[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: My first d-i installation of a mips.

"Bernhard R. Link" <brl@pcpool00.mathematik.uni-freiburg.de> writes:

> * Goswin von Brederlow <brederlo@informatik.uni-tuebingen.de> [031006 16:36]:
> > > [/sbin/init vs /linuxrc]
> > Patch pending a successfull build and boot on i386 and alpha. I don't
> > want to break anything.
> Ok, thanks. Then I can just ignore that when I have some more time to
> look into it.
> > > In the shell (that is on tty2) when beeing in the last line
> > > and pressing tab or cursor up, the whole screen is blanked. 
> > > (The content gets restored when switching ttys) This might be 
> > > a problem with the kernel's framebuffer-console, though the sh 
> > > after chrooting in a partly installed base-system does not 
> > > have this problem.
> > 
> > From what I was told by the mips guys at the D-I Debcamp in oldenburg
> > its the framebuffer.
> As I fear the kernel would be hard to fix, do you think is is appropiate
> to file a wishlist-bug against busybox for a workaround for this?
> > > network-configuration is utterly strange
> > > (It first asks for a IP, then waits long time, then
> > > again asks for a IP and some more information. Only after
> > > selecting configure a static network twice it has asked
> > > for all information and still insists on "something has gone wrong".)
> > > When I start with DEBCONF_PRIORITY=low I only have to select
> > > netcfg-static once and everything works...
> > 
> > If something goes wrong the priority is lowered and you can try again
> > with more options.
> Ok, that explains the behaviour. Now my confusion is limited to the
> question why anyone might think an IP-address without the dns and/or 
> gateway address is of any value.

Given an IP, say I would guess the gateway to be, then netmask to be /24 and the DNS to be too.

DNS being the wildest guess here. I would prefer a multiline formular
there. One field per question and while you change a field the other,
not yet edited fields, would change according tothe guessing

So normaly one would just enter in the IP field and skip
ahead to "continue".

Any volunteres to add a new widget to (c)debconf?
> > > The installers userfriendlyness is still missing worlds compared
> > > to the the old b-fs. One often does not know, where one currently
> > > is, but is unexpectently asked questions so one can only guess what it
> > > currently plans to do. (I hope this is not intentional). I also
> > > hope things like availability of a "Install kernel" item when this
> > > only triggers the already failed "install basesystem" will
> > > go away.
> > 
> > Items will trigger all items they depend on on. Thats actually a nice
> > feature.
> > 
> > You boot up and select "install kernel" and it ill automatically go
> > through all the steps needed till it can install a kernel.
> While I can agree that this is a feature, I'm not conviced it is nice.
> Perhaps the permanent failure[1] of netcfg-static confused the menu, but 
> it looked as if "Install the base system" and "Install kernel" were
> both qualiy availabe actions. If both actions start the same way, this
> violates the principle of least surprise. 
> Or to say in fewer words: I think this might be convient for installing
> but not friendly to new users.

Maybe it could be limited to items with no menu and not yet executable
items be drawn in grey.

But then people will wonder why something is grey and how to satisfy
its dependencies. Of cause being grey could just be a visible clue
saying it will do more than just this. A "+" mark in front of all
items being selected by an option could also be used. But again, who
will programm the widget for that?

> > That also means you can't skip a step that half worked but said it
> > failed. With boot-floppies you could ignore steps.
> Ah, that is something to put in the installation manuals, once they are
> written: "How do I ingore steps." Is there any easier way than editing
> some status file with an uncommon editor?
> Hochachtungsvoll,
> 	Bernhard R. Link
> [1]: While it finaly configured the network properly, it never
> exited without error, as it stumbled over doing things twice.

That should never happen :) Bad, bad, bad. Happy bug hunting.


Reply to: