Bug#379526: papersize is a4 for US install - should be letter

[Rick Thomas]

On Jul 24, 2006, at 4:14 PM, Christian Perrier wrote:


plan is to also support en_US.


I assume this means that if I do whatever magic is required to get
en_US locale, then I'll get "letter" in /etc/papersize.  What if I
choose "C" locale?



This should be discussed with the localization-config maintainer but
that would require extra code in localization-config to not only take
the locale into account but also the chosen "country".

However my personal feeling about this is that locales are exactly
meant for that purpose: set parameters that depend on language AND
country.

So, if you actually want a paper size adapted to your country, I'd
recommend to set the locale properly.


Far be it from me to make extra work for the localization-config  
maintainer.  And I agree with you that locales are exactly for the  
purpose of setting things that depend on country and language.  I  
guess my problem is that (for the intended application of this  
system) I want a locale that says "language" is "C" and "country" is  
"US" -- but there is no such locale.



Rick

-- If wishes were horses, then beggars would ride!
-- anonymous English proverb

-- If a frog had wings, he'd have a long tail, 'cause he wouldn't rub  
it off bumping his backside on the ground all the time!

-- me


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

Bug#379526: papersize is a4 for US install - should be letter

[Rick Thomas]

On Tue, 2006-07-25 at 06:36 +0200, Christian Perrier wrote:
> > Far be it from me to make extra work for the localization-config  
> > maintainer.  And I agree with you that locales are exactly for the  
> > purpose of setting things that depend on country and language.  I  
> > guess my problem is that (for the intended application of this  
> > system) I want a locale that says "language" is "C" and "country" is  
> > "US" -- but there is no such locale.
> 
> 
> en_US seems to be the closest approximation...
> 
> Apart from the extra cruft bringed by the locales packages you would
> get by choosing English then United States, I actually would recommend
> setting the locale to en_US if you want some correct
> internationalization.
> 
> 
> 
> I actually recommend closing this bug report.

Do what you want, but keep in mind that en_US, and every other natural
language locale except "C", messes with the sort collating order to
satisfy some librarian's idea of 'niceness'.   It robs me of the
simplicy and reliability of the "natural" collating order that I get
from plain old 8-bit ASCII characters.

Oh well... It's just one more thing to go on my list of stuff I have to
fix after the Debian installer gets done.

Sigh,

Rick



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

Bug#379120: lspci present on i386, verify on powerpc

[Rick Thomas]

I booted my test beige G3 (oldworld) from the miboot floppies with  
the July 30'th businesscard CD in the CDrom drive.  I was also able  
to do 'lspci' at keyboard selection time.


Rick

On Jul 31, 2006, at 6:04 AM, Eddy Petrişor wrote:


On 29/07/06, Geert Stappers <[EMAIL PROTECTED]> wrote:


On a i386 is lspci present in the recent builds op debian-installer.

While the installer waits for a keyboard selection,
I switched to the second console and typed `lspci`
and got the output of lspci.

Could this be verified on a powerpc computer?
(no need to do a re-installed ( no computer harm ))


The gtk-miniiso (downloaded a week ago, IIRC) contained it, just  
cjhecked.


--
Regards,
EddyP
=
"Imagination is more important than knowledge" A.Einstein


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

Bug#258907: This bug can be closed. It's no longer relevant.

[Rick Thomas]

Thanks!

Rick


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

Bug#262201: Please close this bug. It's no longer relevant.

[Rick Thomas]

Please close this bug.  It's no longer relevant.

I believe it has been fixed in the released "sarge" installer.

Thanks!

Rick



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

Bug#262198: Please close this bug. It's no longer relevant.

[Rick Thomas]

Please close this bug.  It's no longer relevant.

Thanks!

Rick


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

Bug#380718: lspci command not present after reboot in non-Desktop configuration

[Rick Thomas]

Package: installation-reports

Note: This is not the same bug as #379120.

This bug report is about the condition after the reboot.  It's not  
about the time during installation -- before the reboot.


It seems that lspci is pulled in by the "Desktop" task.  It's absent  
after the reboot in a "bare-bones" ("Desktop" un-checked) install.


Leaving "Desktop" unchecked in tasksel is something that one might  
reasonably want to use for a simple server.  However, a simple server  
is a classic case where you might actually want "lspci" and friends  
to be present...


So to put the problem as simply as possible:

The command "lspci" is present during the installation (see bug  
#379120) before the reboot.  But after the reboot, whether lspci is  
present or absent depends on whether the user left the "Desktop  
environment" checked when the installer was running tasksel.  If it  
was checked, then lspci will be present in the installed system after  
the reboot.  If it was not checked, then lspci will be absent after  
the reboot.


It seems to me that lspci is such a basic tool that it ought to be  
present in the bare-bones system, not dragged in as an afterthought  
by the Desktop task.




Enjoy!

Rick


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

Bug#382070: Long pauses during configuring portmap and configuring printconf

[Rick Thomas]

Package: installation-reports

I recently did a businesscard install of the etch daily.

There were two unexpectedly long pauses in the later part of the  
installation.   I've seen these before, so they aren't just a fluke.   
So I'm reporting them now.


The first was during "Configuring portmap".  Looking on the VT02  
console, "ps" showed that it was doing "rpcinfo -p".  As an  
experiment, I did a chroot to /target myself and tried "rpcinfo -p".   
It hung, though a ctl-C killed it.


The second was during "Configuring printconf".  Again, looking on the  
VT02 console with "ps", it seemed to be doing "/usr/bin/lpstat -l -d - 
p -v" for a long time -- then later it was doing "/usr/bin/lpstat -d"  
for an equally long time.


Normally these operations take a very short time.  Is there something  
in the configuration process that is being done out-of-order, so that  
a vital system component isn't available yet?


Logs are available on request...

Rick



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

Bug#382129: daily wont boot on OldWorld PowerPC Mac

[Rick Thomas]

Package: installation-reports

kernel and initrd from latest daily refuses to boot on OldWorld beige  
G3 Mac


Hardware used is
PowerMac G3 (Gossamer)
this is an OldWorld powerpc Macintosh

Boot loader used is
MacOS-9.2 with the BootX extension.

Kernel version used is
2.6.16-2-powerpc (Debian 2.6.16-17) ([EMAIL PROTECTED])...
taken directly from the businesscard daily install CD

The boot process gets only a small ways in and freezes

The screen contains about half a screen worth of kernel startup  
messages starting with



Total Memory = 384MB; using 1024kB for hash table (at cff0)
Linux version 2.6.16-2-powerpc (Debian 2.6.16-17)  
([EMAIL PROTECTED]) (gcc version...

#1 Sat Jul 15 17:39:41 CEST 2006


and so on... ending with


GMT Delta read from XPRAM: 60 minutes, DST: on
time_init: decrementer frequency = 16.708016 MHz
time_init: processor frequency   = 200.69 MHz

MMU:exit



then it hangs.

I can provide more details if necessary, but obviously, there are no  
log files.



Rick


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

Bug#382129: daily wont boot on OldWorld PowerPC Mac

[Rick Thomas]

I should have added that sarge boots and runs just fine on this  
machine with BootX.



On Aug 9, 2006, at 2:05 AM, Rick Thomas wrote:


Package: installation-reports

kernel and initrd from latest daily refuses to boot on OldWorld  
beige G3 Mac


Hardware used is
PowerMac G3 (Gossamer)
this is an OldWorld powerpc Macintosh

Boot loader used is
MacOS-9.2 with the BootX extension.

Kernel version used is
2.6.16-2-powerpc (Debian 2.6.16-17) ([EMAIL PROTECTED])...
taken directly from the businesscard daily install CD

The boot process gets only a small ways in and freezes

The screen contains about half a screen worth of kernel startup  
messages starting with



Total Memory = 384MB; using 1024kB for hash table (at cff0)
Linux version 2.6.16-2-powerpc (Debian 2.6.16-17)  
([EMAIL PROTECTED]) (gcc version...

#1 Sat Jul 15 17:39:41 CEST 2006


and so on... ending with


GMT Delta read from XPRAM: 60 minutes, DST: on
time_init: decrementer frequency = 16.708016 MHz
time_init: processor frequency   = 200.69 MHz

MMU:exit



then it hangs.

I can provide more details if necessary, but obviously, there are  
no log files.



Rick


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






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

Re: Switch to 2.6.17 kernel udebs (was: D-I Beta 3 - release today)

[Rick Thomas]

Will this fix the problem of the kernel not booting on OldWorld  
PowerPC Macintosh?



On Aug 11, 2006, at 5:12 AM, Frans Pop wrote:


On Thursday 10 August 2006 12:30, Frans Pop wrote:

- switch to 2.6.17 kernel udebs (long term 2.6.16 maintenance no
  longer seems to be an option for the kernel team)


The kernel team is planning a new upload, possibly today, that will
include an ABI bump.
It probably makes sense to wait for that before we start on kernel  
udebs.


Cheers,
FJP

P.S. The release was delayed a bit by a very late mirror sync. Release
announcement later today.



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

Bug#382129: Beta3 won't boot on OldWorld PowerPC Mac

[Rick Thomas]

I just tried the etch Beta3 kernel on my Beige G3 OldWorld PowerPC  
Macintosh.


Same results as noted in the original bugreport by this number.  It  
hangs almost immediately.  The picture of Tux never appears at the  
top of the screen.  See the original bug report for details.


Seriously,

Has *anybody* had any success getting the "Linux version 2.6.16-2- 
powerpc (Debian 2.6.16-17)" kernel to boot an OldWorld machine with  
BootX?  Or *any* kernel after 2.6.15?


What has changed between the Sarge kernel (which boots just fine) and  
this one that would have such a catastrophic effect?




Enjoy!

Rick



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

Bug#382129: Beta3 won't boot on OldWorld PowerPC Mac

[Rick Thomas]

On Aug 18, 2006, at 7:59 AM, Aurélien GÉRÔME wrote:


Hi,

On Sun, Aug 13, 2006 at 03:19:18AM -0400, Rick Thomas wrote:

Has *anybody* had any success getting the "Linux version 2.6.16-2-
powerpc (Debian 2.6.16-17)" kernel to boot an OldWorld machine with
BootX?  Or *any* kernel after 2.6.15?


I think the anwser is plainly no. :(


Perhaps the Debian PTBs should simply admit that PowerPC OldWorld  
Macintosh is an unsupported sub-arch for Debian etch?   /-8




Well, you would be better off using miBoot anyway. I think I will
produce miBoot floppies for the last Debian kernel for you to try...


One thing I've discovered:  The miboot floppy at http:// 
people.debian.org/~wouter/d-i/powerpc-miboot/daily/powerpc/floppy/  
contain the "2.6.16-2-powerpc-miboot (Debian 2.6.16-7)" kernel, which  
is the same base version (different config, presumably) as the kernel  
on the beta3 CDs.  And that kernel (the one on the miboot floppy)  
boots just fine(*) on my beige G3.  So the problem seems to be a  
configuration issue, not anything fundamental.


(*) Well... almost "just fine".  There is a small problem with the  
video parameters -- some columns of the text on the screen seem to  
flicker.  If I could figure out how to modify the kernel command line  
arguments (specifically the video options) on the floppy, I might be  
able to fix that.  Interestingly, "works just fine" only applies to  
the "boot" floppy.  The "ofonlyboot" floppy does not work -- it gets  
the same symptoms as I described in Bug#369760, and as someone else  
has described in Bug#380187.






What has changed between the Sarge kernel (which boots just fine) and
this one that would have such a catastrophic effect?


Maybe the image is too heavy, some kernel defines changed, or BootX
cannot find the sections of the image anymore? We cannot fiddle with
BootX source code to see what happens, but we will be able with miBoot
source code when it will be ready. :)


Currently, miboot (leaving aside issues of "freeness") only works for  
floppies?  Do you think there is any chance of it's working for  
booting off of CDs or hard-disks, once we can fiddle with the source  
code?




Cheers,
--
 .''`.   Aurélien GÉRÔME
: :'  :
`. `'`   Free Software Developer
  `- Unix Sys & Net Admin

Bug#382129: Beta3 won't boot on OldWorld PowerPC Mac

[Rick Thomas]

On Aug 18, 2006, at 2:31 PM, Aurélien GÉRÔME wrote:


What is PTBs? :)


Powers that be...

Bug#382129: Beta3 won't boot on OldWorld PowerPC Mac

[Rick Thomas]

On Aug 18, 2006, at 2:31 PM, Aurélien GÉRÔME wrote:





Wow!  great!  I'm really glad somebody's working on that.

Do you think it will be ready for inclusion in etch at release time?

Rick

Bug#382129: Beta3 won't boot on OldWorld PowerPC Mac

[Rick Thomas]

On Aug 18, 2006, at 2:31 PM, Aurélien GÉRÔME wrote:


Well, aside the issue of booting, the kernel runs just
fine and so does the user space... As you can look at
, I think
it is partially admitted in "Oldworld powerpc boot floppies will not
work as miboot is not included".


I think that is just referring to the fact that miboot is not (yet)  
"free" so it can't be included in an official debian distribution.   
As noted, wouter provides floppy images with a working miboot and a  
working kernel (almost -- modulo the video problems I've mentioned.)


I don't think the errata takes any notice of the fact that, quite  
independent of "freeness" considerations, there is *no* working boot  
loader for the 2.6.16 kernel on an OldWorld Mac.  I've tried both  
bootx and quik and neither of them work on my beige G3.  In it's  
current state, even if it were free, miboot would not be a workable  
solution because it only works from floppies.


Rick

Bug#382129: Beta3 won't boot on OldWorld PowerPC Mac

[Rick Thomas]

On Aug 18, 2006, at 5:58 PM, Aurélien GÉRÔME wrote:


On Fri, Aug 18, 2006 at 02:52:24PM -0400, Rick Thomas wrote:

On Aug 18, 2006, at 2:31 PM, Aurélien GÉRÔME wrote:

What is PTBs? :)

Powers that be...


On Fri, Aug 18, 2006 at 11:44:23AM -0400, Rick Thomas wrote:

Perhaps the Debian PTBs should simply admit that PowerPC OldWorld
Macintosh is an unsupported sub-arch for Debian etch?   /-8


Okay, in that case, I strongly disagree. That would be like saying
the efforts to get it working until now are worthless. The individual
behind debian-installer has already kicked out Sven without any
concern for the PowerPC port well being.

Killing the OldWorld subarch in the PowerPC port would simply get  
Piotr

(I think) and I (surely) not caring anymore about miBoot... After
all, I know how to proceed with it the hard way, so the installation
easiness does not matter to me. It is only more fun for me for it to
become easy. :)

I sincerely hope I will succeed getting an access to the
debian-installer SVN repository to work on miboot targets and on
miboot-installer...


I have to agree that your work on miboot is the only real light I can  
see in the darkness surrounding Debian support of OldWorld  
PowerMacs.  I sincerely wish you luck in getting the resources  
(physical and political) that you need to finish the job.  If there  
is anything I can do to help with the process, please do not hesitate  
to ask.



Rick

Bug#382129: Beta3 won't boot on OldWorld PowerPC Mac

[Rick Thomas]

On Aug 18, 2006, at 6:43 PM, Sven Luther wrote:


On Sat, Aug 19, 2006 at 12:18:23AM +0200, Sven Luther wrote:

On Fri, Aug 18, 2006 at 11:58:07PM +0200, Aurélien GÉRÔME wrote:

On Fri, Aug 18, 2006 at 02:52:24PM -0400, Rick Thomas wrote:

On Aug 18, 2006, at 2:31 PM, Aurélien GÉRÔME wrote:

What is PTBs? :)

Powers that be...


On Fri, Aug 18, 2006 at 11:44:23AM -0400, Rick Thomas wrote:

Perhaps the Debian PTBs should simply admit that PowerPC OldWorld
Macintosh is an unsupported sub-arch for Debian etch?   /-8


Okay, in that case, I strongly disagree. That would be like saying
the efforts to get it working until now are worthless. The  
individual

behind debian-installer has already kicked out Sven without any
concern for the PowerPC port well being.


There is already an alternative installer planed once etch is  
released, which
will contain fixes for all those issues the d-i team left dying on  
the

road-side.


all those powerpc issues.


Thanks!  If there is anything I can do to help, please let me know.

Rick

Re: Bug#352610: Please create a udeb for ntpdate

[Rick Thomas]

On Aug 19, 2006, at 2:52 PM, Steve Langasek wrote:


On Sat, Aug 19, 2006 at 04:47:44PM +0200, Frans Pop wrote:


We mainly need to determine how we are going to use rdate:
- for all installations;
- only for some (sub)arches like nlsu;
- only if difference between system date/time and rdate date/time is
  greater than x.


Seems to me there needs to be a link with clock-setup as I guess  
rdate
would set the time to UTC. So either clock-setup would need to  
reset the

clock if "local time" is selected, or it would have to default to UTC
too.
We could even use rdate to guess if clock is set to local time or UTC
based on the difference between system time and rdate.


For countries with more than 1 timezone, I want to be able to set a  
default
timezone for the system based on the offset between the system  
clock and the

remote timeserver as well... :)


It should be noted that the dhcp protocol has an option to provide  
the local offset from UTC, though it does not (currently) have a way  
of communicating things like Daylight Saving Time and so on.


There is a draft proposal in the IETF (http://www.ietf.org/internet- 
drafts/draft-ietf-dhc-timezone-option-03.txt) to add full time-zone  
info (as an option) to dhcp.  Full RFC status is probably many months  
(years?) away.  Implementation in production dchp servers will most  
likely wait for that.


Rick


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

Re: Bug#352610: Please create a udeb for ntpdate

[Rick Thomas]

On Aug 19, 2006, at 6:59 PM, Steve Langasek wrote:


Using DHCP will only tell you what the
DHCP admin's preferred timezone setting is, which won't necessarily  
match
and also doesn't give you any idea whether the user wants the  
system's clock

to be set in UTC or not.


The proposed RFC allows the dhcp administrator to tailor the response  
based on the MAC address of the client.  Most won't, but it should be  
possible if you want to.


Rick


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

Re: Bug#352610: Please create a udeb for ntpdate

[Rick Thomas]

On Aug 19, 2006, at 2:52 PM, Steve Langasek wrote:


On Sat, Aug 19, 2006 at 04:47:44PM +0200, Frans Pop wrote:


We mainly need to determine how we are going to use rdate:
- for all installations;
- only for some (sub)arches like nlsu;
- only if difference between system date/time and rdate date/time is
  greater than x.


Seems to me there needs to be a link with clock-setup as I guess  
rdate
would set the time to UTC. So either clock-setup would need to  
reset the

clock if "local time" is selected, or it would have to default to UTC
too.
We could even use rdate to guess if clock is set to local time or UTC
based on the difference between system time and rdate.


For countries with more than 1 timezone, I want to be able to set a  
default
timezone for the system based on the offset between the system  
clock and the

remote timeserver as well... :)


RFC 868 (http://www.rfc-archive.org/getrfc.php?rfc=868) says that the  
rdate protocol delivers time as a 32-bit binary integer in units of  
seconds since midnight on January first 1900 GMT (time 1 is 12:00:01  
am on 1 January 1900 GMT).


The rdate command prints time in the prevailing timezone (as  
specified by the TZ environment variable, to to print the date and  
time in UTC, do "TZ=UTC rdate -p").


There is no option for the rdate command to print anything but a  
formated date/time.  This means that the output will have to be  
parsed back into a binary integer before you do the calculations  
needed to deduce the time zone.  This is, of course, possible.  But  
shell script wouldn't be my language of choice for doing it.   
Regardless of the language, getting the fiddlly bits just right (leap  
years leap seconds) just right is tricky business.  It's best to use  
a pre-existing library to do the hard part.


Do you have access to perl or python at the time you want to do  
this?  Do you have access to the date/time libraries for either of  
those languages?  Would it be better to write a simple, one-purpose,  
C program to do what you want?


Rick


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

Re: Bug#352610: Please create a udeb for ntpdate

[Rick Thomas]

On Aug 20, 2006, at 12:05 AM, Steve Langasek wrote:




The proposed RFC allows the dhcp administrator to tailor the response
based on the MAC address of the client.  Most won't, but it should be
possible if you want to.


I think you're missing the point that the maintainer of the newly- 
installed
Debian system may *disagree* with the TZ information provided by  
the dhcp

admin.


No, that's exactly what I meant.  The dhcp administrator may not  
care, or may have wrong-headed ideas about this (or any) bit of  
data.  The client must be able to tailor for her own particular needs.


Rick


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

Bug#382129: Beta3 won't boot on OldWorld PowerPC Mac

[Rick Thomas]

Well... That's interesting news.  And I'm glad you're having success  
with your system.


But, unfortunately, the bottom line for this bug report is still that  
nobody has yet succeeded in getting the 2.6.16 kernel to boot under  
BootX.


Keep in touch.  I'll be particularly interested in hearing of your  
experiences upgrading to Dapper.


Enjoy!

Rick


On Aug 20, 2006, at 3:25 AM, Harold Johnson wrote:

Oh, and since I forgot to mention it, the kernel version I  
currently (at this moment) have installed is 2.6.12-9; that's after  
simply installing Breezy.  Once I upgrade to Dapper again, I'll be  
using whatever version that is -- or I'll jump to Debian and do the  
same.  I'll try to remember to post that version number here so  
that you'll know if I got up to 2.6.16 using my method.  All I know  
for certain is that I've been able to get to Dapper using this  
installation method in the past.


Harold

On 8/19/06, Harold Johnson <[EMAIL PROTECTED]> wrote:
You have great timing; I wouldn't have had that answer for you  
earlier today, because I wasn't certain which kernel version I had  
installed.  I am writing down the entire process of setting up a  
triple-booting system, which will be running Linux, OS X, and OS  
9.  (This has been done many times before, I'm sure, but not by  
me!)  Thus far, I've installed OS 9, Ubuntu Linux, and I'm  
currently installing OS X (Panther) on the third (and final)  
partition.


I'll be adding to some online documentation, but hopefully the  
following info. will fill in some of the gaps:


-- 2 Mac OS (hfs+) partitions and one unallocated space using an OS  
9 installation disk.  The first partition will be for OS X, the  
second is for OS 9, and the free space is for Linux. 

-- After installing OS 9 and BootX, installed Ubuntu Server  
(Breezy).  This info. is probably the most helpful for you; I've  
tried on multiple occasions to install the latest flavors of Debian  
using BootX, to no avail.  The same goes with the latest Ubuntu  
versions (Dapper); only Breezy installs on my OldWorld PowerBook.   
Once I've installed Breezy, it's easy enough to upgrade to Dapper  
by simply editing the repositories (replacing all instances of  
"Breezy" with "Dapper") and then using apt-get to update the  
system.  I imagine it's possible to use this same technique to  
update to the latest version of Debian; wouldn't you think? 


Thanks for sending along your script,

Harold


On 8/19/06, Rick Thomas <[EMAIL PROTECTED]> wrote:
On Aug 19, 2006, at 6:29 PM, Harold Johnson wrote:

> In the meantime, I can continue using BootX -- not a real elegant
> solution, IMHO, but it works.

Hi Harold,

What magic did you have to use to get BootX to boot a 2.6.16 kernel
for you?

Thanks!

Rick






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

Bug#380105: Show current hour in hardware clock question

[Rick Thomas]

On Aug 25, 2006, at 10:57 AM, Christian Perrier wrote:

I am not sure, but in the graphical installer, we could add a  
clock widget
somewhere from the start, and do clock setting pretty early one  
(we probably
only need hwclock and a little menu thingy), it can even be done  
before

base-install and partman, since there are no extra dependencies.


There has been such kind of suggestion last weeks but it has been
ruled out. I guess that the rationale is mostly avoiding features that
are only available in some D-I flavours.

So, if a method to set the clock is offerred, it has to work for all
interfaces.


In some cases it's not possible (e.g. no hardware clock.)  In some  
cases it's possible but not desirable (e.g. hardware clock exists but  
is known to be wrong and porters have decided to ignore it during  
system initialization until access to an external clock is available.)


But it seems a shame to ban a useful tool just because some people  
can't or won't use it...


My two cents worth,

Rick



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

Is something wrong with CDIMAGE.debian.org daily builds?

[Rick Thomas]

The "latest" builds for all architectures are from Dec 15th.

Is there a problem with the daily build process?

Rick



Index of /cdimage/daily-builds/daily/arch-latest
Name Last modified  Size

Parent Directory  -
alpha/   15-Dec-2006 22:08-
amd64/   15-Dec-2006 22:11-
arm/ 15-Dec-2006 22:14-
hppa/15-Dec-2006 22:17-
i386/15-Dec-2006 22:21-
ia64/15-Dec-2006 22:24-
m68k/15-Dec-2006 22:27-
mips/15-Dec-2006 22:30-
mipsel/  15-Dec-2006 22:33-
powerpc/ 15-Dec-2006 22:37-
sparc/   15-Dec-2006 22:40-

Apache/2.2.3 (Unix) Server at cdimage.debian.org Port 80



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

Re: miboot-floppies (powerpc oldworld) for sarge

[Rick Thomas]

On Dec 19, 2006, at 9:54 AM, Holger Levsen wrote:


Hi,

On Saturday 16 December 2006 00:52, Sven Luther wrote:
But why? People can install with 2.4 just fine and then later  
upgrade to

2.6, so why do the work and backport it?
Because even when using 2.4 floppies, they will install the 2.6.8  
kernels,
so it is best to have the same kernel for installation media and  
reboot

kernels.


As I don't know of any hardware that a.) needs to boot from floppy  
and b.)
doesnt work with 2.4 but only with 2.6 I still don't see a good  
reason to do

the work and do a rebuild of the _sarge_ miboot-floppies.

Since almost 10 years (1997 I think) apple switched to newworld,  
which all can
boot from cdrom or have proper OF implementations to boot from  
network - so I

also doubt there is a hardware which needs this and I don't know :)


There are plenty of OldWorld Macs out there (I own 6 of them) that  
would benefit from a kernel 2.6 bootable Sarge floppy-set.


It's a one-time effort, not a continuing commitment.





The change is trivial, please try it :


At the moment I only have/had to run "debuild", everything else,  
takes some of

my time which I rather invest in improving etch and lenny.


It's a one-time effort, not a continuing commitment.  And many people  
will benefit.


Please!



Sorry for not using your patch, but you can always rebuild the  
miboot floppies

yourself ;-)



Sincerely,

Rick


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

Re: miboot-floppies (powerpc oldworld) for sarge

[Rick Thomas]

On Dec 19, 2006, at 10:09 PM, John Schmidt wrote:


On Tuesday 19 December 2006 07:54, Holger Levsen wrote:

Hi,

On Saturday 16 December 2006 00:52, Sven Luther wrote:
But why? People can install with 2.4 just fine and then later  
upgrade

to 2.6, so why do the work and backport it?


Because even when using 2.4 floppies, they will install the 2.6.8
kernels, so it is best to have the same kernel for installation  
media and

reboot kernels.


As I don't know of any hardware that a.) needs to boot from floppy  
and b.)
doesnt work with 2.4 but only with 2.6 I still don't see a good  
reason to

do the work and do a rebuild of the _sarge_ miboot-floppies.


I have an oldworld ppc that can't boot from cdrom that would  
benefit from

sarge miboot floppies.


Actually, you (and I) *can* boot our oldworld PowerMacs from the 2.4  
miboot floppyset.  And we can install Sarge from there.  We can even  
use it to install a 2.6.8 kernel.


Not to put words into his mouth, but I think that is what Holger is  
saying.  For the purpose the Sarge floppyset serves, it's enough to  
have just a 2.4 kernel for the installer.


On the other hand, what Sven is saying seems (again, not to put words  
in his mouth) to be that, as long as we're going to encourage the  
user to install a 2.6 kernel eventually, it would be cleaner/nicer/ 
more-elegant to start them off with a 2.6 kernel from the beginning  
in the installer.


Holger seems to reply "Why bother?"

My thought (admittedly a bit of a stretch) is that there *may* be  
hardware out there (SATA disks?) that a user can plug into an  
OldWorld PowerMac PCI slot that is not supported by the 2.4 kernel,  
but is supported by the 2.6 kernel, and why should we artificially  
limit our user base to exclude such users?


So I mostly agree with Sven on this one, but I don't think it's worth  
a lot of emotion.  There are plenty of bigger issues regarding  
OldWorld PowerMacs that need looking into.  Until recently, just  
getting Etch to boot *at-all* on OldWorld was one of them!



Rick


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

Re: miboot-floppies (powerpc oldworld) for sarge

[Rick Thomas]

Hi Sven,


On Dec 20, 2006, at 2:16 AM, Sven Luther wrote:

What is really needed is for confirmation with the current kernels  
that :


  1) miboot booting works (or not).

  2) bootx booting works (or not).

  3) quik booting works (or not).

Rick, you have been rather active in this, could i ask you to make  
a survey
with the current 2.6.18 kernels in both etch and testing, and tell  
us what
works and what doesn't. And file a bug report against the kernel  
packages

(linux-2.6, title prefaced [powerpc,oldworld]).


Items 1 and 2 are on my list.  I hope to get to them before Friday.   
Item 3 (quik) is more problematical.  I also run MacOS-9 on my  
OldWorld test machine, and quikis incompatible with that.


I'll look thru the junk box and see if I can find parts to cobble  
together a "quik" test box, but I'm not guaranteeing anything, and I  
don't know how long it will take if I do succeed.


To make sure I have it right:   Bug reports (or success reports!)  
should have

Package: linux-2.6
as the first line, and should have
Subject: [powerpc,oldworld]
in the email headers.

Right?


Enjoy!

Rick


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

Re: miboot-floppies (powerpc oldworld) for sarge

[Rick Thomas]

On Dec 20, 2006, at 7:17 AM, Sven Luther wrote:

Well, i think what would be interesting is maybe to have a wiki  
page, listing
a cross table of all tested models, and the different boot methods,  
and
listing the working reports and not working ones, or something.  
Then give out

a call for testers of the last daily-builds.


That would be a good idea.  Unfortunately I don't have the time or  
the HTML skills to do such a thing.  Anybody else willing to set it  
up?  I'll contribute data-points if you do...





To make sure I have it right:   Bug reports (or success reports!)
should have
Package: linux-2.6
as the first line, and should have
Subject: [powerpc,oldworld]


Yes, apart, that after the [powerpc,oldworld] you can add a short  
description.


Cool... That's what I'll do.

Rick


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

Re: kde/gnome/xfce media other than full CDs

[Rick Thomas]

On Dec 20, 2006, at 3:52 PM, Joey Hess wrote:


peter green wrote:

it has recently been announced that there will be seperate CDs for
kde/gnome/xfce with different desktop tasks and package selections.

but what is the plan for other means of installation
(buisnesscard/netinst/floppies/netboot)? will there be 3 seperate
desktop tasks listed? will there be one desktop task that installs  
all

three? or will kde and xfce desktops simply not be availible from
tasksel.


In all cases you get a choice what desktop environment you want,  
before
you enter d-i. d-i/tasksel never requires you to make this choice  
in the

user interface, though. In the case of regular sized CDs, you make the
choice when deciding what CD to get. In the case of all other
installation images, you can make the choice when booting the  
installer,

by typing "install tasks=kde-desktop"


Yeuch!  There's *got* to be a better way to let a user of only  
average expertise specify which desktop environment to install.


It should be an up-front/in-your-face option at "tasksel" time.

Rick


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

Bug#403778: installation-report: sudo password not specified

[Rick Thomas]

On Dec 21, 2006, at 1:16 PM, Tshepang Lekhonkhobe wrote:


Anyways, I didn't know that one has access to man pages during
installation. Is that because I'm a newbie myself?


So far as I know (and I've been using and administering Linux  
machines for 10 years, and UNIX machines for 20 years before that)  
the man pages aren't available during installation.


Rick


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

Re: State of Debian-Installer

[Rick Thomas]

Howabout if somebody volunteered (Geert?) to, on (say) a weekly  
basis, cut and paste the latest of Franz' status reports from the  
debian-boot mailing list into the wiki page?


would that work?

Rick


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

Re: 20061230-2 netinst powerpc iso

[Rick Thomas]

On Jan 1, 2007, at 10:10 AM, Ulrich Teichert wrote:


Hi,

[del]
I'm still downloading the 20070101-1 build. This will take a  
while


You may have been just too early. Please let us know if the new image
still does not work and we'll investigate deeper.


Sure, I just wanted to provide an update and explain why it will at  
least

take another 6 hours until I can test the new years build ;-)

CU,
Uli


From the cdimage.debian.org:cdimage/daily-builds/sid_d-i/20070101-1/ 
powerpc/iso-cd/  "businesscard" CD.


# find /mnt -iname '*prep*' -print
/mnt/install/prep
/mnt/install/prep/vmlinuz-prep.initrd
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/affs-modules-2.6.18-3- 
prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/crypto-modules-2.6.18-3- 
prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/ext2-modules-2.6.18-3- 
prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/ext3-modules-2.6.18-3- 
prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/fat-modules-2.6.18-3- 
prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/firmware- 
modules-2.6.18-3-prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/floppy-modules-2.6.18-3- 
prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/hfs-modules-2.6.18-3- 
prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/ipv6-modules-2.6.18-3- 
prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/irda-modules-2.6.18-3- 
prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/jfs-modules-2.6.18-3- 
prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/loop-modules-2.6.18-3- 
prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/md-modules-2.6.18-3-prep- 
di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/nic-extra- 
modules-2.6.18-3-prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/nic-modules-2.6.18-3- 
prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/nic-pcmcia- 
modules-2.6.18-3-prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/nic-shared- 
modules-2.6.18-3-prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/pcmcia-modules-2.6.18-3- 
prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/pcmcia-storage- 
modules-2.6.18-3-prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/ppp-modules-2.6.18-3- 
prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/reiserfs- 
modules-2.6.18-3-prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/sata-modules-2.6.18-3- 
prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/scsi-extra- 
modules-2.6.18-3-prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/serial-modules-2.6.18-3- 
prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/ufs-modules-2.6.18-3- 
prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/usb-serial- 
modules-2.6.18-3-prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-kernel-di-powerpc-2.6/xfs-modules-2.6.18-3- 
prep-di_1.26_powerpc.udeb
/mnt/pool/main/l/linux-modules-di-powerpc-2.6/loop-aes- 
modules-2.6.18-3-prep-di_1.02_powerpc.udeb

/mnt/pool/main/p/partman-prep
/mnt/pool/main/p/partman-prep/partman-prep_9_powerpc.udeb
/mnt/pool/main/p/prep-installer
/mnt/pool/main/p/prep-installer/prep-installer_0.4_powerpc.udeb
#

Enjoy!

Rick


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

Re: Powerstack II: installation success with some quirks

[Rick Thomas]

On Jan 4, 2007, at 1:28 PM, Frans Pop wrote:


(CC to Colin for comments about disabling prep support on CDs.)

On Thursday 04 January 2007 11:24, Ulrich Teichert wrote:

after using the correct daily netboot image:
cdimage.d.o:/cdimage/daily-builds/sid_d-i/20070103-2/powerpc/iso- 
cd/deb

ian-testing-powerpc-netinst.iso


That is not a netboot image...
The netboot image is:
http://people.debian.org/~wouter/d-i/powerpc/daily/prep/netboot/


Now, to the details:
booted the initrd-image from the CD over tftp with:
`boot net:,vmlinuz-prep.initrd,`


I'm not sure that we want to support booting a CD image over the  
network.
The real netboot method and the hd-media method are valid  
alternatives.


...


Booting from the CD-ROM did not work and I didn't expected it would -
no big deal, IMHO, as long as it's noted in the manual. I've just
checked and prep is listed under supported architectures for CD-ROM
booting. If Powerstacks could be simply excluded there, this would be
appreciated.


If booting of CD is not supported, I'd just as soon remove support for
prep from CD images. The powerpc CDs are already very large because  
they
have multiple kernels and initrds. Removing the prep support would  
free

up some much needed space for regular packages.

AFAICT the only change needed is to disable building the d-i prep  
cdrom

images in d-i build system; debian CD will then do the correct thing.

We'll still need the udebs and kernel image on the CD for the hd-media
installation method though.


...

Thanks for your testing and glad to hear you managed a successful  
install.


I support efforts to decrease the size of the powerpc CDs, but...

Unless the prep/netboot image from Wouter can reproduce this success,  
the changes Frans is suggesting will (If I understand them correctly)  
render this success obsolete.  The process Ulrich described will not  
work with the etch/sid netinst CD if prep support is removed.


Or am I missing something?

Rick


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

Re: RFC: warn user about first accounts privs

[Rick Thomas]

On Jan 11, 2007, at 5:47 PM, dann frazier wrote:


I'm curious if
I'm the only experienced admin who didn't notice and is surprised.


I didn't notice for a while, and I was surprised to find out.

I've been sysadmining UNIX for 25 years, and Linux for 10.

Rick


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

Re: Debian has failed us

[Rick Thomas]

It's sad to see d-i on powerpc, a major architectural variant, being  
eroded and neglected as a result of a few people who can't get past  
their own personal animosity to Sven.  Whatever his merits or  
demerits -- and I'm not going to get drawn into a debate on that  
topic -- the powerpc architecture deserves better than it's seen  
recently.


Rick

Can't we all just get along?
-- Rodney King


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

Re: Debian has failed us

[Rick Thomas]

On Jan 15, 2007, at 12:39 AM, Christian Perrier wrote:


Quoting Rick Thomas ([EMAIL PROTECTED]):


It's sad to see d-i on powerpc, a major architectural variant, being
eroded and neglected as a result of a few people who can't get past
their own personal animosity to Sven.  Whatever his merits or
demerits -- and I'm not going to get drawn into a debate on that
topic -- the powerpc architecture deserves better than it's seen
recently.



Can you give us examples of actions from the D-I team who could be
used as illustrations of negligence towards the powerpc architecture
users?


Well, OK:

The one that particularly got to me was the very long dry spell for  
OldWorld PowerMacs during which etch would not boot on that  
hardware.  Sarge ran fine, but until the advent of the 2.6.18 series  
of kernels, etch never got much beyond the bootloader(*).  This  
resulted in "genetic drift" of applications software (specifically  
gstreamer is the one I know about.  There are probably others.)  
because it couldn't be tested.  The result is that now there's a very  
subtle release critical bug (#404876) and no time to fix it.


As Mathew just pointed out:  Kernels not booting may not be d-i's  
fault, but it is their problem.


The underlying problem was that the infighting in the d-i team  
distracted the in-fighters from taking bug reports on this topic  
seriously.  If Debian on powerpc hardware is to survive, the  
infighting problem has to be solved.  You can blame other people all  
you want, but it won't solve the problem.


I don't expect the people this is about to understand what I'm  
talking about here.  They are blinded by their own hatred, thus  
unable to see the larger problem.


The powerpc architecture deserves better than it's seen recently.


Rick


*) Bootloader == BootX in my case.  I tried miboot as well, and it  
booted with a differently configured kernel, but when it came time to  
reboot into the production kernel, I got the same results -- nothing  
beyond the bootloader messages.



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

Re: Debian has failed us

[Rick Thomas]

On Jan 16, 2007, at 8:46 PM, Mathew Binkley wrote:

As I said earlier, there shouldn't be an arbitrary line between the  
installer and the packages.  People aren't installing "Etch the  
installer" or "Etch the packages", they're installing "Etch the  
release". Test both parts simultaneously, as a single unit, because  
that's how your users are going to use them.


Mat has an important point here.  Too often, I've seen: "I'm sorry  
about the bugs you've encountered post-install.  It sounds like the  
install itself went fine.  I'm closing this report."   Usually, the  
person making the report doesn't have a clue about the fine points of  
which package team is responsible for the bug she has encountered.   
The person best able to make that determination is the person who  
fields the installation report.  But it seems that person can't be  
bothered to pass the report along, and feels no responsibility to try.


Sad.

Rick


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

Bug#408818: cool

[Rick Thomas]

On Jan 28, 2007, at 9:42 PM, Phill Thorpe wrote:


On Mon, 2007-01-29 at 03:09 +0100, Frans Pop wrote:

On Monday 29 January 2007 02:41, Phill Thorpe wrote:

I dont think that you read it correctly.
This install did not detect my nic at first, it only detected my nic
when I booted with:
install interface=eth1
Which I dont believe should be necessary.


No, the "interface=eth1" option does not have _any_ effect on  
detection of

NICs, only on which one is _used_.

So, unless you can point us to a real problem with NIC detection/ 
selection
during your initial install, there is no problem as far as we can  
tell.


Cheers,
FJP


So I wonder why it detected the wrong one, when I only have one nic.

regards
 Phill.



Does looking closely at what Phill supplied for his original report  
help any?:



Output of lspci -nn and lspci -vnn:
phill:/home/phill# lspci -nn
...  ...
00:12.0 Ethernet controller [0200]: VIA Technologies, Inc. VT6102  
[Rhine-II] [1106:3065] (rev 7c)

... 

Output of lspci -vnn:
phill:/home/phill# lspci -vnn
..
00:12.0 Ethernet controller [0200]: VIA Technologies, Inc. VT6102  
[Rhine-II] [1106:3065] (rev 7c)

Subsystem: ASUSTeK Computer Inc. Unknown device [1043:80ed]
Flags: bus master, medium devsel, latency 64, IRQ 217
I/O ports at d800 [size=256]
Memory at febff800 (32-bit, non-prefetchable) [size=256]
Capabilities: [40] Power Management version 2
...


I'm not enough of a PCI hardware guru to know if this is going to be  
recognized as a NIC.  Anybody else?


Also, I saw a bunch of PCI bridges and so on.  Is is possible that  
the NIC is on the far side of one of them and that's confusing the  
issue?  Or am I talking total nonsense?


Just a thought...

Rick



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

Re: debian-installer ("auto" is confusing)

[Rick Thomas]

On Jan 31, 2007, at 1:41 PM, Frans Pop wrote:


On Wednesday 31 January 2007 19:27, Robert Millan wrote:
It just mirrors the same boot options offered by the official  
installer

(except "rescue" which doesn't make much sense in this context).


No, with official images "auto" is not really offered as a boot  
option. It

is not described in the syslinux help screens.
It is an option that is only to be used for fully automated  
installs using

preseeding and is only described in the relevant appendix in the
installation guide.


...I was under the assumption that questions undisplayed due to
prio=crit always reverted to defaults.  Shouldn't the default mirror
for Australia be used instead?


Maybe, but it currently does not and that will not change for Etch.  
And

that still does not solve the problem of not being able to select a
mirror, nor a lot of other useful stuff.
Basically, priority critical has never been intended for interactive
installs.


Then maybe 'auto" should be renamed to something like 'preseeded' ?   
It might avoid confusion in the future...


Rick


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

Re: Errors on downloading d-i images

[Rick Thomas]

On Feb 7, 2007, at 2:47 PM, Geert Stappers wrote:


Op 06-02-2007 om 11:50 schreef hugo vanwoerkom:


11:40:09 (8.04 KB/s) - Connection closed at byte 2746856. Retrying.


 


Do other people have this trouble?



For what it's worth: It is the first report.


I've had it in the past.  Not so much recently.

Rick


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

Bug#410625: G3 B/W pcilynx firewire blues

[Rick Thomas]

On Feb 12, 2007, at 7:22 PM, Benjamin Herrenschmidt wrote:


On Mon, 2007-02-12 at 15:30 -0500, Rick Thomas wrote:


Maybe I should try the whole thing again and write down the details.

Can you give me a URL for the linux ieee1394 mailing list?


https://lists.sourceforge.net/lists/listinfo/linux1394-devel

Ben.


Looking at the archives, it appears that pcilynx on PPC hardware has  
been broken horribly for a long time (at least since 2004).   
Furthermore, it looks like support for pcilynx may be dropped  
entirely from the next generation of ieee1394 drivers.


I'd suggest that pcilynx be disabled in the Debian PPC kernels, and a  
note to that effect be put in the release documents.


To what package should I send a bug report with that payload?

Rick



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

Bug#410971: interactive aptitude wants to remove hfsutils and sudo after etch installation

[Rick Thomas]

Package: installation
Severity: important


After installation on a powerpc system (PowerMac Blue&White G3) when aptitude 
is run in interactive (curses) mode, told 
to do update and finish any pending operations ("g") command, it turns out that 
the "hfsutils" and "sudo" packages are 
marked for deletion.  If aptitude is told to go ahead (a second "g" command) it 
will delete them.  This has bad effects 
on the health of the system.

*) hfsutils is used by ybin to setup the boot partition on Macs.  Without it, I 
can't use ybin to change bootstrap 
parameters.

*) sudo is needed if I choose "no root login" at installation time (which I 
usually do).  Without it I can't do system 
administration of just about any kind.

Interestingly,  non-interactive )commandline) aptitude does not seem to share 
this odd proclivity.  E.g. "aptitude 
install ntp" does not automatically delete hfsutils or sudo.

If it's important (this has been going on for a while, so it's not just a 
glitch in the current daily install CDs) this 
shows up when installing from [Debian GNU/Linux testing _Etch_ - Official 
Snapshot powerpc NETINST Binary-1 
20070214-00:13]




-- System Information:
Debian Release: 4.0
  APT prefers testing
  APT policy: (500, 'testing')
Architecture: powerpc (ppc)
Shell:  /bin/sh linked to /bin/bash
Kernel: Linux 2.6.18-3-powerpc
Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968)


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

Bug#410971: interactive aptitude wants to remove hfsutils and sudo after etch installation

[Rick Thomas]

On Feb 14, 2007, at 8:00 PM, Frans Pop wrote:


On Thursday 15 February 2007 01:10, Joey Hess wrote:

Rick Thomas wrote:

After installation on a powerpc system (PowerMac Blue&White G3) when
aptitude is run in interactive (curses) mode, told to do update and
finish any pending operations ("g") command, it turns out that the
"hfsutils" and "sudo" packages are marked for deletion.  If aptitude
is told to go ahead (a second "g" command) it will delete them.   
This

has bad effects on the health of the system.


I have seen this occasionally in the past, especially when I used  
'apt-get

builddeps' or some such on an installed system. However, I have never
seen this after installs.

I also cannot reproduce it if I install d-i with sudo option. If I  
login
afterwards and do 'sudo aptitude' followed by 'u' and 'g', I just  
get "No

packages are scheduled to be installed, removed or upgraded.


OK, here's the exact sequence of events as near as I can remember --  
I didn't write it down.  If it's critical, I can re-do the install  
and write *everything* down.  Also, if it would be helpful, I can  
supply log files.


Start with a bog-standard Blue&White PowerMac G3 (but I've seen this  
on other PowerMac machines, so the exact hardware may not be important.)


Installed from the Netinst CD (I've seen it with Businesscard CDs  
too) at the boot prompt chose the default type of install (Non- 
expert.  Though I've seen it with an "expert" install too.)  Accepted  
all the defaults to all the questions except...


When it asked for a root password, I hit "go back" and told it I  
didn't want to allow root logins.  Then I proceeded with default  
answers from then on.  I installed a standard system at tasksel time.  
(un-checked "Desktop Environment", checked "Standard System", and  
left unchecked everything else.  But I've seen this with "Desktop  
Environment" checked as well.)


After the reboot, I logged in as normal user, did "sudo su -" to get  
to root.  I edited the sources.list file to get rid of the line that  
calls for the netinst CD.  I then did "aptitude install rsnapshot  
enscript ntp lynx-cur openssh-server lpr gpm bzip2 fbset mouseemu"  
and let it run it's course.


Then I did a bare "aptitude" and in the aptitude interactive mode I  
hit "u" and "g".  It told me it wanted to delete sudo and hfsutils.   
I told it to "install" those two instead (hit "+" for each of them)  
and gave it a "g" which installed nothing and removed nothing.


Would log files help?

Rick



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

Re: Bug#407460: USB ethernet interface renamed after installation on NSLU2 which causes the system to be inaccessible

[Rick Thomas]

On Feb 27, 2007, at 6:23 PM, Joey Hess wrote:


Gordon Farquharson wrote:

I tried a build of the installer from trunk (revision 45431) and
nearly fell of my chair:


Known problem at the top of DebianInstaller/Today in the wiki.


I don't see any reference at all on the wiki to Gordon Farquharson  
nearly falling off his chair...  <-8




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

Bug#411446: Clock not set correctly; MacOS9 not detected?

[Rick Thomas]

So the problem is that (aside from the possibility that OS9 is not  
being recognized) if the other OS is OS9, the default for hardware  
clock should be "local time", but if the other OS is OS-X, the  
default for the hardware clock should be "UTC".  Is that a correct  
assessment?


I solve it by running my OS9 in UTC.  Others are not so likely to  
appreciate that solution!


Rick



On Feb 28, 2007, at 12:14 PM, Wouter Verhelst wrote:


On Wed, Feb 28, 2007 at 05:53:16PM +0100, Frans Pop wrote:
Note that Debian people tell me that OSx systems _are supposed_ to  
run

with internal clock on UTC.


This seems correct, yes; My PowerBook happily runs both Debian and  
OS X

at this time, with clock set correctly, and with 'tail -n 1
/etc/adjtime' returning 'UTC'.




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

Re: Etch netinstaller has no eth0 in qemu

[Rick Thomas]

On Mar 5, 2007, at 5:40 PM, Joey Hess wrote:


Something wrong with rsync? I think zsync can also be used.


zsync works great.  It doesn't work with DVD images -- something  
about files larger than 2GB (31 bit byte offsets).


Rick


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

Bug#413814: installing Debian GNU/Linux 4.0 on a Power Macintosh G3 Server

[Rick Thomas]

Hi Alex!

Welcome to an elite minority of those of us who have got this to work!

Below are a couple of hints from my own experience in doing this.

Rick

On Mar 7, 2007, at 5:21 AM, Alex Teclo wrote:


Package: installation-reports

Boot method: BootX
Image version: Debian etch powerpc weekly build
Date: February 4th 2007

Machine: Power Macintosh G3 Server
Architecture: powerpc
Processor: PowerPC 740/750 (G3)
Memory: 256 MB
Comments/Problems:

Here's a feedback of my installation of Debian 4.0 "etch" on a  
Power Macintosh G3 MiniTower


Since it's an oldworld macintosh, it cannot boot from the Debian  
CDs. I tried booting from floppies, but it never worked, the floppy  
drive is probably dead.

So my solution was to install a legit copy of MacOS 9.2

Here's what I did:
- boot into MacOS 9.2
- launch BootX with the files vmlinux and initrd.gz
- debian-installer runs smoothly and everything goes well, except  
one thing: when debian-installer attempts to install quik as a boot  
loader, the operation fails, and the error messages states that  
"the partition is not ext2". This error messages seems odd to me.  
What partition does d-i mean ? The partition where MacOS 9.2 is,  
which is HFS ? One of the Linux partitions, which are ext3 and not  
ext2 ? Anyway, isn't ext2 the same thing as ext3 without  
journalization ?
- anyway I'm not worried at all that quik could not install,  
because I still can boot into MacOS 9.2 and then fire up BootX to  
boot into Debian GNU/Linux

- debian-installer finished up its work and reboots the machine
... and then something goes wrong: I see on the power macintosh a  
flashing "?" in a floppy. This means "I can't find any operating  
system to boot" ... ouch ! Now that's a problem.

- I tried zapping the PRAM, it did not help.
- Finally I took the "Outil Disque Dur" diskette and chose the menu  
"Fonction" and then "Mise à jour". Pardon my French, this should  
translate to "Apple Disk Tool", menu "Functions" and then "Update".  
I don't know the exact wording since I've never used MacOS 9.2 in  
any other language than French :)

... and *yes*, now I can boot into MacOS 9.2



At some point after the installer has stepped you through choosing  
language and keyboard, but before running the partitioner, you can  
choose "go back" from one of the question screens. That will get you  
to a top-level menu which has an option (near the bottom) of changing  
the priority level of the questions that the installation asks you.   
Change it to "low" or "medium".  From now on, the installer will  
return to that top level menu screen between installation steps.   It  
will also ask you more questions than it did at default priority --  
just choose the defaults if you don't know the answers.


This will give you a chance to prevent it from trying to install the  
quik bootloader (choose "proceed without installing a boot loader"  
instead.)  This will keep it from messing up the OS-9 bootstrap stuff.


When it ejects the CD and pauses one last time before rebooting,  
switch to VT2 (hit -F2) where you can start up a limited shell  
by hitting the  key.  The root partition for the installation  
is mounted on "/target".  have a look around if you like (do "ls",  
"df", "ps" and anything else non-destructive.)  Then do the following  
stuff:


mkdir /target/MacOS
# you *may* need to do "modprobe hfs" here (or "modprobe hfsplus").   
I don't recall.
# In any case, you've got the installer kernel and it's associated  
modules in the

# installer ramdisk, where you currently are, so it will work.
	mount -t hfsplus /dev/hda6 /target/MacOS  # or "-t hfs" if that's  
what it is.

chroot /target
# Now you're in "bash" inside the installed system.
# If you like, you can poke around a bit now to get a feel for the  
environment.
	cp /boot/vmlinux /MacOS/System\ Folder/Linux\ Kernels/vmlinux-new
# or whatever you like to call it
	cp /boot/initrd.img /MacOS/System\ Folder/Linux\ Kernels/initrd- 
new.img  # or whatever

exit
# now you're back into the installer environment with it's limited shell
umount /target/MacOS

Then switch back to the VT1 (-F1) console and proceed with the  
final stages of the installation.


It will reboot into MacOS-9.2 and pause at the BootX screen.  Enter  
your "root=/dev/hda9" in the parameter box and choose your new kernel  
and initrd.img files.  Then allow it to boot into Linux.  You should  
be good to go.





... but now when I boot into Debian GNU/Linux with BootX, there is  
another, more serious problem:
When I boot into Debian GNU/Linux with BootX, I can only use the  
vmlinux and initrd.gz of debian-installer. So instead of booting  
into my system, I boot into debian-installer.
Attempt to solve this problem: start a shell from debian-installer,  
and chroot to my Debian GNU/Linux system. From there, find the  
vmlinux and initrd.gz that will allow me to boot directly into  
Debian GNU/Linux (they are in /boot) an

Bug#389881: RC-ness of this bug

[Rick Thomas]

On Mar 15, 2007, at 1:12 PM, Frans Pop wrote:


On Thursday 15 March 2007 17:44, Colin Watson wrote:

Personally I also feel that all possible solutions effectively make
/etc/fstab unreadable and unmaintainable.


The approach we took in Ubuntu was to put comments above each UUID
entry in /etc/fstab documenting which traditional device name they
corresponded to at the point of installation. Of course this can get
out of date, but I don't think there's really any sensible way around
that.


My main point is not that the UUID itself is not readable, but  
rather that

the lines get way too long and, depending on your editor (settings),
either get wrapped or disappear off screen. You loose the easy  
overview

of what's in fstab.
/etc/fstab used to be fairly maintainable because the info could be  
kept
in columns fairly easily for most cases and because the info would  
mostly

fit on one line [1].

IMO with a switch to UUIDs we are going to need an fstab editor  
(console

based) that:
- does the translation to the "normal" device names on the fly (and  
thus

  does always reflect the actual situation)
- provides different 'views' of what's in fstab
- allows to select what representation for the file system should be
  used in the fstab (traditional, path, uuid, id, ...)
- allows to set mount point, type, mount options, etc.
- sorts partitions into a logical order
- maybe knows about removable devices
- has a simple interface to add new entries for e.g. USB sticks
- ...

[1] Yeah, I know this is not true for NFS volumes and if a lot of  
options

are used, but in general it was true.



If you're going to all the trouble of a smart fstab editor, why not  
simply define a more modern format (e.g. like that of dhcpd.conf) for  
the information that can accommodate line breaks and nesting.  Change  
the name to something else, don't call it fstab; if the new file  
doesn't exist the mount command (and the rest of them that currently  
read fstab) will default to /etc/fstab if present.


The biggest problem will be identifying all the places where fstab is  
currently *assumed* to be present and making them all use the new  
file.  A library is probably needed that does the deciding of which  
format to use.


Are there POSIX/LSB/etc ramifications?

just a thought,

Rick


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

Re: the future of the netinst image

[Rick Thomas]

On Mar 25, 2007, at 7:10 AM, Geert Stappers wrote:



Back to 'netinst'

This image is one that does disappoint people often.


What could be done to prevent that?


I see two possibilities:

(A) Eliminate it entirely
(B) Treat it as an expanded "businesscard" image.

Discussion:

(A) I'm serious.  Get rid on it.  It's not needed and it just  
confuses people.  I've personally encountered quite a few times, and  
observed on this list many more times, that the "netinst" image  
violates the "principle of least astonishment" in a number of ways.


	First, it's misnamed.  People expect it to somehow involve booting  
from the LAN, which it doesn't.  That confuses people.  Once they get  
that figured out, the existence of a "network install" CD makes them  
wonder what the "businesscard" CD is, if not a network install CD.   
That confuses people.
	Second, it's significantly larger (factor of 3) than the  
"businesscard" image, but still not complete -- even for a small  
server (non-X) install.  You still need access to the Internet (or a  
local mirror repository) to complete the installation.
	Third, It insists on leaving remnants of itself in the installed  
sources.list file.  This just serves to confuse people by demanding  
to have the install CD inserted any time you want to do an update.   
That's the kind of stuff you expect from Microsoft, not free software!
	Fourth, it's too big to download without broadband.  At 56 Kbits/ 
sec, it takes about 9 hours to download a netinst image, and you  
still need to get more stuff from the network during the install if  
you want more than a very minimal system.  If you do have broadband,  
you still don't want the netinst image.  You might as well either  
download a full CD image with everything you need, or install from a  
businesscard image where you will always get the latest-and-greatest  
stuff from the network.
	Finally, nobody has a good idea of what should be on it.  It's  
neither fish nor fowl, and everybody expects something different of  
it.  This recent business of not having the K7 kernel is a good example.



(B) If we still want to keep it around, we need to make it less  
"astonishing" by at least eliminating the curiosa listed above.  The  
easiest way to do that, seems to be to have it do exactly what the  
businesscard image does, but more of it.


Rick


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

etch/volatile -- public key is not available: NO_PUBKEY EC61E0B0BBE55AB3

[Rick Thomas]

Anybody know what the problem is here?  Or how to fix it?


W: GPG error: http://volatile.debian.org etch/volatile Release: The  
following signatures couldn't be verified because the public key is  
not available: NO_PUBKEY EC61E0B0BBE55AB3

W: You may want to run apt-get update to correct these problems

Thanks!

Rick


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

Re: etch/volatile -- public key is not available: NO_PUBKEY EC61E0B0BBE55AB3

[Rick Thomas]

On Apr 14, 2007, at 6:39 AM, Steve Langasek wrote:


On Fri, Apr 13, 2007 at 11:21:56PM -0400, Rick Thomas wrote:

Anybody know what the problem is here?  Or how to fix it?



W: GPG error: http://volatile.debian.org etch/volatile Release: The
following signatures couldn't be verified because the public key is
not available: NO_PUBKEY EC61E0B0BBE55AB3
W: You may want to run apt-get update to correct these problems


You need to import the public key for volatile.debian.org  
separately using

apt-key if you want to use this repo with secure apt, the key is not
included in the default debian-archive-keyring package.


Thanks,

Two questions:

1) How do I import the public key for volatile?  (If it's in the fine  
manual, then please point me to it so I can RTFM)


2) What's the reasoning behind not including it in the default debian- 
archive-keyring package?  Is it not recommended that everyone use  
volatile for the things like security and spam updates that it provides?


Enjoy!

Rick


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

Re: Red X -- was: Re: new oldworld pmac miboot 2.6 floppies, root size should be ok, net_drivers still too big, please test.

[Rick Thomas]

On Sunday, August 29, 2004, at 04:31 AM, Rick_Thomas wrote:
I tried the new 2.4 PowerMac floppys today.  Now I get the Red "X" on
the 2.4 boot floppy as well.
I did an experiment...
I mounted the 2.4 boot floppy and extracted the zImage file,
uncompressed it, and compared it to the 2.4.25-powerpc-small kernel.
They both claim to be the same kernel in that
strings $file | grep 'Linux version'
produces the same output for each file. Namely:
Linux version 2.4.25-powerpc-small ([EMAIL PROTECTED]) (version gcc 3.3.3
(Debian 20040401)) #1 mer avr 14 17:26:11 CEST 2004
But the two files differ according to "cmp".
So I wonder if the process of compressing the kernel and putting it on
the miboot floppy is somehow corrupting it?
Rick

OOOps...
I missed the "objcopy -O binary" that was occurring in the log file.
When I did that, the two files were identical.
So the compression and copying to the floppy image are happening 
correctly.

Is it possible the objcopy is corrupting it?
mumble...
Rick
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Red X -- was: Re: new oldworld pmac miboot 2.6 floppies, root size should be ok, net_drivers still too big, please test.

[Rick Thomas]

Sven Luther wrote:
> Rick Thomas wrote:
> > Is it possible the objcopy is corrupting it?
> 
> This is indeed a possibility.
> 
> I will disable this again for the 2.4 floppies, and we will see tomorrow what
> happens.
> 
> That said, the 2.6 floppies are too big to work with miboot without the
> objcopy -O binary call, so we need to investigate this more in detail.


I downloaded the latest 2.4 floppy images, but it looks like you didn't get around to 
disabling the objcopy yet.

In any case, I still get a red X.

The objcopy and the bfd library on my most recent debian sarge install are dated May 
19th, 2004.  Is it possible that we haven't had a successful powerpc OldWorld floppy 
boot since that time?

Rick


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

Re: Red X -- not quite gone yet...

[Rick Thomas]

On Monday, August 30, 2004, at 04:41 AM, Sven Luther wrote:
On Mon, Aug 30, 2004 at 08:23:28AM +0200, Sven Luther wrote:
On Mon, Aug 30, 2004 at 01:58:34AM -0400, Rick Thomas wrote:
Sven Luther wrote:
Rick Thomas wrote:
Is it possible the objcopy is corrupting it?
This is indeed a possibility.
I will disable this again for the 2.4 floppies, and we will see 
tomorrow what
happens.

That said, the 2.6 floppies are too big to work with miboot 
without the
objcopy -O binary call, so we need to investigate this more in 
detail.
I downloaded the latest 2.4 floppy images, but it looks like you 
didn't get around to disabling the objcopy yet.
Yeah, alioth and thus the d-i svn repo was dead yesterday. It is 
up again, and i will remove it for tomorrow.
Fixed, please try tomorrows floppy-2.4 builds.
Friendly,
Sven Luther
Not yet...  When I boot the ofonlyboot (and the boot) floppy, I get 
the tuxmac icon for only a few seconds.  The red X appears almost 
immediately.  When I looked at the filesystem, it has a zero length 
zImage file, and no vmlinu* file.  Commenting out the objcopy means 
that there is no vmlinux.bin file for gzip to work on.

Rick

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

Re: Red X -- not quite gone yet...

[Rick Thomas]

Sven,
It's 2:15 AM, and I've got a meeting tomorrow at work, so I won't 
be able to test these tonight.  I'll try to get to them tomorrow 
(9/1) in the evening (US East Coast time).

Please do me a favor and loop-mount the images to see if they have 
all the expected pieces and the pieces are of the expected sizes.  
That way we won't loose another round of debugging for a problem 
that doesn't require an actual bootstrap to discover.  Thanks!

Take care!
Rick
On Tuesday, August 31, 2004, at 11:03 AM, Sven Luther wrote:
On Tue, Aug 31, 2004 at 04:15:01PM +0200, Wouter Hanegraaff wrote:
On Tue, Aug 31, 2004 at 03:41:50PM +0200, Sven Luther wrote:
Please try the (2.6) floppies at :
http://people.debian.org/~luther/miboot/floppy-2.6-2004.08.31
boot.img only contains vmlinuz and doesn't boot, ofonlyboot.img starts
loading a kernel and ends with the tux/red cross.
http://people.debian.org/~luther/miboot/floppy-2.4-2004.08.31
boot.img and ofonlyboot.img start loading a kernel and end with the
tux/red cross.
http://people.debian.org/~luther/miboot/floppy-2.4-old-2004.08.31
boot.img only contains vmlinuz and doesn't boot, ofonlyboot.img starts
loading a kernel and ends with the tux/red cross.
Could you try both daily-builds tomorrow ? and todays 2.4 one too.
I think miboot _never_ worked for you, right ? What is your box again ?
Friendly,
Sven Luther
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact 
[EMAIL PROTECTED]


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

Re: Red X saga -- success!!! (up to a point...)

[Rick Thomas]

On Wednesday, September 1, 2004, at 04:45 AM, Sven Luther wrote:
Ok, try out :
  http://people.debian.org/~luther/miboot/floppy-2.6-2004.09.01
I have checked the ofonlyboot, boot and root floppies.
I tried this -- the usual 30 seconds or so of floppy noises 
followed by "Red X" for both the boot and ofonlyboot floppies.

also
http://people.debian.org/~luther/miboot/floppy-2.4-2004.08.31
same result - 30 seconds or so or floppy reading then "Red X" for 
both the boot and ofonlyboot floppies.

I'm going to test the "daily" 2.4 floppies at
http://people.debian.org/~luther/d-
i/images/daily/powerpc/floppy-2.4/
then go to bed.
... pause while I go into the other room to try things out ...
Wonder of wonders!  It boots and reads the "root" and two drivers 
floppies!  It got as far as finding the network interface card and 
the CD-rom drive and reading the d-i udebs from the CD.  It even 
tried to run the partitioner, but there it failed to find my hard 
disk.

Oh well -- too much success in one day is bad for your karma...  (<-8)
We'll work on why it didn't find my hard disk another day.  It's 
almost 4 AM and I'm going to bed!

Enjoy!
Rick
Here's where I got the successful boot floppy from:

Index of /~luther/d-i/images/daily/powerpc/floppy-2.4
 NameLast modified   Size  Description

 Parent Directory26-Aug-2004 21:35  -
 asian-root.img  01-Sep-2004 21:26   1.1M
 boot.img01-Sep-2004 21:27   1.4M
 cd-drivers.img  01-Sep-2004 21:27   1.4M
 net-drivers.img 01-Sep-2004 21:27   1.4M
 ofonlyboot.img  01-Sep-2004 21:27   1.4M
 root.img01-Sep-2004 21:28   1.2M

Apache/1.3.26 Server at people.debian.org Port 80

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

Bug#269529: OldWorld pmac installation -- several problems

[Rick Thomas]

Joey Hess wrote:
> 
> > Note 6:
> >
> > Prior to the reboot, I tried to "Save debug logs" to a floppy disk.
> > It failed because "No floppy device was found".  The oldworld Macs all
> > use the "swim3" floppy controller.
> 
> Do you know if the floppy module is supposed to support this controller?

The eventually installed 2.6.7 system can access the floppy drive (read
and write, eject, sense the write-protect tab, etc)  so the Mac floppy
controller *is* supported by *something* in the 2.6.7 kernel/modules suite.

On the installed system (after the reboot), the date on the "floppy.ko"
module is reasonable (Aug 5 18:32 -- exactly the same as all the other
modules in that directory).  Is that a clue?



In  the original installation report, (with reference to conditions
during the installation and before the reboot), I wrote:

> Attempting to manually "modprobe floppy" from the F2 console gives the
> error:
> 
> FATAL: Error inserting floppy
> (/lib/modules/2.6.8-powerpc/kernel/drivers/block/floppy.ko),
> Device or resource busy
> 
> Interestingly, floppy.ko has date "Jan 1 1970".  All other modules in
> that directory have date "Aug 26 14:53".  Strange...


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

Bug#269529: OldWorld pmac installation -- several problems

[Rick Thomas]

Joey Hess wrote:
> 
> Can you mail the /var/log/debian-installer/syslog and messages to this

The complete set of logs and other system info from that install are
available at:

http://rcthomas.org:7879/~rbthomas/logfiles/install-6500/

I can mail them to the bug if you like, but why bother?

Rick


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

Bug#269529: OldWorld pmac installation -- several problems

[Rick Thomas]

Joey Hess wrote:
> 
> Rick Thomas wrote:
> >   http://rcthomas.org:7879/~rbthomas/logfiles/install-6500/

> Permissions prevent me from reading the syslog.

Fixed.  Sorry!

Rick


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

Bug#270239: Timezone configuration should not show 'UTC' after system time

[Rick Thomas]

Frans Pop wrote:
> 
> On Monday 06 September 2004 19:53, Joey Hess wrote:
> > AfAIK hwclock output never includes the timezone.
> 
> I'm afraid it does.
> 
> On a system installed with LANG=en_US (on which I based my report):
> # hwclock --show --localtime | awk '{NF-=2; print $0}'
> Mon 06 Sep 2004 09:00:31 PM CEST
> 
> On a system installed with [EMAIL PROTECTED]:
> # hwclock --show --localtime | awk '{NF-=2; print $0}'
> ma 06 sep 2004 18:53:11 CEST
> 
> This is of course after I selected my timezone to be Europe/Amsterdam. hwclock
> apparently (from the output in base-config) shows 'UTC' before timezone
> selection.
> 
> I've got no clue why it does not show the timezone in your situation.

Apparently, it looks at the "LANG" environment variable...

LANG=C /sbin/hwclock --show --localtime
gives
Tue Sep  7 02:22:16 2004  -0.463550 seconds
but
LANG=en_US /sbin/hwclock --show --localtime
gives
Tue 07 Sep 2004 02:23:24 AM EDT  -0.950496 seconds

Suggested fix: Use a "LANG=C" prefix.

Interesting...

I have no idea why the maintainers of hwclock care two figs about the LANG envariable. 
 But it seems they do.

Moral of the story: In a multicultural system like Debian, *everything* has to be 
tested with a variety of LANG values.  Welcome to the twenty-first century!

Rick


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

Bug#270239: Timezone configuration should not show 'UTC' after system time

[Rick Thomas]

On Monday, September 6, 2004, at 11:00 PM, Joey Hess wrote:
Rick Thomas wrote:
Apparently, it looks at the "LANG" environment variable...
LANG=C /sbin/hwclock --show --localtime
gives
Tue Sep  7 02:22:16 2004  -0.463550 seconds
but
LANG=en_US /sbin/hwclock --show --localtime
gives
Tue 07 Sep 2004 02:23:24 AM EDT  -0.950496 seconds
Ok, I see: I have LC_TIME=C while LANG=en_US.
Suggested fix: Use a "LANG=C" prefix.
That really doesn't work, the time display needs to be localised along
with everything else.
Unfortunatly I can't think of a good way to remove the time zone from
the display that'll work for all LC_TIME settings.
--
see shy jo
OK: How about this
	/sbin/hwclock --show --localtime --debug | sed -n 's/Time read 
from Hardware Clock: //p'
which prints this
	2004/09/07 05:03:27
Which is pretty much language/locale independent.

Rick

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

Bug#270599: Floppy install on Oldworld PowerMac

[Rick Thomas]

On Wednesday, September 8, 2004, at 05:18 AM, Sven Luther wrote:
On Wed, Sep 08, 2004 at 04:17:55AM -0400, Rick_Thomas wrote:
Package: installation-reports
powerpc boot-floppy 20040906 OldWorld PowerMac
...
Then I tried the "boot" floppy.  It gave me the tuxmac and made
reading noises.  After a while it ejected the boot floppy and switched
to a text mode screen at (I think) 640x480 resolution.  This is good
enough for installing -- but not satisfactory for long term usage.
Yes, i guess quik-installer should allow you to use further kernel 
options.
You could create your own miboot floppy, and then you can add the 
options you
want, more on this below.
For the time being, just adding "DEBCONF_PRIORITY=medium" to the 
standard "boot" floppy would be a major good-thing.

Frankly, I think that "medium" priority would be acceptable as a 
default mode for floppy boots.  It's nice to minimize user 
interaction and all, but if I'm going to all the trouble of burning 
a bunch of floppies, I want a reasonable amount of control over the 
details of the installation process.  I don't think inexperienced 
users will find the "medium" priority dialogue any more confusing 
than the existing "woody" installer.  Just my opinion -- YMMV, of 
course.


It called for the root floppy, so I fed it that, which it read
happily.  After reading the root floppy and asking me some questions
about languages and locations, it asked if I wanted to read a driver
floppy.  I said yes and fed it the "root-2" floppy.
Well, i fixed the root-2 thingy earlier, so it is nice that it 
works, even if
there is no real support for this in the installer yet. Feel free to
participate in the fixing of this, be it only by suggesting what 
the root-2
asking question should be, and where it should be asked. Ideally 
we would add
a load-second-root-floppy .udeb, which would present a menu and 
load the
second floppy, and which would be part of the first root floppy.
A special "load-root-2.udeb" (or whatever) may not be necessary:  
Just think of everything after the "root" as "extra installer 
component" floppies (or some such) rather than "driver" floppies 
specifically.  Simply modify the existing dialogue to end with a 
question "Do you want to load another installer component?"  If the 
answer is "yes" loop back and re-execute.  If the answer is "no", 
make a normal return.  The first pass through is mandatory and 
loads the "root-2" floppy.  All subsequent passes are optional, 
based on what kind of installation you want to do.

One thing to be careful of:  It will be necessary to craft the 
wording of the dialogue questions very carefully so that the user 
understands fully what is going on, and what is required of them, 
at each step.

The root-2 floppy contains stuff (namely netcfg and co) that was 
spilled out
from the first root floppy.
That's pretty much what I figured was going on.

My choice of "root-2" at this point was based on a hunch.  There was
no indication of which driver floppy it was expecting (Indeed, it was
not clear at all that "root-2" was a "driver" floppy.  My hunch was
that it would be needed immediately and that the easiest way to add
files to the ram-disk root was to emulate a driver floppy.)  It would
be better to out-and-out say "root-2" if that's what is wanted.
Like Joeyh mentioned, right now there is support for loading only 
one drivers
floppy, which may well be buggy in itself, and maybe a question 
for asking for
an additional floppy like asking for additional apt sources later 
on may be
welcome. Also, the root-2 is not really a drivers floppy, but 
should be loaded
earlier on, maybe.
I have no problem with answering questions about language and 
location before loading additional installer components.  I'd go 
ahead and leave it right where it is in the sequence, if that's 
easiest.


It then tried to detect my network interface and failed, so it asked
Because the net drivers are not on the floppy.
for the network drivers floppy, which I gave it.  This time it
succeeded in finding my network interface and configured it via DHCP
Woaw. I was under the impression that this would fail, from 
joeyh's comment
about only one driver floppy, but this is great.
Right.  This is what makes me think we can use the existing 
framework to load an arbitrary number of extra "installer 
component" floppies.


(I would have preferred the option to do this manually, but there 
is no
way to specify "DEBCONF_PRIORITY=medium" in booting an oldworld pmac
Like said, if you build your own miboot floppy, you can add any kernel
arguments you like. As miboot is non-free, users may be forced to 
do this
anyway, so ...
Yuch!  Please don't force inexperienced users to build their own 
"boot" floppy.  You'll loose a large class of potential users if 
you do.


machine from floppy.)  It asked for a mirror, and I specified the
uchicago one since it seems to be fastest and most reliable from my
little corner of the Internet.
Overcool.
Note 2:

Bug#270599: Floppy install on Oldworld PowerMac

[Rick Thomas]

On Wednesday, September 8, 2004, at 09:18 AM, Rick Thomas wrote:
On Wednesday, September 8, 2004, at 05:18 AM, Sven Luther wrote:
Please try again with todays floppies,
and if it doesn't fix the problem, we need to investigate what 
driver is
missing or something.
I'll try the new floppies tonight.
Ummm... The contents of
http://people.debian.org/~luther/d-i/images/daily/powerpc/floppy-2.4/
haven't changed in the last few days.  Is the build process stalled 
somewhere?

Thanks!
Rick

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

Bug#270599: Floppy install on Oldworld PowerMac

[Rick Thomas]

On Thursday, September 9, 2004, at 12:45 PM, Rick Thomas wrote:
On Wednesday, September 8, 2004, at 09:18 AM, Rick Thomas wrote:
On Wednesday, September 8, 2004, at 05:18 AM, Sven Luther wrote:
Please try again with todays floppies,
and if it doesn't fix the problem, we need to investigate what 
driver is
missing or something.
I'll try the new floppies tonight.
I guess whatever it was is fixed now.  Because I was able to 
download a set of floppies from

Index of /~luther/d-i/images/daily/powerpc/floppy-2.4
 NameLast modified   Size  Description

 Parent Directory26-Aug-2004 21:35  -
 asian-root.img  10-Sep-2004 05:46   1.2M
 boot.img10-Sep-2004 05:47   1.4M
 cd-drivers.img  10-Sep-2004 05:49   1.4M
 net-drivers.img 10-Sep-2004 05:50   1.4M
 ofonlyboot.img  10-Sep-2004 05:51   1.4M
 root-2.img  10-Sep-2004 05:52   1.4M
 root.img10-Sep-2004 05:53   1.3M

Apache/1.3.26 Server at people.debian.org Port 80
The "ofonlyboot" has not changed.  It reads and inverts the colors 
of the tuxmac, but never switches to text-mode screen from the 
inverted color tuxmac.

The boot floppy reads and switches to the text screen then asks for 
the root floppy, which it reads.  It then asks for language 
(English) and location (US) (but not keyboard layout) then invites 
me to load drivers from a floppy.  I gave it the "root-2" floppy 
and it complained about not being able to find any kernel drivers 
on that floppy.  I chose  and re-executed "load drivers 
from a floppy".  This time I gave it the net-drivers floppy, and it 
was happy.  Still thinking that we wouldn't get any where without 
the root-2 floppy loading (and being a bit bull headed anyway) I 
tried "load drivers from floppy" for the third time, and again fed 
it the root-2.  It complained again about not finding any kernel 
modules.  This time I told it to "continue without loading drivers" 
and to my amazement, it started decoding the stuff from the root-2 
floppy!  Curioser and curioser!

I think it was at this point that it asked for my keyboard layout, 
and suggested "European" as default, even though I had given it 
every reason to suspect that US-English was my preferred locale.  
I've reported this violation of the principle of least astonishment 
before.

It proceeded then to find my ethernet interface (remember I'd 
loaded the net-drivers floppy earlier) and do DHCP discovery on 
it.  This succeeded, as expected.  When it asked, I chose the 
uchicago mirror as usual, and it loaded the installer-components 
list (I think -- I didn't get the exact words) after which it 
*again* complained about not finding any kernel modules!  I told it 
to continue anyway, and it started downloading and unpacking 
installer components from the uchicago mirror (presumably).

When it got done with that and moved on to the partitioner, it 
couldn't find any of my disks (not my IDE main disk or my SCSI Zip 
disk).  The only IDE think it knew about was the CD-ROM drive.  
Exploring on the F2 console showed that it wasn't just the 
partitioner that was confused.  There was no evidence of IDE or 
SCSI disks in /proc or /dev.  (Same as last time -- no progress on 
that front...)

So I wrapped it up and took a tea break to write this report.
I have to consider the check for kernel modules at inappropriate 
times to be a serious bug...

Enjoy!
Rick

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

Bug#270599: Floppy install on Oldworld PowerMac

[Rick Thomas]

On Saturday, September 11, 2004, at 07:11 AM, Russell Hires wrote:

P.S. When are you finally going to start work on the 
manual?
All joking aside... That is an important task!  But I kinda figured it
was less important than getting the software working at all on 
oldworld
hardware, since I seem to be the only one on the list who has oldworld
pmac hardware available for testing.  If there were someone else who
could do the testing part, I'd have time to work on the documentation
part...  Any takers out there?

Rick
I've offered my services before in writing up some docs (or simply
modifying the woody ones to fit the sarge install) in a couple of
previous threads. Meanwhile, I'm running on a G3/266 that I'd be 
willing
to test with. Also, I need help with setting my G3 to send output to a
serial console, since the 2.6.x kernels don't give my voodoo3 card any
console data.

Russell
OK Russell, you're on!
Here's what you need to do:
First, google a bit to find and get copies of the distribution 
files for BootX, miboot, and quik -- the three boot-loaders that 
work on OldWorld PowerMacs.  Read and try to understand the 
documentation that comes with the package distributions.  Most of 
it is sketchy, but if you combine it with more googling for stuff 
in the various mailinglist archives (debian and YellowDog Linux, in 
particular, but also the PowerPC Linux mailing list and any other 
distros that support PowerPC, such as SuSE and Fedora)  and there's 
useful (if anecdotal) stuff in several people's personal home web 
pages as well.

Retrieve and read the Apple Tech notes on Open Firmware.  Start 
with TN1061 and follow pointers from there.  There's also lots of 
useful stuff on the web.  Google for "Open Firmware Apple 
macintosh".  There are also some very useful docs about using 
OpenFirmware with NetBSD.

If you're really dedicated, the first stage should take you a 
couple of weeks.

Second, partition your disk so that you have plenty of free space 
to install test releases of Debian into.  Each installation takes a 
minimum of about 1.5 GB -- more if you want to make it actually 
useful.  So multiply 2-3 GB by the maximum number of test 
installations you intend to make before you wipe the disk and start 
over clean.  On that disk (or another one dedicated to the purpose) 
also set up an HFS partition (*not* HFS-plus -- Debian does not at 
this time support access to HFS-plus filesystems from inside the 
installer) of about 1 GB (more if you want it to be actually useful 
other than as an intermediate boot loader.  If you're going to run 
Toast here, you should allow plenty of space [gigabytes] for 
CD-images).  Install MacOS-9 there.  Then install BootX (both the 
BootX extension and the BootX.app application) according to the 
instructions you got with the BootX distribution.  MacOS-X does not 
support BootX.  (Unfortunately, part of the MacOS_X boot loader is 
called "bootx".  It's not related to the one we are interested in 
here.)

Third, download the latest d-i businesscard iso, and burn it (I use 
Toast) to a CD-RW (don't waste a CD-R on it -- you're probably only 
going to use it a couple of times at most).  Copy the kernel of 
your choice from the CD (install:powerpc:vmlinux or 
install:powerpc:2.4:vmlinux) into the "System Folder:Linux Kernels" 
folder of your MacOS-9 partition, and the initrd.gz file from the 
same place to the "System Folder:Linux Ramdisks" folder.  Invoke 
BootX.app, set the appropriate parameters and let her rip.  Answer 
the questions and file an installation report.

Dig out my previous d-i installation reports for OldWorld PowerPC 
installations from the mailing list archives.  They may give you 
some useful hints.

Fourth, try a floppy disk install.  Contact Sven for instructions 
on where to download the latest floppy images.  Let it try to 
install the quik bootloader, and see if you can figure out how to 
make that work.  If you succeed in this, let me know.  I haven't 
gotten this far yet.

When you're completely familiar with all the various aspects of 
booting, you can start on re-writing a "D-I on OldWorld 
installation manual".

Contact me if you have questions at any point.  I've left out a 
massive amount of detail!

Enjoy!
Rick

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

Re: Floppy install on Oldworld PowerMac

[Rick Thomas]

On Thursday, September 9, 2004, at 01:23 PM, Sven Luther wrote:
On Thu, Sep 09, 2004 at 12:45:29PM -0400, Rick Thomas wrote:
Ummm... The contents of
http://people.debian.org/~luther/d-i/images/daily/powerpc/floppy-2.4/
haven't changed in the last few days.  Is the build process stalled
somewhere?
[DIR] 2004-09-08_RSYNC_IN_PROGRESS/ 07-Sep-2004 22:05  -
[DIR] 2004-09-09_RSYNC_IN_PROGRESS/ 08-Sep-2004 22:05  -
Hmmm.. It seems to be doing it again...
 2004-09-11/   10-Sep-2004 23:13  -
 2004-09-12_RSYNC_IN_PROGRESS/ 11-Sep-2004 22:05  -
 daily/10-Sep-2004 23:13  -
It's been this way for over an hour...
Rick
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: HELP: Failed powerpc autobuild upload again, p.d.o is rejecting my rsync.

[Rick Thomas]

Then it's probably not a hardware problem.
On Sunday, September 12, 2004, at 06:00 AM, Sven Luther wrote:
On Sun, Sep 12, 2004 at 05:10:53AM -0400, Rick_Thomas wrote:
On Sun, 2004-09-12 at 03:06, Sven Luther wrote:
Here is the error in my log :
powerpc/netboot/2.4/vmlinuz-chrp.initrd
powerpc/netboot/2.4/vmlinuz-prep.initrd
Read from remote host people.debian.org: Connection reset by peer
rsync: writefd_unbuffered failed to write 4 bytes: phase 
"unknown": Broken pipe
rsync error: error in rsync protocol data stream (code 12) at 
io.c(839)
Sun Sep 12 05:28:26 UTC 2004

Anyone has an idea on what is going on ?
Whenever I've seen error messages similar to those, it's been 
because an
ethernet interface has gotten confused.  Usually rebooting the 
relevant
machine clears it up (but sometimes only for a while.)  If it 
comes back
after a reboot, it's likely a sign that the interface hardware is
failing and should be replaced.
Err, the box in question is always online, gets my mail and host 
my irc client
under screen. I am thus continously connected to it or something, 
and i don't
see any failure in it. Furthermore, if i do the upload by hand, it 
always
seems to work.

Friendly,
Sven Luther

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

Bug#271417: "Select keyboard" defaults to "European" when "American English" is expected on OldWorld PowerPC

[Rick Thomas]

Package: installation-reports
powerpc BootX 20040911 businesscard OldWorld PowerMac
See Note 2
INSTALL REPORT
Debian-installer-version: 

Index of /pub/cdimage-testing/daily/powerpc/current

 Name   Last modified   Size  Description


 Parent Directory   12-Sep-2004 00:13  -
 MD5SUMS12-Sep-2004 00:13 1k
>>> sarge-powerpc-businesscard.iso 12-Sep-2004 00:07   140M
 sarge-powerpc-netinst.iso  12-Sep-2004 00:13   297M

Apache/1.3.31 Server at cdimage.debian.org Port 80

uname -a: 
	Linux debian 2.6.8-powerpc #1 Fri Aug 27 18:02:26 CEST 2004 ppc 
GNU/Linux

Date: 
3:00 AM US East Coast time Sept 12, 2004
Method: How did you install?  What did you boot off?  If network
  install, from where?  Proxied?
MacOS BootX using the "2.6" kernel and initrd.gz
from the install/powerpc folder on the indicated
businesscard CD image with an assist from the
uchicago "testing" mirror
Machine: 
PowerMac G3/300 MHz

Processor:
processor   : 0
cpu : 740/750
temperature : 36-41 C (uncalibrated)
clock   : 300MHz
revision: 2.2 (pvr 0008 0202)
bogomips: 601.29
machine : Power Macintosh
motherboard : AAPL,Gossamer MacRISC
detected as : 48 (PowerMac G3 (Gossamer))
pmac flags  : 
L2 cache: 1024K unified pipelined-syncro-burst
memory  : 384MB
pmac-generation : OldWorld
Memory:
384 MB
Root Device: 
root is on /dev/hdg10.  Swap is on /dev/hdg8. macOS is on /dev/hdg6
	/dev/hdg
		#type name  length   
base  ( size )  system
	/dev/hdg1 Apple_partition_map Apple 63 
@ 1 ( 31.5k)  Partition map
	/dev/hdg2  Apple_Driver43 Macintosh 54 
@ 64( 27.0k)  Driver 4.3
	/dev/hdg3  Apple_Driver43 Macintosh 74 
@ 118   ( 37.0k)  Driver 4.3
	/dev/hdg4  Apple_Driver_IOKit Macintosh512 
@ 192   (256.0k)  Unknown
	/dev/hdg5   Apple_Patches Patch Partition  512 
@ 704   (256.0k)  Unknown
	/dev/hdg6   Apple_HFS untitled 2097152 
@ 1216  (  1.0G)  HFS
	/dev/hdg7 Apple_UNIX_SVR2 root19531251 
@ 2098368   (  9.3G)  Linux native
	/dev/hdg8 Apple_UNIX_SVR2 swap 1953126 
@ 21629619  (953.7M)  Linux swap
	/dev/hdg9 Apple_UNIX_SVR2 root19531251 
@ 23582745  (  9.3G)  Linux native
	/dev/hdg10Apple_UNIX_SVR2 Root-10  5859376 
@ 43113996  (  2.8G)  Linux native
	/dev/hdg11 Apple_Free Extra   52734377 
@ 48973372  ( 25.1G)  Free space
	/dev/hdg12 Apple_Free Extra  113607707 
@ 206565349 ( 54.2G)  Free space
	/dev/hdg13Apple_UNIX_SVR2 SuZq   104857600 
@ 101707749 ( 50.0G)  Linux native
	
	Block size=512, Number of Blocks=320173056
	DeviceType=0x0, DeviceId=0x0
	Drivers-
	1: @ 64 for 23, type=0x1
	2: @ 118 for 36, type=0x


Root Size/partition table:  Feel free to paste the full partition
  table, with notes on which partitions are mounted where.
Output of lspci and lspci -n:
	:00:00.0 Host bridge: Motorola MPC106 [Grackle] (rev 40)
	:00:0d.0 Ethernet controller: D-Link System Inc RTL8139 
Ethernet (rev 10)
	:00:0e.0 SCSI storage controller: Artop Electronic Corp 
ATP865 (rev 06)
	:00:0f.0 PCI bridge: Hint Corp HB6 Universal PCI-PCI bridge 
(non-transparent mode) (rev 13)
	:00:10.0 ff00: Apple Computer Inc. Heathrow Mac I/O (rev 01)
	:00:12.0 VGA compatible controller: ATI Technologies Inc 3D 
Rage I/II 215GT [Mach64 GT] (rev 9a)
	:01:08.0 USB Controller: NEC Corporation USB (rev 41)
	:01:08.1 USB Controller: NEC Corporation USB (rev 41)
	:01:08.2 USB Controller: NEC Corporation USB 2.0 (rev 02)
	:01:0b.0 FireWire (IEEE 1394): Texas Instruments TSB12LV26 
IEEE-1394 Controller (Link)
	
	:00:00.0 0600: 1057:0002 (rev 40)
	:00:0d.0 0200: 1186:1300 (rev 10)
	:00:0e.0 0100: 1191:0009 (rev 06)
	:00:0f.0 0604: 3388:0021 (rev 13)
	:00:10.0 ff00: 106b:0010 (rev 01)
	:00:12.0 0300: 1002:4754 (rev 9a)
	:01:08.0 0c03: 1033:0035 (rev 41)
	:01:08.1 0c03: 1033:0035 (rev 41)
	:01:08.2 0c03: 1033:00e0 (rev 02)
	:01:0b.0 0c00: 104c:8020

Base System Installation Checklist:
[O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it
Initial boot worked:[o]
Configure network HW:   [?]
Config network: [?]
Detect CD:  [o]
Load installer modules: [o]
Detect hard drives: [o]
Partitio

Bug#271418: After reboot wrong ethernet interface is primary on Oldworld PowerPC Macintosh

[Rick Thomas]

Package: installation-reports
powerpc BootX 20040911 businesscard OldWorld PowerMac
See note 3
INSTALL REPORT
Debian-installer-version: 

Index of /pub/cdimage-testing/daily/powerpc/current

 Name   Last modified   Size  Description


 Parent Directory   12-Sep-2004 00:13  -
 MD5SUMS12-Sep-2004 00:13 1k
>>> sarge-powerpc-businesscard.iso 12-Sep-2004 00:07   140M
 sarge-powerpc-netinst.iso  12-Sep-2004 00:13   297M

Apache/1.3.31 Server at cdimage.debian.org Port 80

uname -a: 
	Linux debian 2.6.8-powerpc #1 Fri Aug 27 18:02:26 CEST 2004 ppc 
GNU/Linux

Date: 
3:00 AM US East Coast time Sept 12, 2004
Method: How did you install?  What did you boot off?  If network
  install, from where?  Proxied?
MacOS BootX using the "2.6" kernel and initrd.gz
from the install/powerpc folder on the indicated
businesscard CD image with an assist from the
uchicago "testing" mirror
Machine: 
PowerMac G3/300 MHz

Processor:
processor   : 0
cpu : 740/750
temperature : 36-41 C (uncalibrated)
clock   : 300MHz
revision: 2.2 (pvr 0008 0202)
bogomips: 601.29
machine : Power Macintosh
motherboard : AAPL,Gossamer MacRISC
detected as : 48 (PowerMac G3 (Gossamer))
pmac flags  : 
L2 cache: 1024K unified pipelined-syncro-burst
memory  : 384MB
pmac-generation : OldWorld
Memory:
384 MB
Root Device: 
root is on /dev/hdg10.  Swap is on /dev/hdg8. macOS is on /dev/hdg6
	/dev/hdg
		#type name  length   
base  ( size )  system
	/dev/hdg1 Apple_partition_map Apple 63 
@ 1 ( 31.5k)  Partition map
	/dev/hdg2  Apple_Driver43 Macintosh 54 
@ 64( 27.0k)  Driver 4.3
	/dev/hdg3  Apple_Driver43 Macintosh 74 
@ 118   ( 37.0k)  Driver 4.3
	/dev/hdg4  Apple_Driver_IOKit Macintosh512 
@ 192   (256.0k)  Unknown
	/dev/hdg5   Apple_Patches Patch Partition  512 
@ 704   (256.0k)  Unknown
	/dev/hdg6   Apple_HFS untitled 2097152 
@ 1216  (  1.0G)  HFS
	/dev/hdg7 Apple_UNIX_SVR2 root19531251 
@ 2098368   (  9.3G)  Linux native
	/dev/hdg8 Apple_UNIX_SVR2 swap 1953126 
@ 21629619  (953.7M)  Linux swap
	/dev/hdg9 Apple_UNIX_SVR2 root19531251 
@ 23582745  (  9.3G)  Linux native
	/dev/hdg10Apple_UNIX_SVR2 Root-10  5859376 
@ 43113996  (  2.8G)  Linux native
	/dev/hdg11 Apple_Free Extra   52734377 
@ 48973372  ( 25.1G)  Free space
	/dev/hdg12 Apple_Free Extra  113607707 
@ 206565349 ( 54.2G)  Free space
	/dev/hdg13Apple_UNIX_SVR2 SuZq   104857600 
@ 101707749 ( 50.0G)  Linux native
	
	Block size=512, Number of Blocks=320173056
	DeviceType=0x0, DeviceId=0x0
	Drivers-
	1: @ 64 for 23, type=0x1
	2: @ 118 for 36, type=0x


Root Size/partition table:  Feel free to paste the full partition
  table, with notes on which partitions are mounted where.
Output of lspci and lspci -n:
	:00:00.0 Host bridge: Motorola MPC106 [Grackle] (rev 40)
	:00:0d.0 Ethernet controller: D-Link System Inc RTL8139 
Ethernet (rev 10)
	:00:0e.0 SCSI storage controller: Artop Electronic Corp 
ATP865 (rev 06)
	:00:0f.0 PCI bridge: Hint Corp HB6 Universal PCI-PCI bridge 
(non-transparent mode) (rev 13)
	:00:10.0 ff00: Apple Computer Inc. Heathrow Mac I/O (rev 01)
	:00:12.0 VGA compatible controller: ATI Technologies Inc 3D 
Rage I/II 215GT [Mach64 GT] (rev 9a)
	:01:08.0 USB Controller: NEC Corporation USB (rev 41)
	:01:08.1 USB Controller: NEC Corporation USB (rev 41)
	:01:08.2 USB Controller: NEC Corporation USB 2.0 (rev 02)
	:01:0b.0 FireWire (IEEE 1394): Texas Instruments TSB12LV26 
IEEE-1394 Controller (Link)
	
	:00:00.0 0600: 1057:0002 (rev 40)
	:00:0d.0 0200: 1186:1300 (rev 10)
	:00:0e.0 0100: 1191:0009 (rev 06)
	:00:0f.0 0604: 3388:0021 (rev 13)
	:00:10.0 ff00: 106b:0010 (rev 01)
	:00:12.0 0300: 1002:4754 (rev 9a)
	:01:08.0 0c03: 1033:0035 (rev 41)
	:01:08.1 0c03: 1033:0035 (rev 41)
	:01:08.2 0c03: 1033:00e0 (rev 02)
	:01:0b.0 0c00: 104c:8020

Base System Installation Checklist:
[O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it
Initial boot worked:[o]
Configure network HW:   [?]
Config network: [?]
Detect CD:  [o]
Load installer modules: [o]
Detect hard drives: [o]
Partitio

Bug#271419: "mesh" SCSI driver should be loaded by default on OldWorld Powermac

[Rick Thomas]

Package: installation-reports
powerpc BootX 20040911 businesscard OldWorld PowerMac
See Note 1 below...
INSTALL REPORT
Debian-installer-version: 

Index of /pub/cdimage-testing/daily/powerpc/current

 Name   Last modified   Size  Description


 Parent Directory   12-Sep-2004 00:13  -
 MD5SUMS12-Sep-2004 00:13 1k
 sarge-powerpc-businesscard.iso 12-Sep-2004 00:07   140M
 sarge-powerpc-netinst.iso  12-Sep-2004 00:13   297M

Apache/1.3.31 Server at cdimage.debian.org Port 80

uname -a: 
	Linux debian 2.6.8-powerpc #1 Fri Aug 27 18:02:26 CEST 2004 ppc 
GNU/Linux

Date: 
3:00 AM US East Coast time Sept 12, 2004
Method: How did you install?  What did you boot off?  If network
  install, from where?  Proxied?
MacOS BootX using the "2.6" kernel and initrd.gz
from the install/powerpc folder on the indicated
businesscard CD image with an assist from the
uchicago "testing" mirror
Machine: 
PowerMac G3/300 MHz

Processor:
processor   : 0
cpu : 740/750
temperature : 36-41 C (uncalibrated)
clock   : 300MHz
revision: 2.2 (pvr 0008 0202)
bogomips: 601.29
machine : Power Macintosh
motherboard : AAPL,Gossamer MacRISC
detected as : 48 (PowerMac G3 (Gossamer))
pmac flags  : 
L2 cache: 1024K unified pipelined-syncro-burst
memory  : 384MB
pmac-generation : OldWorld
Memory:
384 MB
Root Device: 
root is on /dev/hdg10.  Swap is on /dev/hdg8. macOS is on /dev/hdg6
	/dev/hdg
		#type name  length   
base  ( size )  system
	/dev/hdg1 Apple_partition_map Apple 63 
@ 1 ( 31.5k)  Partition map
	/dev/hdg2  Apple_Driver43 Macintosh 54 
@ 64( 27.0k)  Driver 4.3
	/dev/hdg3  Apple_Driver43 Macintosh 74 
@ 118   ( 37.0k)  Driver 4.3
	/dev/hdg4  Apple_Driver_IOKit Macintosh512 
@ 192   (256.0k)  Unknown
	/dev/hdg5   Apple_Patches Patch Partition  512 
@ 704   (256.0k)  Unknown
	/dev/hdg6   Apple_HFS untitled 2097152 
@ 1216  (  1.0G)  HFS
	/dev/hdg7 Apple_UNIX_SVR2 root19531251 
@ 2098368   (  9.3G)  Linux native
	/dev/hdg8 Apple_UNIX_SVR2 swap 1953126 
@ 21629619  (953.7M)  Linux swap
	/dev/hdg9 Apple_UNIX_SVR2 root19531251 
@ 23582745  (  9.3G)  Linux native
	/dev/hdg10Apple_UNIX_SVR2 Root-10  5859376 
@ 43113996  (  2.8G)  Linux native
	/dev/hdg11 Apple_Free Extra   52734377 
@ 48973372  ( 25.1G)  Free space
	/dev/hdg12 Apple_Free Extra  113607707 
@ 206565349 ( 54.2G)  Free space
	/dev/hdg13Apple_UNIX_SVR2 SuZq   104857600 
@ 101707749 ( 50.0G)  Linux native
	
	Block size=512, Number of Blocks=320173056
	DeviceType=0x0, DeviceId=0x0
	Drivers-
	1: @ 64 for 23, type=0x1
	2: @ 118 for 36, type=0x


Root Size/partition table:  Feel free to paste the full partition
  table, with notes on which partitions are mounted where.
Output of lspci and lspci -n:
	:00:00.0 Host bridge: Motorola MPC106 [Grackle] (rev 40)
	:00:0d.0 Ethernet controller: D-Link System Inc RTL8139 
Ethernet (rev 10)
	:00:0e.0 SCSI storage controller: Artop Electronic Corp 
ATP865 (rev 06)
	:00:0f.0 PCI bridge: Hint Corp HB6 Universal PCI-PCI bridge 
(non-transparent mode) (rev 13)
	:00:10.0 ff00: Apple Computer Inc. Heathrow Mac I/O (rev 01)
	:00:12.0 VGA compatible controller: ATI Technologies Inc 3D 
Rage I/II 215GT [Mach64 GT] (rev 9a)
	:01:08.0 USB Controller: NEC Corporation USB (rev 41)
	:01:08.1 USB Controller: NEC Corporation USB (rev 41)
	:01:08.2 USB Controller: NEC Corporation USB 2.0 (rev 02)
	:01:0b.0 FireWire (IEEE 1394): Texas Instruments TSB12LV26 
IEEE-1394 Controller (Link)
	
	:00:00.0 0600: 1057:0002 (rev 40)
	:00:0d.0 0200: 1186:1300 (rev 10)
	:00:0e.0 0100: 1191:0009 (rev 06)
	:00:0f.0 0604: 3388:0021 (rev 13)
	:00:10.0 ff00: 106b:0010 (rev 01)
	:00:12.0 0300: 1002:4754 (rev 9a)
	:01:08.0 0c03: 1033:0035 (rev 41)
	:01:08.1 0c03: 1033:0035 (rev 41)
	:01:08.2 0c03: 1033:00e0 (rev 02)
	:01:0b.0 0c00: 104c:8020

Base System Installation Checklist:
[O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it
Initial boot worked:[o]
Configure network HW:   [?]
Config network: [?]
Detect CD:  [o]
Load installer modules: [o]
Detect hard drives: [o]
Pa

Bug#271421: Printserver task setup defaults to "a4" paper when "letter" is expected.

[Rick Thomas]

Package: installation-reports
powerpc BootX 20040911 businesscard OldWorld PowerMac
see note 5
INSTALL REPORT
Debian-installer-version: 

Index of /pub/cdimage-testing/daily/powerpc/current

 Name   Last modified   Size  Description


 Parent Directory   12-Sep-2004 00:13  -
 MD5SUMS12-Sep-2004 00:13 1k
>>> sarge-powerpc-businesscard.iso 12-Sep-2004 00:07   140M
 sarge-powerpc-netinst.iso  12-Sep-2004 00:13   297M

Apache/1.3.31 Server at cdimage.debian.org Port 80

uname -a: 
	Linux debian 2.6.8-powerpc #1 Fri Aug 27 18:02:26 CEST 2004 ppc 
GNU/Linux

Date: 
3:00 AM US East Coast time Sept 12, 2004
Method: How did you install?  What did you boot off?  If network
  install, from where?  Proxied?
MacOS BootX using the "2.6" kernel and initrd.gz
from the install/powerpc folder on the indicated
businesscard CD image with an assist from the
uchicago "testing" mirror
Machine: 
PowerMac G3/300 MHz

Processor:
processor   : 0
cpu : 740/750
temperature : 36-41 C (uncalibrated)
clock   : 300MHz
revision: 2.2 (pvr 0008 0202)
bogomips: 601.29
machine : Power Macintosh
motherboard : AAPL,Gossamer MacRISC
detected as : 48 (PowerMac G3 (Gossamer))
pmac flags  : 
L2 cache: 1024K unified pipelined-syncro-burst
memory  : 384MB
pmac-generation : OldWorld
Memory:
384 MB
Root Device: 
root is on /dev/hdg10.  Swap is on /dev/hdg8. macOS is on /dev/hdg6
	/dev/hdg
		#type name  length   
base  ( size )  system
	/dev/hdg1 Apple_partition_map Apple 63 
@ 1 ( 31.5k)  Partition map
	/dev/hdg2  Apple_Driver43 Macintosh 54 
@ 64( 27.0k)  Driver 4.3
	/dev/hdg3  Apple_Driver43 Macintosh 74 
@ 118   ( 37.0k)  Driver 4.3
	/dev/hdg4  Apple_Driver_IOKit Macintosh512 
@ 192   (256.0k)  Unknown
	/dev/hdg5   Apple_Patches Patch Partition  512 
@ 704   (256.0k)  Unknown
	/dev/hdg6   Apple_HFS untitled 2097152 
@ 1216  (  1.0G)  HFS
	/dev/hdg7 Apple_UNIX_SVR2 root19531251 
@ 2098368   (  9.3G)  Linux native
	/dev/hdg8 Apple_UNIX_SVR2 swap 1953126 
@ 21629619  (953.7M)  Linux swap
	/dev/hdg9 Apple_UNIX_SVR2 root19531251 
@ 23582745  (  9.3G)  Linux native
	/dev/hdg10Apple_UNIX_SVR2 Root-10  5859376 
@ 43113996  (  2.8G)  Linux native
	/dev/hdg11 Apple_Free Extra   52734377 
@ 48973372  ( 25.1G)  Free space
	/dev/hdg12 Apple_Free Extra  113607707 
@ 206565349 ( 54.2G)  Free space
	/dev/hdg13Apple_UNIX_SVR2 SuZq   104857600 
@ 101707749 ( 50.0G)  Linux native
	
	Block size=512, Number of Blocks=320173056
	DeviceType=0x0, DeviceId=0x0
	Drivers-
	1: @ 64 for 23, type=0x1
	2: @ 118 for 36, type=0x


Root Size/partition table:  Feel free to paste the full partition
  table, with notes on which partitions are mounted where.
Output of lspci and lspci -n:
	:00:00.0 Host bridge: Motorola MPC106 [Grackle] (rev 40)
	:00:0d.0 Ethernet controller: D-Link System Inc RTL8139 
Ethernet (rev 10)
	:00:0e.0 SCSI storage controller: Artop Electronic Corp 
ATP865 (rev 06)
	:00:0f.0 PCI bridge: Hint Corp HB6 Universal PCI-PCI bridge 
(non-transparent mode) (rev 13)
	:00:10.0 ff00: Apple Computer Inc. Heathrow Mac I/O (rev 01)
	:00:12.0 VGA compatible controller: ATI Technologies Inc 3D 
Rage I/II 215GT [Mach64 GT] (rev 9a)
	:01:08.0 USB Controller: NEC Corporation USB (rev 41)
	:01:08.1 USB Controller: NEC Corporation USB (rev 41)
	:01:08.2 USB Controller: NEC Corporation USB 2.0 (rev 02)
	:01:0b.0 FireWire (IEEE 1394): Texas Instruments TSB12LV26 
IEEE-1394 Controller (Link)
	
	:00:00.0 0600: 1057:0002 (rev 40)
	:00:0d.0 0200: 1186:1300 (rev 10)
	:00:0e.0 0100: 1191:0009 (rev 06)
	:00:0f.0 0604: 3388:0021 (rev 13)
	:00:10.0 ff00: 106b:0010 (rev 01)
	:00:12.0 0300: 1002:4754 (rev 9a)
	:01:08.0 0c03: 1033:0035 (rev 41)
	:01:08.1 0c03: 1033:0035 (rev 41)
	:01:08.2 0c03: 1033:00e0 (rev 02)
	:01:0b.0 0c00: 104c:8020

Base System Installation Checklist:
[O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it
Initial boot worked:[o]
Configure network HW:   [?]
Config network: [?]
Detect CD:  [o]
Load installer modules: [o]
Detect hard drives: [o]
Partitio

Bug#271420: Default kernel image is 2.6.7 -- should be 2.6.8 -- on OldWorld PowerPC Mac

[Rick Thomas]

Package: installation-reports
powerpc BootX 20040911 businesscard OldWorld PowerMac
See note 4
INSTALL REPORT
Debian-installer-version: 

Index of /pub/cdimage-testing/daily/powerpc/current

 Name   Last modified   Size  Description


 Parent Directory   12-Sep-2004 00:13  -
 MD5SUMS12-Sep-2004 00:13 1k
>>> sarge-powerpc-businesscard.iso 12-Sep-2004 00:07   140M
 sarge-powerpc-netinst.iso  12-Sep-2004 00:13   297M

Apache/1.3.31 Server at cdimage.debian.org Port 80

uname -a: 
	Linux debian 2.6.8-powerpc #1 Fri Aug 27 18:02:26 CEST 2004 ppc 
GNU/Linux

Date: 
3:00 AM US East Coast time Sept 12, 2004
Method: How did you install?  What did you boot off?  If network
  install, from where?  Proxied?
MacOS BootX using the "2.6" kernel and initrd.gz
from the install/powerpc folder on the indicated
businesscard CD image with an assist from the
uchicago "testing" mirror
Machine: 
PowerMac G3/300 MHz

Processor:
processor   : 0
cpu : 740/750
temperature : 36-41 C (uncalibrated)
clock   : 300MHz
revision: 2.2 (pvr 0008 0202)
bogomips: 601.29
machine : Power Macintosh
motherboard : AAPL,Gossamer MacRISC
detected as : 48 (PowerMac G3 (Gossamer))
pmac flags  : 
L2 cache: 1024K unified pipelined-syncro-burst
memory  : 384MB
pmac-generation : OldWorld
Memory:
384 MB
Root Device: 
root is on /dev/hdg10.  Swap is on /dev/hdg8. macOS is on /dev/hdg6
	/dev/hdg
		#type name  length   
base  ( size )  system
	/dev/hdg1 Apple_partition_map Apple 63 
@ 1 ( 31.5k)  Partition map
	/dev/hdg2  Apple_Driver43 Macintosh 54 
@ 64( 27.0k)  Driver 4.3
	/dev/hdg3  Apple_Driver43 Macintosh 74 
@ 118   ( 37.0k)  Driver 4.3
	/dev/hdg4  Apple_Driver_IOKit Macintosh512 
@ 192   (256.0k)  Unknown
	/dev/hdg5   Apple_Patches Patch Partition  512 
@ 704   (256.0k)  Unknown
	/dev/hdg6   Apple_HFS untitled 2097152 
@ 1216  (  1.0G)  HFS
	/dev/hdg7 Apple_UNIX_SVR2 root19531251 
@ 2098368   (  9.3G)  Linux native
	/dev/hdg8 Apple_UNIX_SVR2 swap 1953126 
@ 21629619  (953.7M)  Linux swap
	/dev/hdg9 Apple_UNIX_SVR2 root19531251 
@ 23582745  (  9.3G)  Linux native
	/dev/hdg10Apple_UNIX_SVR2 Root-10  5859376 
@ 43113996  (  2.8G)  Linux native
	/dev/hdg11 Apple_Free Extra   52734377 
@ 48973372  ( 25.1G)  Free space
	/dev/hdg12 Apple_Free Extra  113607707 
@ 206565349 ( 54.2G)  Free space
	/dev/hdg13Apple_UNIX_SVR2 SuZq   104857600 
@ 101707749 ( 50.0G)  Linux native
	
	Block size=512, Number of Blocks=320173056
	DeviceType=0x0, DeviceId=0x0
	Drivers-
	1: @ 64 for 23, type=0x1
	2: @ 118 for 36, type=0x


Root Size/partition table:  Feel free to paste the full partition
  table, with notes on which partitions are mounted where.
Output of lspci and lspci -n:
	:00:00.0 Host bridge: Motorola MPC106 [Grackle] (rev 40)
	:00:0d.0 Ethernet controller: D-Link System Inc RTL8139 
Ethernet (rev 10)
	:00:0e.0 SCSI storage controller: Artop Electronic Corp 
ATP865 (rev 06)
	:00:0f.0 PCI bridge: Hint Corp HB6 Universal PCI-PCI bridge 
(non-transparent mode) (rev 13)
	:00:10.0 ff00: Apple Computer Inc. Heathrow Mac I/O (rev 01)
	:00:12.0 VGA compatible controller: ATI Technologies Inc 3D 
Rage I/II 215GT [Mach64 GT] (rev 9a)
	:01:08.0 USB Controller: NEC Corporation USB (rev 41)
	:01:08.1 USB Controller: NEC Corporation USB (rev 41)
	:01:08.2 USB Controller: NEC Corporation USB 2.0 (rev 02)
	:01:0b.0 FireWire (IEEE 1394): Texas Instruments TSB12LV26 
IEEE-1394 Controller (Link)
	
	:00:00.0 0600: 1057:0002 (rev 40)
	:00:0d.0 0200: 1186:1300 (rev 10)
	:00:0e.0 0100: 1191:0009 (rev 06)
	:00:0f.0 0604: 3388:0021 (rev 13)
	:00:10.0 ff00: 106b:0010 (rev 01)
	:00:12.0 0300: 1002:4754 (rev 9a)
	:01:08.0 0c03: 1033:0035 (rev 41)
	:01:08.1 0c03: 1033:0035 (rev 41)
	:01:08.2 0c03: 1033:00e0 (rev 02)
	:01:0b.0 0c00: 104c:8020

Base System Installation Checklist:
[O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it
Initial boot worked:[o]
Configure network HW:   [?]
Config network: [?]
Detect CD:  [o]
Load installer modules: [o]
Detect hard drives: [o]
Partitio

Bug#271423: Floppy driver not loaded during install process on OldWorld PowerPC Macs

[Rick Thomas]

Package: installation-reports
powerpc BootX 20040911 businesscard OldWorld PowerMac
see note 6
INSTALL REPORT
Debian-installer-version: 

Index of /pub/cdimage-testing/daily/powerpc/current

 Name   Last modified   Size  Description


 Parent Directory   12-Sep-2004 00:13  -
 MD5SUMS12-Sep-2004 00:13 1k
>>> sarge-powerpc-businesscard.iso 12-Sep-2004 00:07   140M
 sarge-powerpc-netinst.iso  12-Sep-2004 00:13   297M

Apache/1.3.31 Server at cdimage.debian.org Port 80

uname -a: 
	Linux debian 2.6.8-powerpc #1 Fri Aug 27 18:02:26 CEST 2004 ppc 
GNU/Linux

Date: 
3:00 AM US East Coast time Sept 12, 2004
Method: How did you install?  What did you boot off?  If network
  install, from where?  Proxied?
MacOS BootX using the "2.6" kernel and initrd.gz
from the install/powerpc folder on the indicated
businesscard CD image with an assist from the
uchicago "testing" mirror
Machine: 
PowerMac G3/300 MHz

Processor:
processor   : 0
cpu : 740/750
temperature : 36-41 C (uncalibrated)
clock   : 300MHz
revision: 2.2 (pvr 0008 0202)
bogomips: 601.29
machine : Power Macintosh
motherboard : AAPL,Gossamer MacRISC
detected as : 48 (PowerMac G3 (Gossamer))
pmac flags  : 
L2 cache: 1024K unified pipelined-syncro-burst
memory  : 384MB
pmac-generation : OldWorld
Memory:
384 MB
Root Device: 
root is on /dev/hdg10.  Swap is on /dev/hdg8. macOS is on /dev/hdg6
	/dev/hdg
		#type name  length   
base  ( size )  system
	/dev/hdg1 Apple_partition_map Apple 63 
@ 1 ( 31.5k)  Partition map
	/dev/hdg2  Apple_Driver43 Macintosh 54 
@ 64( 27.0k)  Driver 4.3
	/dev/hdg3  Apple_Driver43 Macintosh 74 
@ 118   ( 37.0k)  Driver 4.3
	/dev/hdg4  Apple_Driver_IOKit Macintosh512 
@ 192   (256.0k)  Unknown
	/dev/hdg5   Apple_Patches Patch Partition  512 
@ 704   (256.0k)  Unknown
	/dev/hdg6   Apple_HFS untitled 2097152 
@ 1216  (  1.0G)  HFS
	/dev/hdg7 Apple_UNIX_SVR2 root19531251 
@ 2098368   (  9.3G)  Linux native
	/dev/hdg8 Apple_UNIX_SVR2 swap 1953126 
@ 21629619  (953.7M)  Linux swap
	/dev/hdg9 Apple_UNIX_SVR2 root19531251 
@ 23582745  (  9.3G)  Linux native
	/dev/hdg10Apple_UNIX_SVR2 Root-10  5859376 
@ 43113996  (  2.8G)  Linux native
	/dev/hdg11 Apple_Free Extra   52734377 
@ 48973372  ( 25.1G)  Free space
	/dev/hdg12 Apple_Free Extra  113607707 
@ 206565349 ( 54.2G)  Free space
	/dev/hdg13Apple_UNIX_SVR2 SuZq   104857600 
@ 101707749 ( 50.0G)  Linux native
	
	Block size=512, Number of Blocks=320173056
	DeviceType=0x0, DeviceId=0x0
	Drivers-
	1: @ 64 for 23, type=0x1
	2: @ 118 for 36, type=0x


Root Size/partition table:  Feel free to paste the full partition
  table, with notes on which partitions are mounted where.
Output of lspci and lspci -n:
	:00:00.0 Host bridge: Motorola MPC106 [Grackle] (rev 40)
	:00:0d.0 Ethernet controller: D-Link System Inc RTL8139 
Ethernet (rev 10)
	:00:0e.0 SCSI storage controller: Artop Electronic Corp 
ATP865 (rev 06)
	:00:0f.0 PCI bridge: Hint Corp HB6 Universal PCI-PCI bridge 
(non-transparent mode) (rev 13)
	:00:10.0 ff00: Apple Computer Inc. Heathrow Mac I/O (rev 01)
	:00:12.0 VGA compatible controller: ATI Technologies Inc 3D 
Rage I/II 215GT [Mach64 GT] (rev 9a)
	:01:08.0 USB Controller: NEC Corporation USB (rev 41)
	:01:08.1 USB Controller: NEC Corporation USB (rev 41)
	:01:08.2 USB Controller: NEC Corporation USB 2.0 (rev 02)
	:01:0b.0 FireWire (IEEE 1394): Texas Instruments TSB12LV26 
IEEE-1394 Controller (Link)
	
	:00:00.0 0600: 1057:0002 (rev 40)
	:00:0d.0 0200: 1186:1300 (rev 10)
	:00:0e.0 0100: 1191:0009 (rev 06)
	:00:0f.0 0604: 3388:0021 (rev 13)
	:00:10.0 ff00: 106b:0010 (rev 01)
	:00:12.0 0300: 1002:4754 (rev 9a)
	:01:08.0 0c03: 1033:0035 (rev 41)
	:01:08.1 0c03: 1033:0035 (rev 41)
	:01:08.2 0c03: 1033:00e0 (rev 02)
	:01:0b.0 0c00: 104c:8020

Base System Installation Checklist:
[O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it
Initial boot worked:[o]
Configure network HW:   [?]
Config network: [?]
Detect CD:  [o]
Load installer modules: [o]
Detect hard drives: [o]
Partitio

Bug#271424: Partitioner is checking active swap files!!!

[Rick Thomas]

Package: installation-reports
powerpc BootX 20040911 businesscard OldWorld PowerMac
See note 7
INSTALL REPORT
Debian-installer-version: 

Index of /pub/cdimage-testing/daily/powerpc/current

 Name   Last modified   Size  Description


 Parent Directory   12-Sep-2004 00:13  -
 MD5SUMS12-Sep-2004 00:13 1k
>>> sarge-powerpc-businesscard.iso 12-Sep-2004 00:07   140M
 sarge-powerpc-netinst.iso  12-Sep-2004 00:13   297M

Apache/1.3.31 Server at cdimage.debian.org Port 80

uname -a: 
	Linux debian 2.6.8-powerpc #1 Fri Aug 27 18:02:26 CEST 2004 ppc 
GNU/Linux

Date: 
3:00 AM US East Coast time Sept 12, 2004
Method: How did you install?  What did you boot off?  If network
  install, from where?  Proxied?
MacOS BootX using the "2.6" kernel and initrd.gz
from the install/powerpc folder on the indicated
businesscard CD image with an assist from the
uchicago "testing" mirror
Machine: 
PowerMac G3/300 MHz

Processor:
processor   : 0
cpu : 740/750
temperature : 36-41 C (uncalibrated)
clock   : 300MHz
revision: 2.2 (pvr 0008 0202)
bogomips: 601.29
machine : Power Macintosh
motherboard : AAPL,Gossamer MacRISC
detected as : 48 (PowerMac G3 (Gossamer))
pmac flags  : 
L2 cache: 1024K unified pipelined-syncro-burst
memory  : 384MB
pmac-generation : OldWorld
Memory:
384 MB
Root Device: 
root is on /dev/hdg10.  Swap is on /dev/hdg8. macOS is on /dev/hdg6
	/dev/hdg
		#type name  length   
base  ( size )  system
	/dev/hdg1 Apple_partition_map Apple 63 
@ 1 ( 31.5k)  Partition map
	/dev/hdg2  Apple_Driver43 Macintosh 54 
@ 64( 27.0k)  Driver 4.3
	/dev/hdg3  Apple_Driver43 Macintosh 74 
@ 118   ( 37.0k)  Driver 4.3
	/dev/hdg4  Apple_Driver_IOKit Macintosh512 
@ 192   (256.0k)  Unknown
	/dev/hdg5   Apple_Patches Patch Partition  512 
@ 704   (256.0k)  Unknown
	/dev/hdg6   Apple_HFS untitled 2097152 
@ 1216  (  1.0G)  HFS
	/dev/hdg7 Apple_UNIX_SVR2 root19531251 
@ 2098368   (  9.3G)  Linux native
	/dev/hdg8 Apple_UNIX_SVR2 swap 1953126 
@ 21629619  (953.7M)  Linux swap
	/dev/hdg9 Apple_UNIX_SVR2 root19531251 
@ 23582745  (  9.3G)  Linux native
	/dev/hdg10Apple_UNIX_SVR2 Root-10  5859376 
@ 43113996  (  2.8G)  Linux native
	/dev/hdg11 Apple_Free Extra   52734377 
@ 48973372  ( 25.1G)  Free space
	/dev/hdg12 Apple_Free Extra  113607707 
@ 206565349 ( 54.2G)  Free space
	/dev/hdg13Apple_UNIX_SVR2 SuZq   104857600 
@ 101707749 ( 50.0G)  Linux native
	
	Block size=512, Number of Blocks=320173056
	DeviceType=0x0, DeviceId=0x0
	Drivers-
	1: @ 64 for 23, type=0x1
	2: @ 118 for 36, type=0x


Root Size/partition table:  Feel free to paste the full partition
  table, with notes on which partitions are mounted where.
Output of lspci and lspci -n:
	:00:00.0 Host bridge: Motorola MPC106 [Grackle] (rev 40)
	:00:0d.0 Ethernet controller: D-Link System Inc RTL8139 
Ethernet (rev 10)
	:00:0e.0 SCSI storage controller: Artop Electronic Corp 
ATP865 (rev 06)
	:00:0f.0 PCI bridge: Hint Corp HB6 Universal PCI-PCI bridge 
(non-transparent mode) (rev 13)
	:00:10.0 ff00: Apple Computer Inc. Heathrow Mac I/O (rev 01)
	:00:12.0 VGA compatible controller: ATI Technologies Inc 3D 
Rage I/II 215GT [Mach64 GT] (rev 9a)
	:01:08.0 USB Controller: NEC Corporation USB (rev 41)
	:01:08.1 USB Controller: NEC Corporation USB (rev 41)
	:01:08.2 USB Controller: NEC Corporation USB 2.0 (rev 02)
	:01:0b.0 FireWire (IEEE 1394): Texas Instruments TSB12LV26 
IEEE-1394 Controller (Link)
	
	:00:00.0 0600: 1057:0002 (rev 40)
	:00:0d.0 0200: 1186:1300 (rev 10)
	:00:0e.0 0100: 1191:0009 (rev 06)
	:00:0f.0 0604: 3388:0021 (rev 13)
	:00:10.0 ff00: 106b:0010 (rev 01)
	:00:12.0 0300: 1002:4754 (rev 9a)
	:01:08.0 0c03: 1033:0035 (rev 41)
	:01:08.1 0c03: 1033:0035 (rev 41)
	:01:08.2 0c03: 1033:00e0 (rev 02)
	:01:0b.0 0c00: 104c:8020

Base System Installation Checklist:
[O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it
Initial boot worked:[o]
Configure network HW:   [?]
Config network: [?]
Detect CD:  [o]
Load installer modules: [o]
Detect hard drives: [o]
Partitio

Bug#271419: "mesh" SCSI driver should be loaded by default on OldWorld Powermac

[Rick Thomas]

The following is from the installed system, so the "mesh" driver is 
installed on this system, unlike the installing system before I 
manually did "modprobe mesh".  I don't know if this changes 
anything.

[EMAIL PROTECTED]:~$ ls -l /proc/device-tree
total 9
-r--r--r--   1 root root   4 Sep 13 07:09 #address-cells
-r--r--r--   1 root root   4 Sep 13 07:09 #size-cells
dr-xr-xr-x   2 root root   0 Sep 13 07:09 AAPL,ROM
-r--r--r--   1 root root   4 Sep 13 07:09 AAPL,cpu-id
-r--r--r--   1 root root  12 Sep 13 07:09 AAPL,original-name
dr-xr-xr-x   2 root root   0 Sep 13 07:09 aliases
dr-xr-xr-x   2 root root   0 Sep 13 07:09 chosen
-r--r--r--   1 root root   4 Sep 13 07:09 clock-frequency
-r--r--r--   1 root root  22 Sep 13 07:09 compatible
dr-xr-xr-x   3 root root   0 Sep 13 07:09 cpus
dr-xr-xr-x   2 root root   0 Sep 13 07:09 memory
-r--r--r--   1 root root  16 Sep 13 07:09 model
-r--r--r--   1 root root  12 Sep 13 07:09 name
dr-xr-xr-x   2 root root   0 Sep 13 07:09 offscreen-display
dr-xr-xr-x   2 root root   0 Sep 13 07:09 openprom
dr-xr-xr-x   2 root root   0 Sep 13 07:09 options
dr-xr-xr-x  12 root root   0 Sep 13 07:09 packages
dr-xr-xr-x   7 root root   0 Sep 13 07:09 pci
-r--r--r--   1 root root 256 Sep 13 07:09 pci-OF-bus-map
dr-xr-xr-x   2 root root   0 Sep 13 07:09 perch
[EMAIL PROTECTED]:~$ for i in $( find /proc/device-tree/ -type f | xargs 
grep -l mesh ); do ls -ld $i; cat -v $i; echo; done
-r--r--r--  1 root root 17 Sep 13 07:09 /proc/device-tree/aliases/scsi
/pci/mac-io/mesh^@
-r--r--r--  1 root root 17 Sep 13 07:09 /proc/device-
tree/aliases/scsi-int
/pci/mac-io/mesh^@
-r--r--r--  1 root root 5 Sep 13 07:09 /proc/device-tree/pci/mac-
io/mesh/name
mesh^@
-r--r--r--  1 root root 5 Sep 13 07:09 /proc/device-tree/pci/mac-
io/mesh/compatible
mesh^@
[EMAIL PROTECTED]:~$

Enjoy!
Rick

On Monday, September 13, 2004, at 06:36 AM, Colin Watson wrote:
On Mon, Sep 13, 2004 at 12:45:00AM -0400, Rick Thomas wrote:
Note 1:
This machine has a SCSI Zip drive is on the apple
"mesh" scsi controller.  Before the "discover disks"
phase, I had to go to the F2 console and manually
"modprobe mesh" to get it to recognize the Zip disk.
 Because the mesh driver module was loaded behind
d-i's back (so to speak), d-i didn't know about it,
and as a result, "mesh" wasn't carried forward to
/etc/modules after the reboot. (see note 3)
Many (most?) oldworld PowerMac's have the "mesh"
scsi controller as their *only* (and in any case
*primary*) mass-storage interface.  Failure to load
the mesh driver module will make it impossible for
inexperienced users to install Debian on their
machines.  It seems to me that the mesh driver
should be loaded by default on *all* oldworld
PowerMac machines.
The problem is made more complicated because the
"mesh" chip is on the motherboard, and so doesn't
show up in the output of "lspci".  This only
strengthens the argument for loading the mesh driver
by default.
Not necessarily. Does it show up in the mac-io bus? Send me a 
tarball of
/proc/device-tree if you aren't sure.

--
Colin Watson   
[EMAIL PROTECTED]

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

Bug#271419: "mesh" SCSI driver should be loaded by default on OldWorld Powermac

[Rick Thomas]

Thanks!
I await the fix with baited breath...  (Like the cat beside the 
mouse hole.  <-8)

Enjoy!
Rick
On Monday, September 13, 2004, at 07:49 AM, Colin Watson wrote:
reassign 271419 hw-detect
tags 271419 pending
thanks
On Mon, Sep 13, 2004 at 07:15:17AM -0400, Rick Thomas wrote:
The following is from the installed system, so the "mesh" driver is
installed on this system, unlike the installing system before I
manually did "modprobe mesh".  I don't know if this changes
anything.
/proc/device-tree is exported straight from the firmware; the set of
drivers you have loaded doesn't matter.
[EMAIL PROTECTED]:~$ for i in $( find /proc/device-tree/ -type f | xargs
grep -l mesh ); do ls -ld $i; cat -v $i; echo; done
-r--r--r--  1 root root 17 Sep 13 07:09 /proc/device-tree/aliases/scsi
/pci/mac-io/mesh^@
-r--r--r--  1 root root 17 Sep 13 07:09 /proc/device-
tree/aliases/scsi-int
/pci/mac-io/mesh^@
-r--r--r--  1 root root 5 Sep 13 07:09 /proc/device-tree/pci/mac-
io/mesh/name
mesh^@
-r--r--r--  1 root root 5 Sep 13 07:09 /proc/device-tree/pci/mac-
io/mesh/compatible
mesh^@
[EMAIL PROTECTED]:~$
A fix to autodetect this hardware is in my local tree now waiting for
the Subversion repository to come back up.
Thanks,
--
Colin Watson   
[EMAIL PROTECTED]

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

Bug#270599: Floppy install on Oldworld PowerMac

[Rick Thomas]

On Friday, September 10, 2004, at 05:04 PM, Rick Thomas wrote:
The "ofonlyboot" has not changed.  It reads and inverts the colors 
of the tuxmac, but never switches to text-mode screen from the 
inverted color tuxmac.

The boot floppy reads and switches to the text screen then asks 
for the root floppy, which it reads.  It then asks for language 
(English) and location (US) (but not keyboard layout) then invites 
me to load drivers from a floppy.  I gave it the "root-2" floppy 
and it complained about not being able to find any kernel drivers 
on that floppy.  I chose  and re-executed "load drivers 
from a floppy".  This time I gave it the net-drivers floppy, and 
it was happy.  Still thinking that we wouldn't get any where 
without the root-2 floppy loading (and being a bit bull headed 
anyway) I tried "load drivers from floppy" for the third time, and 
again fed it the root-2.  It complained again about not finding 
any kernel modules.  This time I told it to "continue without 
loading drivers" and to my amazement, it started decoding the 
stuff from the root-2 floppy!  Curioser and curioser!

I think it was at this point that it asked for my keyboard layout, 
and suggested "European" as default, even though I had given it 
every reason to suspect that US-English was my preferred locale.  
I've reported this violation of the principle of least 
astonishment before.

It proceeded then to find my ethernet interface (remember I'd 
loaded the net-drivers floppy earlier) and do DHCP discovery on 
it.  This succeeded, as expected.  When it asked, I chose the 
uchicago mirror as usual, and it loaded the installer-components 
list (I think -- I didn't get the exact words) after which it 
*again* complained about not finding any kernel modules!  I told 
it to continue anyway, and it started downloading and unpacking 
installer components from the uchicago mirror (presumably).

When it got done with that and moved on to the partitioner, it 
couldn't find any of my disks (not my IDE main disk or my SCSI Zip 
disk).  The only IDE think it knew about was the CD-ROM drive.  
Exploring on the F2 console showed that it wasn't just the 
partitioner that was confused.  There was no evidence of IDE or 
SCSI disks in /proc or /dev.  (Same as last time -- no progress on 
that front...)

So I wrapped it up and took a tea break to write this report.
I have to consider the check for kernel modules at inappropriate 
times to be a serious bug...

I tried again with the latest floppies:
Index of /~luther/d-i/images/daily/powerpc/floppy-2.4
 NameLast modified   Size  Description

 Parent Directory26-Aug-2004 21:35  -
 asian-root.img  13-Sep-2004 03:16   1.2M
 boot.img13-Sep-2004 03:17   1.4M
 cd-drivers.img  13-Sep-2004 03:17   1.4M
 net-drivers.img 13-Sep-2004 03:17   1.4M
 ofonlyboot.img  13-Sep-2004 03:17   1.4M
 root-2.img  13-Sep-2004 03:17   1.4M
 root.img13-Sep-2004 03:18   1.3M

Apache/1.3.26 Server at people.debian.org Port 80
No change.
1) The ofonlyboot still doesn't give me a text screen.
2) Loading the root-2 floppy still gives me an error message about 
not finding any driver modules, which I ignore and it loads the 
root-2 floppy anyway.

3) No reasonable combination of choice of mirrors 
(ftp.us.debian.org vs debian.uchicago.edu) and distributions 
(testing vs unstable) give me anything but "no disks found".

4) For what it's worth, at least one combination of mirror and 
distro (I don't remember exactly which -- I *think* it was uchicago 
and testing) complained about not being able to find any driver 
modules (presumably) on the mirror.  But the other combinations 
didn't complain.  (So maybe the uchicago unstable distro does have 
driver modules for the 2.4.27 kernel, but  their testing 
doesn't...  Does that make sense?  Is there any way to check this?)

5) Two of the 2.6 floppy images are still too large to fit on a 
physical 1.44 MB disk.

Any thoughts?
Rick

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

Bug#270599: Testing newoldworld pmac miboot 2.6 floppies

[Rick Thomas]

Sven Luther wrote:
> 
> > Oh, and btw, I'm just installing my oldworld 4400/200 mac with d-i and the
> > daily built 2.4-floppy images (with root.img and root-2.img) and it seems to
> > work fine - right now. I'll keep you informed in another mail.
> 
> BTW, as of tomorrows build, the 2.6 images should be fine also.

The 2.6 images now fit on a physical floppy, so that's good. Unfortunately, the 
resulting floppy doesn't boot on my G3.  It reads and gives me a tux-mac icon, but 
when it gets to the end the screen colors invert, and it just sits there.  No text 
screen.  The boot floppy doesn't eject.  When I eject it manually, feed it the "root" 
floppy, and hit , nothing happens -- specifically, it doesn't start reading 
the root floppy.  This same behavior happens for both the "boot" and "ofonlyboot" 
floppies.

Meanwhile, back at the 2.4 ranch...

The 2.4 boot floppy read, switched to text mode, asked for root, which read, asked for 
language (English), then gave me a blue screen which lasted for more than a minute.  I 
switched to the F2 console, killed 4 processes: "udpkg --configure --force-configure 
countrychooser", two more "countrychooser", and "grep US" [this may be a clue].  Back 
on the main menu on F1 console, I told it to "load drivers" and fed it the "root-2". 
It read that and decoded it, then put me in the country chooser screen (not blue, this 
time) I chose "US".  [possible clue:  There is probably a file (the one the "grep US" 
was looking for) that is on the "root-2" floppy, but is needed by the country chooser, 
so should be on the "root" floppy...]

It asked for an ethernet driver.  The 8139too wasn't listed so I said "none of the 
above" to get it to read the net-drivers floppy.  It did, and things continued 
normally til we got to choose a mirror.  I chose "ftp.us.debian.org" but it didn't ask 
for protocol type or debian version, and when it tried to read stuff, I got the "no 
driver modules" message.  I hit "go back" and re-did the mirror choice (presumably at 
lower priority).  This time it did ask for Debian version.  I said "unstable", and it 
proceeded without problems until it got to the partitioner.

As with previous attempts, the partitioner said "no disks found".  Also as with 
previous attempts, poking around on the F2 console shows that it really hasn't found 
any disks.

Back at the main menu, I changed installer priority to "low", and re-ran detect 
hardware.  It said "unable to load some modules" listing: ide-scsi, ide-mod, 
ide-probe-mod, ide-detect, ide-generic, ide-floppy.  Presumably one of those is needed 
to get it to see my IDE disk.

Just for fun, I did a "modprobe mesh", and re-ran detect hardware.  This time it found 
my SCSI ZIP disk, and offered to partition it for me. I declined and rebooted to write 
up this report.

Hope this helps!

Rick


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

Bug#270599: Testing newoldworld pmac miboot 2.6 floppies

[Rick Thomas]

On Friday, September 17, 2004, at 01:00 PM, Sven Luther wrote:
On Fri, Sep 17, 2004 at 09:42:09AM -0700, Brad Boyer wrote:
 [Lots of stuff about what's where in a beige G3 ...]

Bottom line, Sven, what pieces of information about the G3 do you 
need from me?

Enjoy!
Rick

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

Bug#270599: Testing newoldworld pmac miboot 2.6 floppies

[Rick Thomas]

On Friday, September 17, 2004, at 03:55 AM, Sven Luther wrote:
On Fri, Sep 17, 2004 at 02:31:09AM -0400, Rick Thomas wrote:
Meanwhile, back at the 2.4 ranch...
The 2.4 boot floppy read, switched to text mode, asked for root, 
which read, asked for language (English), then gave me a blue 
screen which lasted for more than a minute.  I switched to the F2 
console, killed 4 processes: "udpkg --configure --force-configure 
countrychooser", two more "countrychooser", and "grep US" [this 
may be a clue].  Back on the main menu on F1 console, I told it 
to "load drivers" and fed it the "root-2". It read that and 
decoded it, then put me in the country chooser screen (not blue, 
this time) I chose "US".  [possible clue:  There is probably a 
file (the one the "grep US" was looking for) that is on the 
"root-2" floppy, but is needed by the country chooser, so should 
be on the "root" floppy...]
Indeed. Do you know the name of the floppy in question ?
I'm not sure what you're asking, but here's the table of contents 
of the directory I got the floppies from...

Index of /~luther/d-i/images/2004-09-16/powerpc/floppy-2.4
 NameLast modified   Size  Description

 Parent Directory26-Aug-2004 21:35  -
 asian-root.img  16-Sep-2004 01:55   1.1M
 boot.img16-Sep-2004 01:55   1.4M
 cd-drivers.img  16-Sep-2004 01:56   1.4M
 net-drivers.img 16-Sep-2004 01:56   1.4M
 ofonlyboot.img  16-Sep-2004 01:56   1.4M
 root-2.img  16-Sep-2004 01:56   1.4M
 root.img16-Sep-2004 01:58   1.2M

Apache/1.3.26 Server at people.debian.org Port 80

It asked for an ethernet driver.  The 8139too wasn't listed so I 
said "none of the above" to get it to read the net-drivers 
floppy.  It did, and things continued normally til we got to 
choose a mirror.  I chose "ftp.us.debian.org" but it didn't ask 
for protocol type or debian version, and when it tried to read 
stuff, I got the "no driver modules" message.  I hit "go back" 
and re-did the mirror choice (presumably at lower priority).  
This time it did ask for Debian version.  I said "unstable", and 
it proceeded without problems until it got to the partitioner.
Yes, this is the infamous 2.6.8 modules not in sarge. I may have a 
solution
for this, but it will need some convincing and work.


Remember, this is the *2.4* floppy set I'm using here.  Does that 
make any difference?




As with previous attempts, the partitioner said "no disks 
found".  Also as with previous attempts, poking around on the F2 
console shows that it really hasn't found any disks.
Ok. We need to know what is your ide controller, and in which udeb 
it is
found, and if discover lists it or not.
It's an "UltraATA 133/100 Pro for Mac" PCI-card, from SIIG, Inc of 
Freemont CA.

It works with kernel 2.4.25 and 2.6.8 installed from a businesscard CD.
Output of lspci and lspci -n:
	:00:00.0 Host bridge: Motorola MPC106 [Grackle] (rev 40)
	:00:0d.0 Ethernet controller: D-Link System Inc RTL8139 
Ethernet (rev 10)
	:00:0e.0 SCSI storage controller: Artop Electronic Corp 
ATP865 (rev 06)
	:00:0f.0 PCI bridge: Hint Corp HB6 Universal PCI-PCI bridge 
(non-transparent mode) (rev 13)
	:00:10.0 ff00: Apple Computer Inc. Heathrow Mac I/O (rev 01)
	:00:12.0 VGA compatible controller: ATI Technologies Inc 3D 
Rage I/II 215GT [Mach64 GT] (rev 9a)
	:01:08.0 USB Controller: NEC Corporation USB (rev 41)
	:01:08.1 USB Controller: NEC Corporation USB (rev 41)
	:01:08.2 USB Controller: NEC Corporation USB 2.0 (rev 02)
	:01:0b.0 FireWire (IEEE 1394): Texas Instruments TSB12LV26 
IEEE-1394 Controller (Link)
	
	:00:00.0 0600: 1057:0002 (rev 40)
	:00:0d.0 0200: 1186:1300 (rev 10)
	:00:0e.0 0100: 1191:0009 (rev 06)
	:00:0f.0 0604: 3388:0021 (rev 13)
	:00:10.0 ff00: 106b:0010 (rev 01)
	:00:12.0 0300: 1002:4754 (rev 9a)
	:01:08.0 0c03: 1033:0035 (rev 41)
	:01:08.1 0c03: 1033:0035 (rev 41)
	:01:08.2 0c03: 1033:00e0 (rev 02)
	:01:0b.0 0c00: 104c:8020

I *think* the driver it needs is aec62xx, but doing "lspci | grep 
aec62xx" on the F2 console during the install show that driver as 
having been loaded.  So I don't kow what to think.  Unless one of 
the "ide-*" drivers it claimed to have not found is the culprit?



Back at the main menu, I changed installer priority to "low", and 
re-ran detect hardware.  It said "unable to load some modules" 
listing: ide-scsi, ide-mod, ide-probe-mod, ide-detect, 
ide-generic, ide-floppy.  Presumably one of those is needed to 
get it to see my IDE disk

Re: No /dev/modem or similar created on PowerPC?

[Rick Thomas]

On Friday, September 17, 2004, at 03:12 PM, Joey Hess wrote:
Russell Hires wrote:
I'm just poking around on my G3/266 and I'm noticing that I don't 
have a
/dev/modem, or any tty that links to it. Does the d-i create such a
device only if you say that you want to install via ppp? This 
could be a
problem if, for example, you want to use your modem to output a serial
console (which is my situation, actually :-) In fact, I can't 
find (and
neither can kppp) which tty connects to my modem.
d-i does not know about modems. I think that pppconfig might set a
/dev/modem link if you choose to use it to install via modem in
base-config, but in general yes, new installs have no /dev/modem link.
Probing for modems is a rather risky business that has been known to
turn off UPSes and do other fun stuff, I don't think the installer 
wants
to go there.

I don't understand what a modem has to do with a serial console BTW.
I think I can clear up a point or two here.
OldWorld Macs have two serial RS-232 ports (actually RS-422, but 
who's counting?) one is marked "Printer" and the other is marked 
"Modem".  In at least some models, the "Printer" port did not have 
the RS-232 modem flow-control signals enabled, but the "Modem" port 
did -- hence the distinction and the naming convention.

Russell is probably *not* talking about using an actual real-live 
modem.  What he's probably talking about is the fact that the Mac 
Open Firmware sometimes uses the "Modem" port as a serial console.  
In order to see the stuff that appears on that line, you need to 
use a serial cross-over cable (also -- confusingly -- called a 
"null-modem" cable) to connect the "Modem" port to another machine 
running a terminal emulator (such as macKermit) or an actual 
real-live terminal from the dark ages.

To answer the question Russell is probably asking:  The "Modem" 
port is called "/dev/ttyS0".  You didn't ask, and you could 
probably figure it out for yourself, but just for completeness, the 
"Printer" port is called "/dev/ttyS1".  Note the upper-case "S".

I've never been able to make it work, but I'm told that when you 
use kernel boot-time command-line arguments to tell the booting 
kernel to send it's messages to the serial console port, you should 
leave off the "/dev".  Thus: "console=ttyS0"

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

Bug#270599: Testing newoldworld pmac miboot 2.6 floppies

[Rick Thomas]

On Fri, Sep 17, 2004 at 09:55:17AM +0200, Sven Luther wrote:
Definitively a bug in discover, could you fill a bug report 
against discover1
with your lspci and lspci -n output ?

You won't get much help out of lspci. These are not PCI devices. 
The macio
chip shows up as one huge PCI device, and the macio layer knows how to
/proc/device-tree/aliases then. And this probably means that there 
is no
chance ever of discover discovering them. Oh well.
Attached is a tar-ball of /proc/device-tree from this machine.  I 
hope it helps!

Rick


device-tree.tar.gz
Description: GNU Zip compressed data

Bug#272310: Floppy install on Oldworld PowerMac G3 tower

[Rick Thomas]

Package: installation-reports
I tried the PowerMac install floppy set from the 18th
   Index of /~luther/d-i/images/2004-09-18/powerpc/floppy-2.4
NameLast modified   Size  Description

 [DIR]  Parent Directory26-Aug-2004 21:35  -
 [   ]  asian-root.img  18-Sep-2004 01:35   1.1M
 [   ]  boot.img18-Sep-2004 01:35   1.4M
 [   ]  cd-drivers.img  18-Sep-2004 01:35   1.4M
 [   ]  net-drivers.img 18-Sep-2004 01:35   1.4M
 [   ]  ofonlyboot.img  18-Sep-2004 01:35   1.4M
 [   ]  root-2.img  18-Sep-2004 01:35   1.4M
 [   ]  root.img18-Sep-2004 01:37   1.2M
   
_

Apache/1.3.26 Server at people.debian.org Port 80

The boot floppy reads OK and calls for the root floppy, which also 
reads OK.
It asks for Language (I gave it English), then the screen started 
blinking.

Switching to the other consoles (opt-F2, -F3, -F4), which are also 
blinking, so it's hard to get any details, it appears that the 
/sbin/debian-installer process is crashing and restarting 
repeatedly.

Enjoy!
Rick

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

No HFS driver, and "change install priority" menu option missing -- and other bugs found while testing 2.4 boot floppies on OldWorld PowerMac

[Rick Thomas]

Package: installation-reports
In addition to the already noted problems with 2.4 PowerPC boot 
floppies, I have two requests for modules to be included on the 
root or root-2:

1) The "change installation priority" menu item should be available 
*very* early in the install process.  Best would be immediately 
after loading the root-2.  Currently, it is not available up thru 
partitioning, at least.

2) The "hfs" and "hfsplus" (MacOS filesystem formats) filesystem 
modules should be available early on.  This would make it much 
easier to save log files to a floppy or a zip disk.  Also, they are 
*required* for support of booting with BootX, so that the kernel 
and initrd can be copied from /target/boot/ to the appropriate 
place on the MacOS partition prior to the reboot.

For the sake of completeness, here's a list of the other things 
I've found that currently don't work about the 2.4 powermac boot 
floppies:

3) Country chooser is called before loading root-2, so it hangs 
trying to do a "grep US" on a file that is on root-2, but should be 
on root.

4) the "ofonlyboot" floppy never switches to text-mode screen.  It 
reads and ejects the boot floppy.  But, since the text mode screen 
never appears, it's impossible to proceed further.  This happens on 
both my test machines, the beige G3 tower, and the 6500.

5) It never finds my disk.  It gets all the way to partitioner 
without loading either the "mesh" driver or the driver for my PCI 
IDE controller card.

For the record, I do not have any of these problems with installing 
via BootX using the latest businesscard CD.  Of course, for 
problems 3 and 4 this is a trivial statement.

Also for the record, these problems occur when using the 
"net-drivers" floppy.  I have not tried to use the "CD-drivers" 
floppy.

I would consider problems 3 and 5 to be "show stoppers".
Life would be *much* easier if 1 and 2 were fixed.
Since the "boot" floppy works on my hardware, I consider problem 4 
to be lower priority than the rest -- though others, for whom the 
"boot" floppy doesn't do the job, may reasonably disagree.

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

NewWorld "Blue & White" Mac Powerpc install -- problems with firewire and aec62xx driver.

[Rick Thomas]

Package: installation-reports
INSTALL REPORT
Debian-installer-version: 

Index of /pub/cdimage-testing/daily/powerpc/current
 Name   Last modified   Size  
Description



 Parent Directory   22-Sep-2004 00:26  -
 MD5SUMS22-Sep-2004 00:26 1k
 sarge-powerpc-businesscard.iso 22-Sep-2004 00:20   153M
 sarge-powerpc-netinst.iso  22-Sep-2004 00:26   310M


Apache/1.3.31 Server at cdimage.debian.org Port 80

uname -a: 
	Linux bluemac 2.6.8-powerpc #1 Fri Aug 27 18:02:26 CEST 2004 
ppc GNU/Linux
	
Date: 

Noon September 22, 2004 US East Coast time
Method: How did you install?  What did you boot off?
If network install, from where?  Proxied?
Booted netinst CD directly
At Yaboot prompt, specified:
"expert-powerpc video=ofonly"
Not proxied
Machine: 
	Apple PowerMac "Blue and White" G3 - Rev A (before IDE chip was fixed)
	
Processor:
	G3 300 MHz
	
Memory:
	384 Mbyte
	
Root Device: 
	/dev/hdc -- 6.4 GB IDE disk on the Mac motherboard IDE controller
	
Root Size/partition table:  Feel free to paste the full partition
  table, with notes on which partitions are mounted where.
	
# mac-fdisk -l /dev/hdc
/dev/hdc
#type name length   
base ( size )  system
/dev/hdc1 Apple_partition_map Apple63 @ 
1( 31.5k)  Partition map
/dev/hdc2Apple_Driver_ATA Macintosh54 @ 
64   ( 27.0k)  Unknown
/dev/hdc3Apple_Driver_ATA Macintosh74 @ 
118  ( 37.0k)  Unknown
/dev/hdc4  Apple_Driver_IOKit Macintosh   512 @ 
192  (256.0k)  Unknown
/dev/hdc5   Apple_Patches Patch Partition 512 @ 
704  (256.0k)  Unknown
/dev/hdc6 Apple_Bootstrap boot 195313 @ 
1216 ( 95.4M)  NewWorld bootblock
/dev/hdc7 Apple_UNIX_SVR2 swap 976563 @ 
196529   (476.8M)  Linux swap
/dev/hdc8 Apple_UNIX_SVR2 root   11421868 @ 
1173092  (  5.4G)  Linux native

Block size=512, Number of Blocks=12594960
DeviceType=0x0, DeviceId=0x0
Drivers-
1: @ 64 for 21, type=0x701
2: @ 118 for 34, type=0xf8ff
# mount
/dev/hdc8 on / type ext3 (rw,errors=remount-ro)
proc on /proc type proc (rw)
sysfs on /sys type sysfs (rw)
devpts on /dev/pts type devpts (rw,gid=5,mode=620)
tmpfs on /dev/shm type tmpfs (rw)
usbfs on /proc/bus/usb type usbfs (rw)
#

Output of lspci and lspci -n:
# lspci
:00:00.0 Host bridge: Motorola MPC106 [Grackle] (rev 40)
:00:0d.0 PCI bridge: Digital Equipment Corporation DECchip 
21154 (rev 02)
:01:00.0 FireWire (IEEE 1394): Texas Instruments 
PCILynx/PCILynx2 IEEE 1394 Link Layer Controller (rev 02)
:01:01.0 IDE interface: Silicon Image, Inc. (formerly CMD 
Technology Inc) PCI0646 (rev 05)
:01:02.0 VGA compatible controller: ATI Technologies Inc 3D 
Rage Pro (rev 5c)
:01:03.0 SCSI storage controller: Adaptec AHA-7850 (rev 03)
:01:04.0 SCSI storage controller: Artop Electronic Corp 
ATP865 (rev 06)
:01:05.0 ff00: Apple Computer Inc. Paddington Mac I/O
:01:06.0 USB Controller: OPTi Inc. 82C861 (rev 10)

# lspci -n
:00:00.0 0600: 1057:0002 (rev 40)
:00:0d.0 0604: 1011:0026 (rev 02)
:01:00.0 0c00: 104c:8000 (rev 02)
:01:01.0 0101: 1095:0646 (rev 05)
:01:02.0 0300: 1002:4749 (rev 5c)
:01:03.0 0100: 9004:5078 (rev 03)
:01:04.0 0100: 1191:0009 (rev 06)
:01:05.0 ff00: 106b:0017
:01:06.0 0c03: 1045:c861 (rev 10)
#
Base System Installation Checklist:
[O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it
Initial boot worked:[o]
Configure network HW:   [o]
Config network: [o]
Detect CD:  [o]
Load installer modules: [o]
Detect hard drives: [e]
Partition hard drives:  [e]
Create file systems:[o]
Mount partitions:   [o]
Install base system:[o]
Install boot loader:[o]
Reboot: [o]
Comments/Problems:

Note 1:
Booting without "video=ofonly" resulted in a scrambled screen.
Note 2:
Had to manually specify "bmac" ethernet interface.
Note 3:
	This machine has two IDE disk drives.  One (6.4 GB) is 
connected to the on-board IDE controller.  The other one (180 GB) 
is connected to a SIIG "Ultra ATA 133/100 PRO for Mac" IDE 
interface PCI card installed.  This card seems to require the 
aec62xx driver, but that driver seems to be broken.  At least, when 
that driver is installed, I get fatal errors in the discover 
hardware section and the partitioner, and I am not able to get 
further.  However, when I refuse to load t

Bug#272967: Acknowledgement (NewWorld "Blue & White" Mac Powerpc install -- problems with firewire and aec62xx driver.)

[Rick Thomas]

I forgot to add:
Note 5:
	Several times during the installation, I got an error message 
telling me that some modules were not loaded because they could not 
be found:

  ide-mod, ide-probe-mod, ide-detect, ide-generic
Note 6:
	I installed the same netinst CD on my beige G3 tower, using 
BootX as boot loader.  This machine also has the SIIG "Ultra ATA" 
controller.  Curiously enough, I did *not* get timeout /lost 
interrupt messages on this install.  Is it possible that the Beige 
G3 somehow declares itself to have a slower bus than the Blue & 
White claims, and this causes the driver to use a slower DMA 
mode -- hence no timeouts?

In other words, the Blue & White is claiming to have a faster PCI 
bus than it really can deliver, but the Beige makes no such 
claims.  The Blue and White pays the price for boasting, the Beige 
reaps the benefits of modesty?

Also, curiously, the Beige G3 had no complaints about missing modules...
Go figure!
Rick

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

Re: No HFS driver, and "change install priority" menu option missing -- and other bugs found while testing 2.4 boot floppies on OldWorld PowerMac

[Rick Thomas]

On Thursday, September 23, 2004, at 11:02 AM, Sven Luther wrote:
On Mon, Sep 20, 2004 at 05:05:58PM -0400, Rick Thomas wrote:
Package: installation-reports
In addition to the already noted problems with 2.4 PowerPC boot
floppies, I have two requests for modules to be included on the
root or root-2:
Ok, actually following up on this from Oldenbourg with an oldworld 
box on
hand.
Glad you finally got an oldworld powermac box to test things on!  (<-8)
I hope Oldenbourg is fun.  It's a bit far for me to travel.

1) The "change installation priority" menu item should be available
*very* early in the install process.  Best would be immediately
after loading the root-2.  Currently, it is not available up thru
partitioning, at least.
Hehe. Well, this is a general thingy. The idea is to change this 
in the kernel
command line with DEBCONF_PRIORITY=medium or low. Maybe i should 
create an
'expert' boot/boot-ofonly floppy set ?
Hmmm... In general, I'd much prefer to solve this problem without a 
multiplication of floppies -- one for each possible boot-time 
option!  It takes several minutes to download and burn a floppy.  
Multiply that by all the options I need to test, and you're talking 
about a serious investment of time.  Can't we come up with some 
more general method of setting boot-time options?

In this particular case, could someone please explain to me the 
logic behind putting the first availability of "change installation 
priority" so late in the game?


2) The "hfs" and "hfsplus" (MacOS filesystem formats) filesystem
modules should be available early on.  This would make it much
easier to save log files to a floppy or a zip disk.  Also, they are
*required* for support of booting with BootX, so that the kernel
and initrd can be copied from /target/boot/ to the appropriate
place on the MacOS partition prior to the reboot.
Yeah, but you have to balance the usefullness of that together 
with the size
limit of the floppies.
I understand the size constraints.  But isn't that the reason why 
we added the "root-2" floppy?  Would adding hfs and/or hfsplus kick 
us over the edge into "root-3" land?


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

Re: No HFS driver, and "change install priority" menu option missing -- and other bugs found while testing 2.4 boot floppies on OldWorld PowerMac

[Rick Thomas]

On Thursday, September 23, 2004, at 06:13 PM, Sven Luther wrote:
I understand the size constraints.  But isn't that the reason why
we added the "root-2" floppy?  Would adding hfs and/or hfsplus kick
us over the edge into "root-3" land?
We could indeed add it to root-2, but i would prefer to get the 
floppy loading
work correctly before i do this rather cosmetic thing working. And 
hfs/hfsplus
without ide/scsi driver is not really all that usefull, isn't it.

OK.  I think see the logic.  I'm not clear on what comes from where 
as far as drivers and install components.  Can you give me a 
general picture?  (Seemingly, "net-drivers" and "cd-drivers" are 
obvious, but maybe not?)  In particular, how are things that 
*aren't* on a particular floppy retrieved, and how does it know 
which of those to retrieve?  Also, is there a general rule for what 
goes on "root", "root-2", one of the "drivers" floppy, or over the 
web?

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

Re: No HFS driver, and "change install priority" menu option missing -- and other bugs found while testing 2.4 boot floppies on OldWorld PowerMac

[Rick Thomas]

On Thursday, September 23, 2004, at 06:45 PM, Sven Luther wrote:
On Thu, Sep 23, 2004 at 06:30:48PM -0400, Rick Thomas wrote:
On Thursday, September 23, 2004, at 06:13 PM, Sven Luther wrote:
I understand the size constraints.  But isn't that the reason why
we added the "root-2" floppy?  Would adding hfs and/or hfsplus kick
us over the edge into "root-3" land?
We could indeed add it to root-2, but i would prefer to get the
floppy loading
work correctly before i do this rather cosmetic thing working. And
hfs/hfsplus
without ide/scsi driver is not really all that useful, isn't it.

OK.  I think see the logic.  I'm not clear on what comes from where
as far as drivers and install components.  Can you give me a
general picture?  (Seemingly, "net-drivers" and "cd-drivers" are
obvious, but maybe not?)  In particular, how are things that
*aren't* on a particular floppy retrieved, and how does it know
which of those to retrieve?  Also, is there a general rule for what
goes on "root", "root-2", one of the "drivers" floppy, or over the
web?
They are not, we need to find out a rule for those. Floppies are 
mainly worked
and tested on x86, which have not the size problem we have.

In root needs to be everything to load the rest of the floppies 
(the floppy
driver and retriever) and the most of the other stuff. That is the only
constraint. We put in root-2 the rest of the non-driver stuff, in 
the net
drivers the network drivers, and in the cd drivers the
ide/scsi/cdrom/disk/filesystem stuff, needed to make the cdrom work and
retrieve more stuff from there.

For the rest of it, it is up to us to take decisions.
Let me see if I've got this right -- please correct me if I've 
misunderstood:

BOOT) The "boot" floppy has the kernel and miboot loader stuff 
(fake "System" that's really a boot-loader, and empty "Finder").  
That's all there is and it pretty much fills up the disk, even with 
a severely stripped-down (anorexic?) kernel.  For what it's worth, 
there is about 169 K left.

ROOT) The "root" floppy has enough on it to talk to the console and 
keyboard (at least as far as being able to tell , the  
key, the arrow-keys, and the  key -- none of which depend on 
locale) set the locale, and finally, load the "root-2" floppy.  
That pretty much fills the floppy.

There's about 160 K left on root of "compressed" space: equivalent 
to [maybe] 570 K of "uncompressed" space assuming a compression 
ratio similar to that of the current contents -- about 1:4.  The 
largest single file is "libc".  After that is /var/lib/dpkg/info, 
/bin/busybox, and /usr/lib/locale.  Together these account for 
about 1/3 of the "uncompressed" space.  The remaining 2/3 is lots 
of small potatoes -- nothing one can point to and say " That's big 
and useless. Let's get rid of it!".

ROOT-2) The "root-2" floppy is a bunch of "udeb"s for all the 
components that wouldn't fit on "root".  [This is a good design, 
but it requires that "root" have everything needed to install a 
udeb, making the "root" floppy even more crowded.]

The root-2 floppy seems to have about 600 K of free space.  Since 
udeb's are already compressed, there's no issue of "compressed" vs 
"uncompressed" space on root-2.

Most of the stuff on root-2 seems to be for setting up the network 
(The largest single file is "nic-extra-modules-2.4.27-powerpc-small-
di.udeb" -- 409 K, over half the total.)  Some of that could, 
theoretically, be moved to the net-drivers floppy, but what would 
be the point?  As long as root-2 is needed at all for overflow from 
root, and there's free space on it, why not use it?  However, it's 
worth keeping this observation in mind if space on root-2 becomes 
tight and net-drivers remains uncrowded.

The only stuff that really is absolutely required to be on root-2 
is that which is (1) not absolutely required to be on root, and (2) 
is needed for *both* a net-install *and* a CD-install.  I'm not 
knowledgeable enough about the details of d-i architecture to tell 
which of the udebs on the present root-2 fits that criterion.  Can 
you give me some clues?

NET-DRIVERS) The "net-drivers" floppy, like root-2, is a collection 
of udeb's. This represents everything (modulo the stuff that goes 
on root-2 "because it's there") that is needed to get to the point 
of being able to download stuff from the web.  This includes 
drivers for *all* sorts of network interfaces (PCI NIC cards, 
wireless PC-card NICs for laptops, serial-IP for people who must 
use modems, etc, etc...).  It also includes things like 
dhcp-client, choose-mirror, software for getting things over the 
web, and so on.  Most of this 

Re: No HFS driver, and "change install priority" menu option missing -- and other bugs found while testing 2.4 boot floppies on OldWorld PowerMac

[Rick Thomas]

On Thursday, September 23, 2004, at 06:13 PM, Sven Luther wrote:
I wrote:
I understand the size constraints.  But isn't that the reason why
we added the "root-2" floppy?  Would adding hfs and/or hfsplus kick
us over the edge into "root-3" land?
We could indeed add it to root-2, but i would prefer to get the 
floppy loading
work correctly before i do this rather cosmetic thing working. And 
hfs/hfsplus
without ide/scsi driver is not really all that usefull, isn't it.
Actually, it's quite useful -- for writing to mac-formatted (HFS) 
floppy disks... as in for saving the log files for later debugging 
when all else has failed.

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

Re: Oldworld Serial Console

[Rick Thomas]

On Friday, September 24, 2004, at 09:17 AM, Gregory Seidman wrote:
In fact, I can attest to using an ordinary Mac printer cable between my
Oldworld Mac running Linux and my dual G4 with a Keyspan adapter and
running MacOS X. I use minicom on the Mac and have set up quik, 
not BootX,
to allow me to have a console over the serial line.

Thanks for the explanation!
Could I prevail upon you to post a detailed step-by-step 
description of how you got quik to let you have a console over the 
serial line?

There are at least a few of us who have been beating our heads 
against this particular wall for some time...

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

Re: No HFS driver, and "change install priority" menu option missing -- and other bugs found while testing 2.4 boot floppies on OldWorld PowerMac

[Rick Thomas]

On Saturday, September 25, 2004, at 04:13 AM, Colin Watson wrote:
On Fri, Sep 24, 2004 at 01:50:34AM -0400, Rick Thomas wrote:
ROOT-2) The "root-2" floppy is a bunch of "udeb"s for all the
components that wouldn't fit on "root".  [This is a good design,
but it requires that "root" have everything needed to install a
udeb, making the "root" floppy even more crowded.]
We're killing root-2; it wasn't a good idea to introduce it 
without good
support for loading it, and everything fits on root and net-drivers if
you juggle things around a bit. Commits will come after the archive
cron.daily run today.
I respectfully disagree.
Hey, I'm just a tester.  You developer guys can do what you want.  
But I predict you'll live to regret this decision.

I think the effort would be better spent in fixing the support for 
loading additional floppies to make it completely open-ended and 
seamless.  It almost works right now -- There are simple 
work-arounds for all existing problems with root-2 as it is today.

Oh well...
Rick

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

Re: No HFS driver, and "change install priority" menu option missing -- and other bugs found while testing 2.4 boot floppies on OldWorld PowerMac

[Rick Thomas]

On Saturday, September 25, 2004, at 04:49 PM, Colin Watson wrote:
Bringing things back into sync with other architectures is worth it on
its own, and reducing the number of floppies required for an
installation was one of the original goals of d-i. After my changes,
there's *loads* of room left for things. We were just being 
inefficient,
that's all.
Glad to hear that you can free up space and still maintain 
functionality.  I look forward to seeing the result!

Enjoy!
Rick
PS - I've often wondered if things would be simpler (maybe *much* 
simpler) if the floppy-boot process just assumed that there will 
always be a CD-rom drive available.  Has anybody else explored that 
possibility?

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

Re: Sarge on OldWorld Mac - No root device

[Rick Thomas]

Hi Duane,
Did you install a 2.6 or 2.4 kernel?  The 2.6 kernel requires an 
initial ramdisk.  The 2.4 kernel (usually) does not.

The kernel installer leaves a kernel and a tailored initrd in the 
/boot partition.  You need to copy them into the appropriate places 
inside the MacOS System folder on your HFS partition.

The way I do this is to go to the "alt-F2" console after the new 
kernel has been installed and just before it reboots. Then I do the 
following:

	modprobe hfs # (see note 1)
	mkdir /MacOS
	mount -t hfs /dev/discs/disc0/part6 /MacOS # (see note 2)
	ls -l /MacOS # (to make sure you've got the right partition mounted!)
	cp /target/boot/vmlinux /MacOS/System\ Folder/Linux\ Kernels/ # 
(see note 3)
	cp /target/boot/initrd.img /MacOS/System\ Folder/Linux\ 
Ramdisks/  # (see note 3)
	sync
	umount /MacOS

Then switch back to the "alt-F1" console and allow the reboot to  
proceed.  This will land you in MacOS with BootX running.  You will 
need to modify the various options to make it pick your new kernel 
and initrd.

Don't worry about the fact that choosing an initrd will take away 
the option to tell it where your root partition is:  The initrd 
that was prepared for you by the kernel install stuff has the right 
code to switch to your real root partition once it has loaded the 
necessary modules, etc.  You just have to point BootX at it and it 
does the rest.


	note 1: You can use "hfsplus" in place of "hfs", if your MacOS 
partition is HFSplus.  I use HFS, personally, because I feel like 
the Linux HFSplus module is still fairly new and untested.  But I'm 
old and paranoid.  You may be braver.

	note 2: The installer uses "devfs" (I think that's the right 
word) device names.  You need to figure out which partition has 
your MacOS system on it, and use that in place of 
"/dev//part6". Shell auto-completion is your friend here!

	note 3: BootX gives you a variety of places to put your kernels 
and initial ramdisk images.  I've used the folder-names that I 
chose when I was getting started.  You should use your own choice, 
of course.

Enjoy!
Rick
On Sunday, September 26, 2004, at 12:28 AM, Duane Cottle wrote:
I successfully installed Sarge using this week's iso CD. I used 
BootX to
access it, as the box won't boot from the CD. After the installation, I
deselected the ramdisk according to README and set root to /dev/hda7
likewise. I get kernel panic and "no root" error. When I use the
ramdisk, the installer begins again.

Has anyone seen this? I prefer BootX over miboot, and yaboot is not an
option on OldWorlds AFAICT.
Thanks for any advice.
Cheers,
Duane
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact 
[EMAIL PROTECTED]


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

Re: Sarge on OldWorld Mac - No root device

[Rick Thomas]

On Sunday, September 26, 2004, at 08:34 AM, Duane Cottle wrote:
I'm going to start another thread concerning the fact that my scsi
hard drive wasn't detected during the entire installation.
You probably have a "mesh" SCSI controller. Almost all OldWorld 
Apple machines had them.

Early on in the installation process -- before the disk/hardware 
discovery phase, switch to the "F2" console and type

modprobe mesh
Then, just before the reboot, switch to the "F2" console again, and 
add the line
	mesh
to the /target/etc/modules file.

I've submitted several bug reports on this topic.  The developers 
know about it, and may fix it sometime.  It's not as easy to fix as 
it sounds, because the mesh controller is not on the regular PCI 
bus, so the normal hardware discovery programs never get a chance 
to see it.

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

Re: floppy root.img sizes

[Rick Thomas]

On Oct 15, 2004, at 11:16 AM, Duane Cottle wrote:
Hi all,
Sorry if this is the wrong list for this...
I have been testing daily sarge floppy install images from
http://people.debian.org/~luther/d-i/images/daily/powerpc/floppy/
for a few weeks on OldWorld Macs.
That is, until October 8. Since then, root.img seems to have swollen to
a size bigger than I can dd onto a floppy.
What's puzzling to me is that the sizes reported by any tool I use to
download the images is less than the ls -l report by around 4.2KB.
Example: in gFTP today's root.img shows 1468006 on the server and
1509557 on my hd.
Hi Duane,
I'm forwarding this to Sven Luther, who generates the floppy images in 
question.  I'm also forwarding this to the "debian-boot" list, where 
there are other folks besides Sven who may be able to help.

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

Bug#276826: PowerPC floppy root.img sizes

[Rick Thomas]

Package: debian-installer
Begin forwarded message:
From: Sven Luther <[EMAIL PROTECTED]>
Date: Sat Oct 16, 2004  05:12:16 AM US/Eastern
To: Rick Thomas <[EMAIL PROTECTED]>
Cc: Duane Cottle <[EMAIL PROTECTED]>, debian-
[EMAIL PROTECTED], [EMAIL PROTECTED], Sven 
Luther <[EMAIL PROTECTED]>
Subject: Re: floppy root.img sizes

On Fri, Oct 15, 2004 at 01:17:45PM -0400, Rick Thomas wrote:
On Oct 15, 2004, at 11:16 AM, Duane Cottle wrote:
Hi all,
Sorry if this is the wrong list for this...
I have been testing daily sarge floppy install images from
http://people.debian.org/~luther/d-i/images/daily/powerpc/floppy/
for a few weeks on OldWorld Macs.
That is, until October 8. Since then, root.img seems to have 
swollen to
a size bigger than I can dd onto a floppy.

What's puzzling to me is that the sizes reported by any tool I use to
download the images is less than the ls -l report by around 4.2KB.
Example: in gFTP today's root.img shows 1468006 on the server and
1509557 on my hd.
Damn it, the root image has again grown out of proportion. Please 
fill a bug
report against debian-installer about this. I will try to have a 
look, but
Joey Hess and maybe Colin Watson would be better placed to fix this,
especially as i will probably not have access to my devel box 
until i put some
order in this room. One month of travelling, and my computer room 
now is
completely crowded. Oh well.

Hi Duane,
I'm forwarding this to Sven Luther, who generates the floppy images in
question.  I'm also forwarding this to the "debian-boot" list, where
there are other folks besides Sven who may be able to help.
Friendly,
Sven Luther
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact 
[EMAIL PROTECTED]


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

Re: Sarge on OldWorld Mac - No root device

[Rick Thomas]

On Monday, September 27, 2004, at 04:40 PM, Joey Hess wrote:
Rick Thomas wrote:
I've submitted several bug reports on this topic.  The developers
know about it, and may fix it sometime.  It's not as easy to fix as
it sounds, because the mesh controller is not on the regular PCI
bus, so the normal hardware discovery programs never get a chance
to see it.
As far as I can tell, it's fixed in an as-yet unreleased version of
hw-detect:
- mac-io bus detection improvements:
  + Detect mesh SCSI controller (closes: #269655, #271419).
  + Detect mace Ethernet controller.
  + Detect mac53c94 SCSI controller.
  + Detect therm_adt746x and therm_windtunnel fan controllers, 
although
only post-reboot for now since those modules aren't in
linux-kernel-di-powerpc-*.
  + Cope with /proc/device-tree/aliases/mac-io pointing to a 
symlink.

Thanks Joey!
When can we expect to see it in our friendly neighborhood "daily" 
CD or boot-floppy?

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

Re: Sarge on OldWorld Mac - No root device

[Rick Thomas]

I'd be perfectly happy (personally) if it only worked in 2.6.  
However, I believe that Sarge d-i has, as one of it's goals, to 
support both 2.4 and 2.6.  Does anybody else on the list know for 
sure?

Rick
On Tuesday, September 28, 2004, at 06:21 AM, Christoph Hellwig wrote:
On Mon, Sep 27, 2004 at 04:40:16PM -0400, Joey Hess wrote:
Rick Thomas wrote:
I've submitted several bug reports on this topic.  The developers
know about it, and may fix it sometime.  It's not as easy to fix as
it sounds, because the mesh controller is not on the regular PCI
bus, so the normal hardware discovery programs never get a chance
to see it.
As far as I can tell, it's fixed in an as-yet unreleased version of
hw-detect:
- mac-io bus detection improvements:
  + Detect mesh SCSI controller (closes: #269655, #271419).
  + Detect mace Ethernet controller.
  + Detect mac53c94 SCSI controller.
  + Detect therm_adt746x and therm_windtunnel fan 
controllers, although
only post-reboot for now since those modules aren't in
linux-kernel-di-powerpc-*.
  + Cope with /proc/device-tree/aliases/mac-io pointing to a 
symlink.
Note that there's a patch on lkml that adds device tables ala
pci,usb,etc.. for macio.  But that's probably useless for you as you'd
want to support 2.4 aswell?

I'd be perfectly happy (personally) if it only worked in 2.6.  
However, I believe that Sarge d-i has, as one of it's goals, to 
support both 2.4 and 2.6 on PowerPC architecture.  Does anybody 
else on the list know for sure?

Rick

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