d games or dvd
player[1].
Okay, I am running Adrian Bunk's backport of Galeon/Mozilla (thanks,
Adrian). But that's it.
  Oh, not this crap again. Or perhaps you're contending that what is 
usefull for you is usefull for everybody.

  Hint: there's more to "useful" than old version of software in early 
stages of development. Lot of desktop oriented apps are changing fast 
and older version are fairly poor (functionality, stability etc.) - open 
office, mozilla (and other browsers), kde, gnome etc. Lot of new HW has 
a better chance to be (better) supported on newer system (are new 
kernels available for stable?)

erik

Re: A case study of a new user turned off debian

[Tom]

On Mon, Nov 03, 2003 at 04:57:34PM -0500, Greg Stark wrote:
> All I want to do is give up on this new version and go to an earlier version,
> most likely the version I had installed five minutes ago. Downgrading to
> testing would probably require a whole new set of libraries and more work.

I keep snapshots of my system states and debs as I upgrade.  It's not 
perfect, but it's easy to roll back massive fuckups immediately, and at 
least possible to roll back one-or-two packages from a couple weeks ago.
(Really complex changes like the recent gnome upgrade still get 
confusing, though).

Re: A case study of a new user turned off debian

[Steve Greenland]

On 03-Nov-03, 14:21 (CST), Erik Steffl <[EMAIL PROTECTED]> wrote: 
>  is even worse than unstable>

Oh, not this crap again. Or perhaps you're contending that I've not
gotten anything done at work in the last two years using my "useless"
Debian stable desktop.

Hint: there's more to "useful" than running the latest version of
everything. Particularly for a sys admin, who I'd expect to be heavily
command line oriented, and who doesn't need the latest 3-d games or dvd
player[1].

Okay, I am running Adrian Bunk's backport of Galeon/Mozilla (thanks,
Adrian). But that's it.

Steve 


[1] Not on their work machine, at least.

-- 
Steve Greenland
The irony is that Bill Gates claims to be making a stable operating
system and Linus Torvalds claims to be trying to take over the
world.   -- seen on the net

Re: A case study of a new user turned off debian

[Greg Stark]

"Julian Mehnle" <[EMAIL PROTECTED]> writes:

> First, I think what Daniel Jacobowitz said is entirely true. Why didn't you
> start with "testing"?

Sure testing is less likely to trigger this. 

But testing isn't infallible either. And it shouldn't be mean Debian shouldn't
have better error handling. The easier it is for people to manage a Debian
system when things go wrong the better, regardless of how much the chances of
going wrong are minimized.

> > All he had to do was install an older version of libc6 and every other
> >package would have been happy. All the infrastructure is there to do this,
> >the old packages are all on the ftp/http sites, the package may even be
> >sitting in apt's cache. But there's no interface for it.
> 
> Wrong. If, on a "unstable" system, Apt sources for "testing" are also listed
> in /etc/apt/sources.list, you can always do a `apt-get -t testing install
> libc6` or `apt-get install libc6/testing`.

Unfortunately not. apt won't downgrade an already installed package like that.

In any case that doesn't really help. What do you do if testing is 2 months
old as is often the case with things like mozilla. Or if installing the
testing version is exactly what caused the problem? 

All I want to do is give up on this new version and go to an earlier version,
most likely the version I had installed five minutes ago. Downgrading to
testing would probably require a whole new set of libraries and more work.

> Or, you could create a file /etc/apt/preferences and pin the "testing" version
> of the package with a high enough priority. See `man apt_preferences`. Then do
> a `apt-get dist-upgrade`.

That's about the last place I would send a new user. I read that man page
about three times during this crisis before I decided it would be hopeless to
try to explain this procedure online. 

This is what I meant about there not being an "interface". 

If apt said "Hm, version 1.2 of libc failed to configure, would you like to
install the previous version (1.1) from testing and hold back the following
packages that depend on the new one (awk, grep, sed) [Yn]?" That would be an
interface. 

If it didn't do anything but "apt-get -f -t testing libc" did that
automatically without explaining what it was doing, that would be an
interface. 

Telling people, go edit this random file with to set "pin priorities" for
things to arbitrary numbers, find out what package dependencies fail, add
those to your list of pin priorities, etc. That's not a useful interface for
this case.

In any case having the granularity of "stable", "testing", "unstable" really
doesn't help. All the package versions are in the pool. I want to be able to
tell apt to try such and such version, or at least put back the version I had
before and restore whatever other packages it must to satisfy dependencies.

> > The only interface for rolling back is switching the entire machine to an
> >earlier distribution and telling apt to try to downgrade -- which is unlikely
> >to work. And worse, every time you run apt it only downloads and unpacks 
> >*more*
> >packages, all of which, of course, fail as well.
> 
> This is probably one of the worst ways of rolling back few or even a single
> package.

Well it's the only way a new admin has idea about. He was told to put
"stable", "testing", or "unstable" in a spot, that didn't work. So he can try
putting a different word in that spot. He can't magically pull
"apt_preferences" from thin air and decide to go editing a file he's never
heard of. And even if it did, it wouldn't really do what he wanted.

I didn't say it was a good idea or that it was going to work. 
My whole point is that that approach sucks and we should make
something more effective rather than leave the admin stuck.

-- 
greg

Re: A case study of a new user turned off debian

[Marc Haber]

On 03 Nov 2003 15:05:56 -0500, Greg Stark <[EMAIL PROTECTED]> wrote:
>What started the chain of events was that a fairly routine minor bug bit the
>latest libc6 release. He's an experienced sysadmin though and wasn't the least
>bit fazed by that. What drove him batty was that it was so hard to recover
>from the mess and all the obvious avenues just made the problem worse.
>
>All he had to do was install an older version of libc6 and every other package
>would have been happy. All the infrastructure is there to do this, the old
>packages are all on the ftp/http sites, the package may even be sitting in
>apt's cache. But there's no interface for it.

An older libc6 than the one in stable? Where did you find that?

Greetings
Marc

-- 
-- !! No courtesy copies, please !! -
Marc Haber  |   " Questions are the | Mailadresse im Header
Karlsruhe, Germany  | Beginning of Wisdom " | Fon: *49 721 966 32 15
Nordisch by Nature  | Lt. Worf, TNG "Rightful Heir" | Fax: *49 721 966 31 29

Re: A case study of a new user turned off debian

[Arnaud Vandyck]

On 03 Nov 2003 15:05:56 -0500
Greg Stark <[EMAIL PROTECTED]> wrote:

> I finally convinced a sysadmin friend of mine that Debian was the way
> and the light.

Great, you are rigth!

[...]

> What started the chain of events was that a fairly routine minor bug
> bit the latest libc6 release. He's an experienced sysadmin though and
> wasn't the least bit fazed by that. What drove him batty was that it
> was so hard to recover from the mess and all the obvious avenues just
> made the problem worse.

[...]

I did have the problem today! ;)

So first, it's really a bad idea to merge stable, testing and unstable,
especially with libraries! A better thing to do is to backport the
package so it can fit in stable.

The problem with recent libc6 is libdb1-compat. The solution is to
download the old libc6 and install it with:
# dpkg --force-overwrite -i libc6_version_arch.deb
# dpkg --purge libdb1-compat

and then, I did reinstall libc6 (to be sure but I don't know if it's
needed).

Even with some bugs, Debian is the good way ;)

Best regards,

-- 
  .''`.   ** Debian GNU/Linux **
 : :' :  Arnaud   Vandyck
 `. `'   http://people.debian.org/~avdyk/
   `-


pgpb5TsK9rhHb.pgp
Description: PGP signature

Re: A case study of a new user turned off debian

[Erik Steffl]

Daniel Jacobowitz wrote:
On Mon, Nov 03, 2003 at 03:05:56PM -0500, Greg Stark wrote:
...
What would be really neat would be if aptitude or perhaps even apt checked for
earlier versions of the package in the pool and offered them as options if the
current one fails to configure.


No, really.  This is what stable and testing releases are for.


  if Greg's friend was installing a server it might have been a good 
idea to use stable (I definitely wouldn't recommend testing)

erik

RE: A case study of a new user turned off debian

[Julian Mehnle]

Greg Stark <[EMAIL PROTECTED]> wrote:
> [...]

First, I think what Daniel Jacobowitz said is entirely true.  Why didn't you 
start with "testing"?

> All he had to do was install an older version of libc6 and every other
> package would have been happy. All the infrastructure is there to do
> this, the old packages are all on the ftp/http sites, the package may
> even be sitting in apt's cache. But there's no interface for it.

Wrong.  If, on a "unstable" system, Apt sources for "testing" are also listed 
in /etc/apt/sources.list, you can always do a `apt-get -t testing install 
libc6` or `apt-get install libc6/testing`.

Or, you could create a file /etc/apt/preferences and pin the "testing" version 
of the package with a high enough priority.  See `man apt_preferences`.  Then 
do a `apt-get dist-upgrade`.

> The only interface for rolling back is switching the entire machine to
> an earlier distribution and telling apt to try to downgrade -- which is
> unlikely to work. And worse, every time you run apt it only downloads
> and unpacks *more* packages, all of which, of course, fail as well.

This is probably one of the worst ways of rolling back few or even a single 
package.

Re: A case study of a new user turned off debian

[Steve Kemp]

On Mon, Nov 03, 2003 at 03:05:56PM -0500, Greg Stark wrote:

> All he had to do was install an older version of libc6 and every other package
> would have been happy. All the infrastructure is there to do this, the old
> packages are all on the ftp/http sites, the package may even be sitting in
> apt's cache. But there's no interface for it.

  Sure there is, 'dpkg --install $foo.deb'.  Doesn't that do exactly the
 correct thing, even at the expense of downgrading?

  If not I'm sure there's a force option for it.

> The only interface for rolling back is switching the entire machine to an
> earlier distribution and telling apt to try to downgrade -- which is unlikely
> to work. And worse, every time you run apt it only downloads and unpacks
> *more* packages, all of which, of course, fail as well.

  In the general case where you have intermingled dependencies perhaps
 that is necessary.  However for a single package that shouldn't be the
 case.
  I know that you cannot remove an essential package and do much
 afterwards (I once resorted to editting the installed list of packages
 such that libc wasn't marked as being installed.  I had a lot of fun
 before getting there and afterwards was even worse).

  However if you have an alternative package you should be able to force
 its installation; even if you have to do someting tedious like download
 a source package and rebuild it ..

Steve
---
# Debian Security Audit Project
http://www.steve.org.uk/Debian/


pgpVNIHdg5B8I.pgp
Description: PGP signature

Re: A case study of a new user turned off debian

[Daniel Jacobowitz]

On Mon, Nov 03, 2003 at 03:05:56PM -0500, Greg Stark wrote:
> 
> I finally convinced a sysadmin friend of mine that Debian was the way and the
> light. He started a new job and showed up on his first day to set up his
> machine by installing Debian. In short, things went horribly wrong and he
> started this new job by wasting two days picking up the pieces. He's now very
> leery of suggesting using Debian on other machines at work or of using it
> himself at home.
> 
> What started the chain of events was that a fairly routine minor bug bit the
> latest libc6 release. He's an experienced sysadmin though and wasn't the least
> bit fazed by that. What drove him batty was that it was so hard to recover
> from the mess and all the obvious avenues just made the problem worse.
> 
> All he had to do was install an older version of libc6 and every other package
> would have been happy. All the infrastructure is there to do this, the old
> packages are all on the ftp/http sites, the package may even be sitting in
> apt's cache. But there's no interface for it.
> 
> The only interface for rolling back is switching the entire machine to an
> earlier distribution and telling apt to try to downgrade -- which is unlikely
> to work. And worse, every time you run apt it only downloads and unpacks
> *more* packages, all of which, of course, fail as well.
> 
> What would be really neat would be if aptitude or perhaps even apt checked for
> earlier versions of the package in the pool and offered them as options if the
> current one fails to configure.



No, really.  This is what stable and testing releases are for.

-- 
Daniel Jacobowitz
MontaVista Software Debian GNU/Linux Developer

A case study of a new user turned off debian

[Greg Stark]

I finally convinced a sysadmin friend of mine that Debian was the way and the
light. He started a new job and showed up on his first day to set up his
machine by installing Debian. In short, things went horribly wrong and he
started this new job by wasting two days picking up the pieces. He's now very
leery of suggesting using Debian on other machines at work or of using it
himself at home.

What started the chain of events was that a fairly routine minor bug bit the
latest libc6 release. He's an experienced sysadmin though and wasn't the least
bit fazed by that. What drove him batty was that it was so hard to recover
from the mess and all the obvious avenues just made the problem worse.

All he had to do was install an older version of libc6 and every other package
would have been happy. All the infrastructure is there to do this, the old
packages are all on the ftp/http sites, the package may even be sitting in
apt's cache. But there's no interface for it.

The only interface for rolling back is switching the entire machine to an
earlier distribution and telling apt to try to downgrade -- which is unlikely
to work. And worse, every time you run apt it only downloads and unpacks
*more* packages, all of which, of course, fail as well.

What would be really neat would be if aptitude or perhaps even apt checked for
earlier versions of the package in the pool and offered them as options if the
current one fails to configure.

-- 
greg