Re: smartmontools

[John Summerfield]

Gleydson Mazioli da Silva wrote:
You can enable the support (even for motherboards without S.M.A.R.T) support with the 
flag "-e" on the command line. If I'm not wrong, this is the default behaviour of it startup 
script.

(IMHO, the addition of smartmontools in the default debian installation is a good idea for 
most part of default Pentium installations and above).
 

 

And Macs. They've just announced support for OSX so I guess it works on 
Macs under Linux.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Suggestion for choose-mirror

[John Summerfield]

Colin Watson wrote:
On Fri, Jul 16, 2004 at 06:26:20AM +0800, John Summerfield wrote:
 

Colin Watson wrote:
   

Since d-i uses debconf, most questions are pre-answerable like this.
 

Where are the Qs and As documented?
   

Unfortunately I suspect the answer right now might be "the source code";
look for templates files. (Alternatively, look through
/var/lib/cdebconf/ in a running d-i.) This should go in the manual, if
it isn't there already.
 

Oh yuk!
I found a spare box and netbooted it: di-barfed on my proxy (haven't 
bothered with why yet) which I specified in my PXE config.

If I understand this right, if I put all the right stuff into the 
environment, then d-i will install the system I want?

It's pretty cleat that I can't put them all in through PXE - the kernel 
just panicked because I already said too much:
label 26
   kernel images/sarge-2.6/vmlinuz
   append initrd=images/sarge-2.6/initrd.gz vga=6 devfs=mount 
root=/dev/ram initrd_size=2 mirror/country=au mirror/protocol=http 
mirror/http/hostname=debian.test.lan 
mirror/http/proxy=http://192.168.9.4:3128/

wretched bootloaders: today I was grubbing a sick machine , typing stuff 
on the initrd line and wondering
a. Why it didn't work
b. Why I thought it would.

I'm only guessing with my answers. Am I right so far? If I can get all 
the answers into environment vars in this fashion, is it going to work?


--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Partman

[John Summerfield]

Margarita Manterola wrote:
-= This is a code fix.  No string change needed =-
It's kind of dumb to show the "Bootable flag" for certain partition
types (like swap), I think it would be better to not show it when it
makes no difference, so as not to confuse the user unnecesarily.
 

To the best of my knowledge, the bootable flag has no significance 
unless you have a DOS-family MBR.

Why ask at all if there is no choice to be made?
If you're installing GRUB or LILO to the boot sector, there is no choice
If you do _not_ have GRUB or LILO on the boot sector,
   and there is only one possible bootable partition there is no choice
   and there are two or three bootable partitions, one containing grub 
is a safe default.

Getting the right answer without asking the user (who might not know 
anyway) might require reading some boot sectors and inspecting their 
contents.

This, of course, points to it being a question to be answered at the 
time of installing (or not) a boot loader, and not when partitioning.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Partman

[John Summerfield]

Steve Langasek wrote:
On Tue, Jul 20, 2004 at 11:51:25AM +0800, John Summerfield wrote:
 

It's kind of dumb to show the "Bootable flag" for certain partition
types (like swap), I think it would be better to not show it when it
makes no difference, so as not to confuse the user unnecesarily.
 

 

To the best of my knowledge, the bootable flag has no significance 
unless you have a DOS-family MBR.
   

It is also significant on Macintosh partition tables, AIUI.
 

:-(( How could I forget.
I don't know about Macintosh, but definitely Darwin on x86,  at present 
I have a Darwin partition whose booter loops.

The point stands, with amendment: there are circumstances when the 
question need not be asked because there is no choice for the user to make.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Partman

[John Summerfield]

Joshua Kwan wrote:
On Mon, 19 Jul 2004 17:33:48 -0300, Margarita Manterola wrote:
 

I know we are really really close to release time, but I'm worried about
partman usability.  We talked about some of these issues during debconf4,
but I've seen that partman is still the same.
   

The problem is that the concept of partman is so difficult to grasp that
it's hard to make UI changes to it for anyone but Anton Zinoviev...
It's a well engineered concept, just that maybe there should be a README
file explaining all of the little directories and files in any given
partman directory, instead of having to read the entire partman manual in
doc/... Anton?
 

_I_ find it so difficult to use that I prefer fdisk and cfdisk.
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Bug#260225: Sarge installer not recognising partition table for disk larger than 137GB

[John Summerfield]

Rick Thomas wrote:
Controllers that don't believe in disks larger than 137 GB(decimal) 
report any disk larger than that as being exactly 137 GB in size.  
This is probably why cfdisk et al are telling you that your partitions 
go beyond the end of the disk -- as far as they know, the disk ends 
before the beginning of the partition: at 137 GB.

It's a good sign that Windows can see the tail of the disk.  That 
means that the controller is capable of seeing it, even if Linux isn't 
forcing the right mode to make it do so.  It's also a hopeful sign 
that Knoppix can mount the partitions on the tail of the disk.

To see if the problem really lies with parted and/or cfdisk you might 
try them under Knoppix...

Is there a jumper or BIOS setting for the controller (or on the disk 
itself) to tell it to always use large disk mode?  (I don't know what 
the official name for that mode is -- maybe somebody on the list 
knows?)  That would be worth a try.

I think the mode is called LBA48. I carefully avoided disks larger than 
120 Mbytes because I expected problems with them, then I got a 120 Gbyte 
LBA48 drive. My Athlon system required a BIOS upgrade to accommodate it.

I think there are now drives down to 80 Gbytes that are LBA48.
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Suggestion for choose-mirror

[John Summerfield]

Colin Watson wrote:
On Thu, Jul 15, 2004 at 08:40:52PM +0800, John Summerfield wrote:
 

Could choose-mirror be changed so that if certain environment variables 
are set then it uses values from those rather than prompt for them?

II suggest that the proxy information be specified thus:
http_proxy=http://example.com:3128/
   

I expect that you can already do this using
mirror/http/proxy=http://example.com:3128/ at the kernel boot prompt.
 

That works fine. However, when I tried to answer some more Qs that way, 
the Kernel panicked.

Instead, I renamed debian-installer to debian-installer.real and 
front-ended it with my own debian-installer which tried to set some 
environment variables and then exec  debian-installer.real.

Unfortunately, that came unstuck because environment variable names 
aren't allowed to contain slashes.

Is there another way to do this?
Is there a document I should be reading?
I suspect I could take old Q&A and poke them into place, but I also 
suspect that fails some time around the first time strings are unfrozen.


--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Use net-install cd as rescue?

[John Summerfield]

Xavier Leoncini wrote:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Sorry to bother I was wondering if it is possible to use the CD net install 
with the beta 4 as a rescue disk
I had to make a dual boot with some windows and cannot get debian to boot 
back.
 

I needed a rescue CD the other day and the best I had to hand was a 
Fedora Core rescue CD. As I had to resuce a RHL system that seemed 
somwhat approbriate, but it transpired it couldn't recscue either the 
RHL system on hdb or Debian on hda (where the real problem was).

Finally I resorted to grub on a floppy.
grub on the CD would be tee riffic!! If it's bootable, I can boot 
anything with Grub.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Some reflections on the future of partman [was: Partman]

[John Summerfield]

Anton Zinoviev wrote:
Hi!
In general Partman suffers from two things.
First - it is very slow.  I'd say it was designed to be slow because
when I started thinking about it I didn't expect that it will become as
slow as it is now.  Moreover adding additional components to it make it
even slower...
Second - its interface is based entirely on debconf.  This means that
its interface is linear, no button bars, no mouse, no keyboard
shortcuts, no menu bars, etc.
 

That, I think, goes far to make it as hard to use as it is. My primary 
concern, even on my P II test machine, is not speed, its ease of use.

It would be easier to users, I think, to  be able to specify all 
decisions about a partition on a single form, so they can see all the 
info that is required and see at a glance what they've said.

Also, any other paritions already specified should also be visible.
I've not tried LVM or RAID, but I have a feeling I could make parallel 
comments there.

In order to solve the first problem we have two choices:
1. Wait until the computers became faster.  At 2009 when sarge+1 will be
released [;-)] they will be about 5 times faster than now.
2. Rewrite it in some other language.  If we take this approach I would
suggest to package some scripting language (perl, python, guile) as udeb
as making big changes in C programs is a pain.  If I wrote partman in C
there would be no big difference between its first version and its
current version.
 

I don't recall now whether Red Hat's previous installer was written in C 
or Perl: whichever, one of the Anaconda developers swore he'd never 
write another installer in .

I'm not familiar with guile, Python I'm starting on. ESR has some 
interesting comments on Python V Perl at 
http://www.linuxjournal.com/article.php?sid=3882

In order to solve the second problem we have these choices:
1. Add to cdebconf additional types of questions.  This would make the
interface considerably better, but not as nice as the interface of the
other installers.
2. Do not use debconf, but write directly to the screen.  I'd not love
this approach.
3. See how this problem is solved by SuSE - YaST supports both text and
graphical installs, both have very nice look and as far as I know there
is almost no duplicate code (due to text+graphics).  Notice that
recently (and finally) SuSE released YaST as free software (GPL).
 

You should always survey the opposition. Anaconda, Red Hat's GPL 
installer, is written in Python. It does both text-mode and graphical 
installs.

Also, Anaconda's already been ported to Debian.
I have no problem with you looking at YaST2 - only with you ignoring 
Anaconda<.> For that matter I believe Mandrake has some nice tools too, 
though I'm not famiiar with them.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Some reflections on the future of partman [was: Partman]

[John Summerfield]

Helmut Wollmersdorfer wrote:
Anton Zinoviev schrieb:
3. See how this problem is solved by SuSE - YaST supports both text and
graphical installs, both have very nice look and as far as I know there
is almost no duplicate code (due to text+graphics).  

Not everything is good, that looks pretty.
Some weeks ago I tried to install Suse 9.0 on my test machine, 
ordinary Celeron 600 with a 106 GB IDE-HDD. hda1 - hda8 are used by 
other installations. Suse partitioner was unable to create a hda9 und 
did not recognize the full 160 GB disk size. Thus I installed Suse 
under VMware with sarge as host, which was straight ahead. There it 
should live, captured on a single virtual disk.

Didn't I see someone with that self-same problem on Debian in the past 
few days? Might on debian-user.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: my first impressions of your installer

[John Summerfield]

Joey Hess wrote:
This is also by design, and is because security updates are used by
default.
 

well it didn't put any security updates line in my sources.list
   

That's because there is currently no security update source for testing.
When used to install an eventual stable sarge, it will add security
updates lines.
 

I think the location for security fixes has been in place for months. 
Whether it's populated is another matter, but I think there is no reason 
the security lines can't be done now.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: my first impressions of your installer

[John Summerfield]

peter green wrote:
im in england i selected english
the language selection list makes no mention of American English or british
english it just says english
 

I just ran the installer here. The _first_ screen gives a choice of 
Australian English, UK English, US English and English for the rest.

It's fairly recent.
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Bug#261873: Move setserial out of base (?)

[John Summerfield]

J.H.M. Dassen (Ray) wrote:
[I'm not on debian-boot; please respect the Reply-To]
On Thu, Jul 29, 2004 at 13:44:48 +0200, Martin Michlmayr wrote:
 

What really has to happen for this is an upload of debootstrap which
doesn't install it.  I'd very like to see such an upload.  I'd also like
to see "#247906: Please don't install boot loaders" addressed (at least on
mips/mipsel, I cannot speak for the other arches with 100% confidence).
JHM, can you talk to -boot and then prepare an urgency high upload?
   

I've refrained from making any changes to debootstrap's list of packages to
be installed for some time now, as a previous upload of mine making such
changes caused quite some irritation with the d-i folks.
I'll be happy to do an upload, provided there is a clear consensus from the
d-i folks as to which changes to the packages list are to be installed.
So, folks, what changes in the list of packages installed by debootstrap
would you like to see?
 

What I as a _user_ would like to see is one or mor separate lists that I 
can easily hack on. The standard list of packages that gets installed is 
not what I want, and so that entails me installing stuff I don't want 
and then replacing it with my choices.

There's a lot of folk running Postfix, Sendmail and other MTAs. Making 
them install exim makes no sense at all.

Then there's the fact there are some packages I _always_ install - vim, 
less and many others. It should be simple for me to get them all in, 
simple for me to create custom install images instead of having to 
rebuild the whole shebang.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Crude Hack, or Elegant Solution

[John Summerfield]

My effort at answering questions by inserting them into the environment 
failed.

However, I've discovered env2debconf.
With this small change:
sed floppy/sbin/env2debconf \
   -e 's/set -e/set -ex/' \
   -e '/for/ s=\=set;cat /etc/env 2>/dev/null='
I can put a list into /etc/env and the answers become the new defaults.
It seems to me this approach goes a good way to automating the process 
of installing with d-i.

These are values I'm using atm:
ns:/mnt# cat floppy/etc/env
mirror/country=Australia
mirror/protocol=http
mirror/http/hostname=debian.demo.lan
mirror/http/proxy=http://debian.demo.lan:3128/
apt-setup/mirror=ftp.wa.au.debian.org
mirror/suite=testing
apt-setup/country=Australia
apt-setup/directory=/debian
apt-setup/hostname=debian.demo.lan
apt-setup/non-free=true
apt-setup/non-us=true
apt-setup/contrib=true
To some extent I'm guessing what to do: atm the choice of mirror doesn't 
work, but country does.

Questions I'd like answered now:
1. How do I get _my_ choices of install language and keyboard?
2. How do I get these questions marked "seen" so I don't see them?
3. Is there a script I can run to get a consolidated list of questions, 
& acceptable answers? A grep over the templates file goes part of the way,
4. How do I tell d-i to install off debian.demo.lan?

I'm trying to get something close to an automatic install, and that 
means no pauses unless there's an error.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

IA32 - missing disk

[John Summerfield]

The 2.6 kernel from Jul 2 (I think from Sid) doesn't detect my IDE 
drive. According to the log, the IDE modules are missing.
2.4 of the same date does detect it.

I don't have logs to offer, and I'm about to look for newer vmlinuz and 
initrd.gz.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#262200: debian-installer: d-i downloads unneeded packages

[John Summerfield]

John wrote:
Package: debian-installer
Severity: normal
I'm doing a network install via modem.
I've discovered d-i downloading, amongst others, jfs and lvm udebs.
I won't be using either, so this is simply wasted time.
I suggest that downloading and installing optional udebs be deferred
until it's clear they are wanted.
 

atm I'm downloading bits of the 2.4 kernel; the above was with 2.6. I 
note its downloading lots of kernel modules. I booted off a LAN, why not 
justput those modules in the initial ram disk?

AFAIK boot image size is only a problem if booting off floppies.
Or, have a second image that cab be downloaded by tftp. Then when the 
kernel and initrd are loaded by tftp, the installer's bootstrap can get 
a second image. The DHCP server can tell which server to use.


--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/

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

Re: IA32 - missing disk

[John Summerfield]

John Summerfield wrote:
The 2.6 kernel from Jul 2 (I think from Sid) doesn't detect my IDE 
drive. According to the log, the IDE modules are missing.
2.4 of the same date does detect it.

I don't have logs to offer, and I'm about to look for newer vmlinuz 
and initrd.gz.

The problem does not occur the Jul 29 version from Sid.
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Progress meter unbelievable

[John Summerfield]

I'm network-installing on an HP Vecra by modem.
Since this is my first (successful so far) install using the Jul 29 Sid 
version, it needed to pull stuff through my modem instead of  of my 
Squid cache.

During most of the time to download the kernel, the progress meter was 
stuck on 87%.

If I'd not been caught on this before, I'd probably have reset the computer.
As it was, I got pretty restless and went investigating to check that it 
was working.


--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Progress meter unbelievable

[John Summerfield]

Sven Luther wrote:
On Fri, Jul 30, 2004 at 05:30:24PM +0800, John Summerfield wrote:
 

I'm network-installing on an HP Vecra by modem.
Since this is my first (successful so far) install using the Jul 29 Sid 
version, it needed to pull stuff through my modem instead of  of my 
Squid cache.

During most of the time to download the kernel, the progress meter was 
stuck on 87%.
   

It is downlaoding stuff in the background.
 

As my infestigations showed.
If I'd not been caught on this before, I'd probably have reset the computer.
As it was, I got pretty restless and went investigating to check that it 
was working.
   

Maybe a good idea would be to add some description of the actual task
happening below, and also the transfer rate, or something such for actual
downloads.
 

The standard apt-get report would be fine.
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Stage 1 info not carried forward

[John Summerfield]

There may be more, not these stand out:
On the boot commandine I specified vga=6
This should be incorporated into the grup config.
For the stage 1 install, I specified a proxy. This should be carried 
forward into the apt config.
atm I'm wondering how to get stuff into it. apt-get says "Connect 113 No 
route to host" even though I can ping the remote site.

This seems to have got me going:
export http_proxy==http://192.168.9.4:3128/
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#262200: debian-installer: d-i downloads unneeded packages

[John Summerfield]

Colin Watson wrote:
On Fri, Jul 30, 2004 at 01:46:37PM +0800, John Summerfield wrote:
 

atm I'm downloading bits of the 2.4 kernel; the above was with 2.6. I 
note its downloading lots of kernel modules. I booted off a LAN, why not 
justput those modules in the initial ram disk?

AFAIK boot image size is only a problem if booting off floppies.
   

If you want an initrd with everything, use the monolithic initrd; you
can probably manage to netboot it. The netboot image will stay small.
 

I went looking to see if I could find this monolithic initrd. I see a 
tarball in the sid installer direcory I've been using. Is this what you 
mean?

Looking at http://www.debian.org/devel/debian-installer/  I see a link 
to CD imaages. Those are out of the question, I'm not going to try to 
pull those through my modem.

The "other boot images" link actually points to the port status page. 
While there seem to be pointers to some more images there, there are 
none appropriate for me. I don't now see the link to Joey's dailies.

I'm pulling from Sid because I can get it locally.
d-i downloads the components you mentioned because, at the moment, it
only gets one chance at downloading components, and therefore downloads
everything it needs to supply e.g. all the partitioning options we want
to make available.
 

What suddenly makes it hard to download stuff? The network's still 
there. Is this a design issue?

 

Or, have a second image that cab be downloaded by tftp. Then when the 
kernel and initrd are loaded by tftp, the installer's bootstrap can get 
a second image. The DHCP server can tell which server to use.
   

Wouldn't it be easier just to have a local mirror of the necessary
udebs?
 

Not for occasional use, and certainly not while you're changing them so 
often, any mirror is soon out of date.

Hopefully I'll have adsl at home again soon and I won't care. Telstra 
said our new home is too far from the exchange, now it's change its 
mind. However, it will be a continuing problem for others.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/

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

Automatic allocation of swap

[John Summerfield]

I have just installed another system, and took the time to try to 
familiarise myself with the partitioning tool.

I was running a 2.6 kernel, and the install kernel's date is Jul 29 06:24.
The install target was a Pentium II, 350 Mhz, 64 Mb RAM and 3.2 Gbytes 
of disk.

It seemed to me that the setup chosen for "multiuser" was entirely 
impractical. Unfortunately, I was more interested in testing other 
aspects, so I just wiped that one and tried some others.

One aspect of that configuration I do recall.
It allocate about 192 Mbytes of swap partition, right at the edge of the 
disk.

I have never thought swap partitions on single-disk systems are a good 
idea, and here is why.

1. Performance
If there is little swapping, then any perfoemance benefit is immaterial.
If swapping is severe (the system is thrashing), there is no good 
alternative to more RAM. RAM is cheap.
That aside, the position of the swap area (in the d-i configuraton as 
in most others) ensures the swap area is far from the data, ensuring 
that operations such a opening an OOo document will send the disk heads 
seeking far, back and forth, back and forth, back and forth, across 
much, maybe most, of the disk surface.

In between, of course, is inbetween: bad sometimes, not so bad at others.
In contrast, if I create a swap file, then it there is some prospect it 
will be near at least some of the data I need to read.

2 Flexibility.
The common Rule of Thumb (ROT) has the appropriate size of swap being 
twice the amount of RAM installed (not three times!).

My own experience is that, mostly, system performance is pretty terrible 
before it's used all the swap area. The sole counterexample I can think 
of is using rsync to do backups. rsync can use enormous amounts of 
virtual memory backing up whole disks, but its working set remains modest.

The stupidity of this ROT is illustrated when adding more RAM. If my 
system is working moderately well with xMbytes of RAM and I, noting that 
RAM is cheaper than formerly, decide to add 2x Mbytes of RAM, should I 
also treble the amount of swap? Of course not, I may well decide I have 
better uses for that disk space.

Either way, the amount of swap is wrong, and recovering or enlarging the 
amount of swap  in a swap partition is not a trivial undertaking.

Finally, the point that I installed a 2.6 kernel is an important one. 
One of the changes I note in the new kernel is that there is no 
performance benefit to using swap partitions.

I recommend that, if there is one disk, a swap file be created rather 
than a swap partition.

And, if /home is a separate partition, then it should be on that 
partition. That is where most I/O activity is likely to be on 
single-user systems.


--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Automatic allocation of swap

[John Summerfield]

Osamu Aoki wrote:
Hi,
On Sun, Aug 01, 2004 at 09:04:50PM +0800, John Summerfield wrote:
 

I have just installed another system, and took the time to try to 
familiarise myself with the partitioning tool.

I was running a 2.6 kernel, and the install kernel's date is Jul 29 06:24.
The install target was a Pentium II, 350 Mhz, 64 Mb RAM and 3.2 Gbytes 
of disk.

It seemed to me that the setup chosen for "multiuser" was entirely 
impractical. Unfortunately, I was more interested in testing other 
aspects, so I just wiped that one and tried some others.

One aspect of that configuration I do recall.
It allocate about 192 Mbytes of swap partition, right at the edge of the 
disk.
   

Hmm.  put it to the edge may not be the best thing to do ... 
Does this have any real impact?
 

For sure. Think about the seek distances. Mostly, in drive specs you see 
'minimum seek" and "average seek."

The latter is a bit vauge, but IBM (at least in the 70s) caclulates it 
over some benchmark workload: it's not the seek time for half the disk.

By spreading data over multiple paritions with great gobs of free space  
between small (after install) amounts of data, you're forcing longer seeks.

 

I have never thought swap partitions on single-disk systems are a good 
idea, and here is why.

1. Performance
If there is little swapping, then any perfoemance benefit is immaterial.
   

So do not complain swap is not at the center.
 

Then you also  don't need any performance benefits of swap partitions 
over swap files.

 

If swapping is severe (the system is thrashing), there is no good 
alternative to more RAM. RAM is cheap.
   

That is different problem. Swap will not be used unless RAM is used up.
 

Indeed, but the system performance will become truly appalling before it 
should.

That aside, the position of the swap area (in the d-i configuraton as 
in most others) ensures the swap area is far from the data, ensuring 
that operations such a opening an OOo document will send the disk heads 
seeking far, back and forth, back and forth, back and forth, across 
much, maybe most, of the disk surface.
   

Then buy a RAM.  Optimization you get with swap location is small.
 

In between, of course, is inbetween: bad sometimes, not so bad at others.
In contrast, if I create a swap file, then it there is some prospect it 
will be near at least some of the data I need to read.
   

I do not understand...
 

Here is a hypothetical layout
/boot
/
/usr
/var
/home
swap
Here, the swap parition is in the wrong location. If the system uses 
synchronous writes to /var (think logs, mail, I guess news) then the 
disk heads will often be somewhere in /var.

If the system has a moderate swapping load, then the heads will be 
travelling clear over /home, maybe twice for each read and write.

I don't know how Linux filesystems allocate space, but likely it's
a) Prefers one edge of the parition
b) Prefers to spread data over the partition
Some may prefer to fill holes, but in the initial state those holes do 
not exist.

If we reverse the order of swap and /home above, we have a better 
situration.

I prefer the use of a swap file. If there's  half-a-gig of data in a 
ten-gig partition, the swap area is right there with the data, as best I 
can do i. Not somewhere probably far from the data.

There's also the problem of just where the major I/O activity will be. 
For tranditional servers, it will be in /var. For interactive workloads, 
/home is more probable. However, a peecee with 50 logged-in users 
running X applications is still a server! For such a workload, a swap 
file in /home might be better than one in /var or in its own partition.

Here is my desktop pc. It's performance sucks bunnies through capilliary 
tubes:
[EMAIL PROTECTED]:~/d-i$ /sbin/swapon -s
FilenameTypeSizeUsed
Priority
/var/swapfile   file524280  455604  -1
/var/swapfile2  file524280  69296   -2
[EMAIL PROTECTED]:~/d-i$ df -lh
FilesystemSize  Used Avail Use% Mounted on
/dev/hda3  29G   27G  639M  98% /
tmpfs 189M 0  189M   0% /dev/shm
/dev/hda1  23M   21M  1.4M  94% /boot
[EMAIL PROTECTED]:~/d-i$

Do you think performance would be improved by forcing the disk to seek 
to the edge of the disk for each access to swap?

I don't.
Two swap files on one disk might not be ideal, but for sure it was easy 
to do, and I didn't have to take the machine down to add more.

 

2 Flexibility.
The common Rule of Thumb (ROT) has the appropriate size of swap being 
twice the amount of RAM installed (not three times!).
   

I thought "at least 2 X RAM".  So 3 X is OK.  Besides, it is less than
10% of disk space.  Who cares lost space.  You can ajust it later if you
care.
 

Most systems I use get bogged down well before the

Re: debian-boot

[John Summerfield]

I thought this:

Content-Type: multipart/alternative; boundary="0-1720438319-1091437058=:65424"
fails Debian list guidelines.



Not quite, it's HTML that's banned.
http://www.debian.org/MailingLists/

I note too that English is mandated and that implies 8-bit character sets.

Do others think a bug report is in order?



-- 

Cheers
John

-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/


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

Re: Automatic allocation of swap

[John Summerfield]

Osamu Aoki wrote:
On Mon, Aug 02, 2004 at 09:21:05AM +0800, John Summerfield wrote:
 

Osamu Aoki wrote:
   

On Sun, Aug 01, 2004 at 09:04:50PM +0800, John Summerfield wrote:
 

By spreading data over multiple partitions with great gobs of free space  
between small (after install) amounts of data, you're forcing longer seeks.
   

Yes.  We all know this.  Let's not argue over belief.
If you want us to change something, few ways to do this.
1. clear facts (bench mark result of different configurations)
2. reference some authoritative documents.
and patch always help.  Really, I did not say putting swap at the edge
of disk is a good thing.  I was annoyed by calling this "stupid".
Please do not call action of DD stupid even if they are.
 

1. You suppose I'm capable of creating a patch. In fact, my C skills 
almost extend far enough that I can read some code.

2. Belief? It seems so obvious to me, it takes me longer to walk across 
two streets than it does one.

In the 70s I was a systems programmer on working on IBM mainframes 
running OS/VS . One of the tasks we took very seriously was the 
minimising of head movement on our disk drives. So far as I know, the 
laws of physics that describe their behaviour have not been repealed.

3. I've reread what I said. You misquoted me. I described the RULE OF 
THUMB as stupid. I don't believe it originated with the DDs.

However, as a ROT, blind adhearance to ROTs is stupid.
Understand the ROT and you will understand its limitations. I rather 
think that ROT was intended to describe an upper limit. Later 
interpretations, like intermpretations of Murphy's Law deviate from the 
truth.


Here is my desktop pc. It's performance sucks bunnies through capilliary 
tubes:
[EMAIL PROTECTED]:~/d-i$ /sbin/swapon -s
FilenameTypeSizeUsed
Priority
/var/swapfile   file524280  455604  -1
/var/swapfile2  file524280  69296   -2
[EMAIL PROTECTED]:~/d-i$ df -lh
FilesystemSize  Used Avail Use% Mounted on
/dev/hda3  29G   27G  639M  98% /
   

 ^
 |
 no wonder, sigh.
 

tmpfs 189M 0  189M   0% /dev/shm
/dev/hda1  23M   21M  1.4M  94% /boot
[EMAIL PROTECTED]:~/d-i$
   

...
 

That's almost certainly bad. You probably don't spend much time reading 
program files and documentation in comparison with the time reading and 
writing variable data.
   

I have been reading and editing documentation a lot.  Please install
debian-reference-en package from Sarge :-)  Then you know what.
http://www.debian.org/doc/manuals/debian-reference/reference.en.html
Oh, please read install guide too.  That has good amount of partitioning
etc. too.  

http://www.debian.org/releases/stable/installmanual
 

Well, I was really referring to your (plural - many DDs) computers.
 

You're a software developer writing in C? Would you say, less than a 
second to read gcc then a minute or so for a compile? Sure, it depends, 
but the ratios will be somewhat like those.
   

I usually do not write C nor compile it much.  (With my old 386, I used
to compile my kernel or patched-X for a day, though.)
Really, Linux or Windows, it is bad idea to fill actively used disk up
to 98%.  (For your case / partition.  /boot may be OK since it is
practically read-only.)  For Linux, 90-95%, for windows 60-70% is my
common sense usage.
 

I agree with that. Breaking the disk into lots of partitions makes it 
more likely that one will fill andcause problems. Recently I installed a 
server running Sarge and I find / is overfilled to the point I have 
problems installing packages:
ns:~# df -t ext3 -m
Filesystem   1M-blocks  Used Available Use% Mounted on
/dev2/root2186   176 1 100% /
/dev/sda1   321813  58% /boot
/dev/sda3 3024   692  2179  25% /usr
/dev/sda4 5378  4714   391  93% /var
/dev/sdb117241 12973  3394  80% /home/users
/dev/hda237572 28382  7664  79% /mnt/rhl
/dev/hdb276905 47356 27987  63% 
/mnt/rhl/var/ftp/pub/linux
ns:~#

Cheers,
Osamu
PS: FYI
  None of these habits should be carried over to Linux and ext2. Linux
  native file systems do not need defragmentation under normal use and this
  includes any condition with at least 5% of free space on a disk. There is
  a defragmentation tool for ext2 called defrag, but users are cautioned
  against casual use. .From: Linux Partitioning mini-FAQ 
 http://pw1.netcom.com/~kmself/Linux/FAQs/partition.html
 

I don't know why you're telling me Linux is better than Windows. I've 
never been a regular Windows user: I us

Re: Introduction

[John Summerfield]

Greg Folkert wrote:
Personally, I use: CTRL+L to reply to list.
 

Thunderbird lacks that option.
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Introduction

[John Summerfield]

Greg Folkert wrote:
Personally, I use: CTRL+L to reply to list.
 

Thunderbird lacks that option.
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Introduction

[John Summerfield]

Matthew A. Nicholson wrote:
John Summerfield wrote:
Greg Folkert wrote:
Personally, I use: CTRL+L to reply to list.
 

Thunderbird lacks that option.
Use the reply all in thunder bird just like I just did. (Ctrl-Shift-R)
Don't forget to prune my address. One's enough.
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Introduction

[John Summerfield]

Steve Langasek wrote:
On Tue, Aug 03, 2004 at 09:08:37AM +0700, Rahmat M. Samik-Ibrahim wrote:
 

On Tue, 2004-08-03 at 04:12, Colin Watson wrote:
   

 

Google for "Reply-To considered harmful".
 

 

Nevertheless: Smoking is considered harmful, Fat is harmful too,
Cholesterol is harmful. But, what if the community prefer those
harmful things?
   

"The community" does not prefer Reply-To.
 

Where are the latest voting results on the matter?
Is voting done per-list, or overall?

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: D-I impressions

[John Summerfield]

Ryan Underwood wrote:
- Is is possible to include memtest86 on the install CD as a boot option?
 One of the first things I do when installing a new machine is to check
 that everything is sane.  In this case, Linux crashed when booting the
 first time.  When I made a memtest floppy, I found that one of the
 memory sticks was seated poorly and causing memory errors (probably
 what caused the machine to be given away).  It would have been nice to
 simply do this from the syslinux menu.
 

I've recommend to  a couple of people on debian-user in the past few 
days that they run memtest86. On the install CD would be a really 
excellent place for it to be.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: D-I impressions

[John Summerfield]

Glenn McGrath wrote:
On Tue, 3 Aug 2004 00:07:16 -0500
Ryan Underwood <[EMAIL PROTECTED]> wrote:
 

- Should ntpdate be included as part of the base install? 
pool.ntp.org
 is rather reliable, and it helps to have one's clock in a sane state
 so that tar does not complain about timestamps being in the future,
 and things like that.
   

rdate would be simpler, busybox has an rdate which is very small.
 

ntp servers abound. rdate servers? Mostly private I suspect. Even my 
Billion routers support ntp.


--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Introduction

[John Summerfield]

Greg Folkert wrote:
On Mon, 2004-08-02 at 22:43, John Summerfield wrote:
 

Greg Folkert wrote:
   

Personally, I use: CTRL+L to reply to list.

 

Thunderbird lacks that option.
   

So did Evo. I added the CTRL+L to do the reply to list.
I really thought T-Bird had a function for it.
/me looks...
/me doesn't find it, much to his chagrin.
 

Further research:
Moz doesn't.
Kmail does. However, kmail's been a little fragile recently. there are 
other reasons to prefer kmail over the others than "reply to list" and 
"reply to author," but atm I don't.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Boot problems with grub and 4GB+

[John Summerfield]

Petter Reinholdtsen wrote:
[Peter Green]
 

or just default to lilo altogether
unless there is some other major advantage to grub ofc
   

There are major advantages with grub.  It understand several file
systems (no need to update the boot block if the kernel or initrd is
updated), and it contains a simple editor (so one can fix the boot
parameters if the current ones are busted).
There are also minor advantages, but I leave them out here.
 

like you can have a completely broken (or not) configuration and still 
boot the system. A week or so my recovery CD proved not to work on the 
system needing recovery and I finally fixed it with nothing more than a 
grub floppy.


--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: release status

[John Summerfield]

Joey Hess wrote:
Joey Hess wrote:
 


   

I guess anyone not on my home network would prefer this url:

 

A partcular concern I have is the reason for so many Apache packages. 
I've never discovered documentation as to why there are apache, 
apache-ssl and apache-perl, and how to decide which one(s) to install.

I set out to find whether the new manual provides that kind of 
information, or pointer to where it may be found.  In the course of my 
search I have discovered:
a)  http://d-i.alioth.debian.org/manual/en.arm/apas03.html contains links to
a.1 The Multi Disk HOWTO 
 which is quite old and 
seriously out of date. I doubt whether any but the most frugal would be 
satisfied, for example, with the partitions sizes recommended on 
http://www.tldp.org/HOWTO/Multi-Disk-HOWTO-10.html#ss10.1. If nobody 
wants to take on the task of revising it, I suggest the link be dropped.
a.2 A link to Partitioning Strategies 
 
which gets a 404.
b) No manuals for PowerPC or s390.


--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Opinion about sarge install

[John Summerfield]

Andrew Pollock wrote:
On Sun, Aug 08, 2004 at 03:01:45PM -0300, [EMAIL PROTECTED] wrote:
 

Hey folks,
Just some opinion about the sarge install process. I installed a july daily
build(I can't remember the exact one, but i think it was 25) on my computer
that has just the very common hardware(i386 arch) with the 2.6 kernel. I
consider myself a getting-to-intemediate-level user and here's some things you
might consider (or not :P):
1)X configuration should be more automatic. I couldn't get the x server to work
on the first try. So I just did some copy and paste for the config file I
backedup from my old Mandrake distro I replaced with debian. If  Mandrake can,
why can't debian get the configuration right with not much user intervention
and technical knowledge? These would make things much easier;
   

It would probably be interesting and useful for the X packaging team to see
a diff between the config that was generated from configuring X and the
config you ultimately had to use to get X working. In my experience,
provided you give the X configuration the right input, it does a reasonable
job of getting X working.
 

I have a DDC2-compliant Sun monitor on my peecee. I've not yet tried d-i 
on it, but there are no questions It should have to ask.


2) I select the desktop package collection and then later when I was trying to
install some software I found out make, gcc and other tools were not installed.
I think these tools should be installed on any install because unfortunetelly
not everything come on .deb packages from a organized  repository;
   

This is a matter of opinion. Not everyone wants a compiler installed by
default. Firewalls don't need a compiler. If you want to build stuff,
apt-get install build-essential. Whether the desktop task should install
build-essential is debateable. I'm inclined to think there should be a
"Development" task that installs at least build-essential packages...
 

I think a software developer task is a fine idea. If you can tune for 
specialisations - GNOME developers probably don't want KDE libraries, 
Kernel or httpd hackers might not evne want a desktop, let alone KDE and 
GNOME libraies -then so much the better.

 

3) It should be a way to find out what's the best mirror. I had to try I few
until I found that was acceptably fast.
   

Well that's going to vary from person to person. I'm not sure how something
like apt-spy can be integrated with the mirror selection process.
 

For me the best mirror is the one that doesn't cost me. I don't care 
about speed, prixiity measured in dollars is all that matters.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: release status

[John Summerfield]

Joey Hess wrote:
John Summerfield wrote:
 

I set out to find whether the new manual provides that kind of 
information, or pointer to where it may be found.  In the course of my 
search I have discovered:
a)  http://d-i.alioth.debian.org/manual/en.arm/apas03.html contains links 
to
a.1 The Multi Disk HOWTO 
<http://www.tldp.org/HOWTO/Multi-Disk-HOWTO.html> which is quite old and 
seriously out of date. I doubt whether any but the most frugal would be 
satisfied, for example, with the partitions sizes recommended on 
http://www.tldp.org/HOWTO/Multi-Disk-HOWTO-10.html#ss10.1. If nobody 
wants to take on the task of revising it, I suggest the link be dropped.
   

It's only two years out of date and looks generally correct to me as far
as workable minimum sizes go. I see no real reason to drop the link.
 

I have a minimimal Sarge install. it's 200 Mbytes.
That rules out Learning.
I don't know what hoppies one can practice with 300 Mbytes of free disk 
and no optional packages, but it seems to me you'd be constrained.

If you want a GUI disktop and a reasonable set of software, 2 gbytes is 
about the going rate: it's what I used on RHL 7.3, it's about what the 
progeny install I just did takes. SuSE 9.0 came out a bit lite at 1.4. I 
don't have a recent equivalent Sarge to look at.

I know my 30 Gbyte drive is bulging at the perimeter. If I thought to 
master DVDs it would be nowhere near enough.


--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Opinion about sarge install

[John Summerfield]

Andrew Pollock wrote:
On Sun, Aug 08, 2004 at 11:56:00PM +0100, peter green wrote:
 

It would probably be interesting and useful for the X packaging
team to see
a diff between the config that was generated from configuring X and the
config you ultimately had to use to get X working. In my experience,
provided you give the X configuration the right input, it does a
reasonable
job of getting X working.
 

i think the point is that the X setup tool needs a full auto mode
almost every other distro has auto setup of X (even some debian based ones
like KNOPPIX)
normal users CANNOT be expected to answer those kind of questions
   

You should jump on [EMAIL PROTECTED] and have a discussion about
this. I think it's been discussed in the past.
The main thing is, Knoppix runs on 1 architecture, the X configuration has
to work on all 11. I believe Branden has said in the past that autodetection
of the monitor and video card can have undesirable results (like hanging the
system) on some combinations of hardware and architectures.
That said, it may be feasible to autodetect everything on i386, which is the
probably the architecture that X is most installed on... I don't know. Talk
to the X folks if you feel strongly about it.
 

The hardware detection Knoppix uses comes from Red Hat Linux, and Red 
Hat Linux supports
IA32
IA64
AMD-64
S/390 & zSeries
IBM pSeries
IBM pSeries
I believe it's also used in Yellow Dog Linux and that runs on Macs.
It also used to run on Sparc - I installed RHL 6.2 on a lunchbox here 
when Debian wouldn't.

Of all those, IA32 is the post complex, and most of those not covered 
have a much smaller variety of hardware to fuss over.

I have a powerbook here. How many different graphics chipsets, firewire 
chipsets, gigabyte NICs, DVD burners, USB2 ports are you likely to 
encounter?

I used to be a sysprog on S370. Yes things might have changed now, but 
then if you wanted to sell me a some disk drives, tapes, printers - 
anything- the onus was on _you_ to ensure it worked with the OS _I_ was 
using at the time, and do so in a way that didn't affect support for my 
OS.  Commonly that was done by making devices look sufficiently like IBM 
devices that no software changes were needed.

Even now, IBM storage devices that are not 3390s present themselves as 
3390s so as to be compatible with older software (MVS).

I do not believe that extending the hardware detection used by Knoppix 
is a big issue.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

di and woody

[John Summerfield]

A few minutes ago Ibooted d-i off my week-old network boot setup and 
tried to install Woody.

It coukln't find the needed netinstall image from the local official mirror.
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: di and woody

[John Summerfield]

Petter Reinholdtsen wrote:
[John Summerfield]
 

A few minutes ago Ibooted d-i off my week-old network boot setup and 
tried to install Woody.

It coukln't find the needed netinstall image from the local official
mirror.
   

I'm not too surprised.  Woody installs are only tested with CDs, as
far as I know.  No-one tested it with network installs yet, and I
expect lots of bugs are left. :)
But the CD install work with Debian-Edu (d-i + Woody + some
upgraded/extra packages), and might work with the normal Woody too. :)
 

At present I'm on a modem. Network installs beat downloading the lot:-)

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: release status

[John Summerfield]

Joey Hess wrote:
John Summerfield wrote:
 

I set out to find whether the new manual provides that kind of 
information, or pointer to where it may be found.  In the course of my 
search I have discovered:
a)  http://d-i.alioth.debian.org/manual/en.arm/apas03.html contains links 
to
a.1 The Multi Disk HOWTO 
<http://www.tldp.org/HOWTO/Multi-Disk-HOWTO.html> which is quite old and 
seriously out of date. I doubt whether any but the most frugal would be 
satisfied, for example, with the partitions sizes recommended on 
http://www.tldp.org/HOWTO/Multi-Disk-HOWTO-10.html#ss10.1. If nobody 
wants to take on the task of revising it, I suggest the link be dropped.
   

It's only two years out of date and looks generally correct to me as far
as workable minimum sizes go. I see no real reason to drop the link.
 

a.2 A link to Partitioning Strategies 
<http://www.tldp.org/HOWTO/mini/Partition/partition-5.html#SUBMITTED> 
which gets a 404.
   

Removed.
 

b) No manuals for PowerPC or s390.
   

I see both in the table at http://d-i.alioth.debian.org/manual/
 

I saw the links there and got errors following them:
Quite possibly I misinterpreted this error:
1091938534.561704 192.168.9.114 TCP_MISS/503 1466 GET 
http://d-i.alioth.debian.org/manual/en.s390/index.html - NONE/- text/html
Host: d-i.alioth.debian.org
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) 
Gecko/20040413 Debian/1.6-5
Accept: text/xml,application/xml,application/xhtml 
xml,text/html;q=0.9,text/plain;q=0.8,image/png,image/jpeg,image/gif;q=0.2,*/*;q=0.1
Accept-Language: en-au,en-us;q=0.7,en;q=0.3
Accept-Encoding: gzip,deflate
Accept-Charset: ISO-8859-15,utf-8;q=0.7,*;q=0.7
Keep-Alive: 300
Proxy-Connection: keep-alive
Referer: http://d-i.alioth.debian.org/manual/

HTTP/1.0 503 Service Unavailable
Server: squid/2.5.STABLE6
Mime-Version: 1.0
Date: Sun, 08 Aug 2004 04:15:34 GMT
Content-Type: text/html
Content-Length: 1117
Expires: Sun, 08 Aug 2004 04:15:34 GMT
X-Squid-Error: ERR_CONNECT_FAIL 101
Sorry for the woft.
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Manuals on S/390

[John Summerfield]

Some of these may apply to otherr arches too. eg kernel version.
Floppies on ?
http://d-i.alioth.debian.org/manual/en.s390/ch05s03.html#unreliable-floppies
http://d-i.alioth.debian.org/manual/en.s390/ch05s02.html#installer-argsThe 
value of the parameter is the path to the device to load the Debian 
installer from. For example, *INSTALL_MEDIA_DEV=/dev/floppy/0*

Do they have framebuffers?
USB?
PCMCIA?
a 2.2.x kernel?
http://d-i.alioth.debian.org/manual/en.s390/ch06s03.html#id2511429
Diskless workstations?
Booting other operating systems?
http://d-i.alioth.debian.org/manual/en.powerpc/ch02s02.html#id2506847
More floppies.
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: monochrome monitor ?

[John Summerfield]

mark david mcCreary wrote:
Thanks to everyone who has worked on the new Sarge installer.
I may be the last person in the USA that uses monochrome monitors, but
perhaps it is of value to people in other parts of the world.
I could install Woody with a monochrome monitor, but not Sarge.
 

VGA? (I have one somewhere)
Hercules?
IBM CGA vintage?

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#265194: resizing needs to be mentioned more in the partitioning utility

[John Summerfield]

Paul Harper wrote:
I agree with this. I found the partitioning to be a
bit confusing. Fortunately I did not have a Windoze
partition to wory about and I had backed up my /home.
 

In contrast, I installed SuSE the other day, intending to trash my 
Windows98. However, the installer explicitly offered to resize the 
partition and gave me a (GUI) knob to drag along a slider to choose the 
size I wanted.

It was clear that I could do it,  easy to do when I chose the option.


--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/

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

Bug#267321: Debian installer and /etc/hosts

[John Summerfield]

Seems to me that this behaviour Thos proposes is thoroughly broken. 
Several times I've had problems because the host name resolves to the 
wrong address.

I'll go further too and propose that sudo's hehaviour is broken. It 
should not be trying to resolve $HOSTNAME if there's no external 
network: note that if there's no external network that implies the user 
has local access. Test rules for 127.0.0.1 by all means.

The brokenness of sudo is no reason to break even more software.
If there is no external network, then the behaviour of programs such as 
ping, telnet, ssh are irrelevant. All are of limited use without a 
functioning network.


Thomas Hood wrote:
On Sun, 2004-08-22 at 22:37, VSJ wrote:
 

I'd go for the following solution:
   

[... not to include the hostname as an alias for 127.0.0.1
if the network is configured via DHCP.]
I am beginning to think that we should bite the bullet and do one of
two things.  Either,
1. to require software to work even if  cannot be resolved,
or,
2. to make the following a standard entry in /etc/hosts:
  127.0.1.1   
Repeating myself from #247734 ...
I think we should continue to consider the alternative of having:
   127.0.0.1localhost.localdomain   localhost
   127.0.1.1
when there is no static IP address we can use instead of '127.0.1.1'.
The advantage of this configuration is that localhost.localdomain is
the canonical hostname for the standard loopback address -- the 
address one gets if one looks up 'localhost', and  is the
canonical host name for the address one gets if one looks up
.

I don't think that there are many disadvantages.  All sudo wants is that
the hostname be resolvable; the address to which the hostname resolves
doesn't have to work.  There may be other programs that want to resolve
the hostname and also use it and for that reason it seems safest to 
make the hostname resolve to a 127.* address.  127.0.1.1 works as well
as 127.0.0.1 for some services: I have tried ping, telnet, ssh and
name lookups using dnsmasq and they all work at 127.0.1.1.  If there
are services for which 127.0.1.1 fails to work then we will either have
to modify those services or tell people not to access those services via
 but via 'localhost'.
--
Thomas


 

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/

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

Bug#267321: Debian installer and /etc/hosts

[John Summerfield]

Thomas Hood wrote:
John Summerfield continued:
 

If there is no external network, then the behaviour of programs such as 
ping, telnet, ssh are irrelevant. All are of limited use without a 
functioning network.
   

That isn't entirely true.  Some programs use the loopback interface to
provide/use services locally.
 

Then it's the user's task to configure the system appropriately. I run a 
webserver. I can configure it to listen to any IP address I choose, 
including 127.0.0.1. I don't need any bogus network interfaces, IP 
addresses or host names misconfigure by the vendor.

I mentioned _those_ programs by name because they were mentioned in the 
email that attracted my attention. None of those require a 
(misconfigured) hostname to resolve to anything.

Better to allow the user to configure the system _properly_ and to have 
it fail if it's broken than to make it work with misconfiguration.



--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/

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

Bug#268158: choose-mirror handles backslashes in http/ftp proxy input differently to rest of system

[John Summerfield]

Paul Hampson wrote:
Package: choose-mirror
Version: N/A; reported 2004-08-26
Severity: normal
Using choose-mirror in a daily netinst tarball downloaded 2004-08-24,
a problem arose trying use to a web-proxy that authenticates against
a windows domain, requiring the following syntax:
http://DOMAN\user:[EMAIL PROTECTED]:port/
If used on the command line, it must be:
a# http_proxy=http://DOMAN\\user:[EMAIL PROTECTED]:port/ wget 
but for the entirety of setup process except the first usage by
choose-mirror, the backslash does not need to be doubled.
Choose-mirror itself is not undoubling the \'s, since it goes into the
debconf database with \\, but I suspect it once calls wget like a#
above, while anna uses net-retriever which only works if the response
from debconf has a single \.
 

First guess, it's getting eaten by a shell. Arguably, this is correct 
behaviour:-)

What happens if you use the syntactically-correct slash instead of the 
backslash?


--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/

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

Re: [PROPOSAL] 2.4.27 as default 2.4 kernel for sarge

[John Summerfield]

Steve Langasek wrote:
 Various tasks are in a hold pattern until this decision is made (ensuring
that d-i uses the proper kernel, removal of other kernel packages from
sarge, rebuilding of some packages to fix build-dep issues[4]), so I'd
like to uncover any problems with this proposal quickly.
   

Because this gives us a shot at having all architectures on the same
version for sarge (where 2.4.26 does not due to arm), I agree that this
is the way to go.
I should be able to get linux-kernel-di-alpha done and uploaded by
Monday.
Sven, how far out are the 2.4.27 powerpc debs?  If these aren't
receiving enough attention because 2.6 is such a priority, I think we
need to seriously consider dropping the 2.4 kernels completely for
powerpc instead of giving them half-hearted support that will delay the
release.
 

All here who have 2.4 and 2.6 kernels on ppc should try disk speed tests 
with hdparm: I found my new Athlon (well the mobo's new, CPU's not) is 
30% faster with the 2.4 kernel.

If something like this is true of ppc too, you wouldn't want to drop 2.4 
kernels.


--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [PROPOSAL] 2.4.27 as default 2.4 kernel for sarge

[John Summerfield]

Sven Luther wrote:
All here who have 2.4 and 2.6 kernels on ppc should try disk speed tests 
with hdparm: I found my new Athlon (well the mobo's new, CPU's not) is 
30% faster with the 2.4 kernel.
   

Have you reported a bug report on this ? And with which 2.6 kernel was it ? 

 

I haven't. I planned on discussing it on debian-user first, and I 
mentioned it in passing in another thread and someone else was going to 
tell me what I did wrong.

I posted further details, but there's been no response.
The kernel is 2.6.7-1-k7; it may be a day or so before I could try 8 as 
I'm on dialup and my modem's rather busy atm.

the mobo's a Gigabyte GA-7S748-L - SiS 748 chipset, LAN.
fwiw I noticed something very like this between 2.2 and 2.4 when 2.4 was 
new: 2.2 was faster on my Pentium system. I think it was a earlier 
version of the same chipset.

Here are results on 2.6.7-1-k7:
/dev/hda:
Timing buffered disk reads:  108 MB in  3.02 seconds =  35.78 MB/sec
/dev/hdg:
Timing buffered disk reads:  120 MB in  3.01 seconds =  39.93 MB/sec
kowari:/etc#
Both drives are WD120 Gb (different models), hdg is on an Abit hotrod66 
(promise chips).


If something like this is true of ppc too, you wouldn't want to drop 2.4 
kernels.
   

Well, it should be fixed instead. None of upstream (all the linux-ppc
devleopers) have any interest left for 2.4, so there is really no sane way to
keep the support going unless we do it all ourselves.
 

It's worth testing; if you find no problem you have no problem:-)

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [PROPOSAL] 2.4.27 as default 2.4 kernel for sarge

[John Summerfield]

Sven Luther wrote:
fwiw I noticed something very like this between 2.2 and 2.4 when 2.4 was 
new: 2.2 was faster on my Pentium system. I think it was a earlier 
version of the same chipset.

Here are results on 2.6.7-1-k7:
/dev/hda:
Timing buffered disk reads:  108 MB in  3.02 seconds =  35.78 MB/sec
/dev/hdg:
Timing buffered disk reads:  120 MB in  3.01 seconds =  39.93 MB/sec
kowari:/etc#
Both drives are WD120 Gb (different models), hdg is on an Abit hotrod66 
(promise chips).
   

And how much you would get in 2.4 ? 

 

Sorry.
kowari:~# hdparm -t /dev/hd{a,g}{,}
/dev/hda:
Timing buffered disk reads:  152 MB in  3.02 seconds =  50.33 MB/sec
/dev/hda:
Timing buffered disk reads:  152 MB in  3.02 seconds =  50.33 MB/sec
/dev/hdg:
Timing buffered disk reads:  120 MB in  3.02 seconds =  39.74 MB/sec
/dev/hdg:  Timing buffered disk reads:  120 MB in  3.00 seconds =  40.00 MB/sec
kowari:~# uname -r
2.4.26-1-k7
kowari:~# 




Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Hacking on d-i

[John Summerfield]

I'd like to make a few minor changes for my own use - specify my local 
mirror, maybe partition layout and such.

I started out by unpacking the initrd to see what's in it, and was 
pretty happy to see debian-installer is a shell script: there's almost 
no limit to the magic one can do there on a good day:-)

There's no clue there as to what to do, but I did try the alternative 
f/e (bogl) and discovered it does nothing useful.

A bit of  finding and grepping and I discovered where the list of 
mirrors is. I've _not_ found a sensible way to change that list, nor  to 
specify my preferred value in any kind of configuration file.

Next, I went looking for the d-i home page. I tried 
www.debian.org/debian-installer - no, not there. Finally, Google gave up 
some clues and I finished up at https://alioth.debian.org/projects/d-i/

There, I find "This project has no visible documents ."
No subprojects have been set up, or you cannot view them.
The question: where is the information I need to change the mirror to be 
chosen,  specify any initial defaults I need?

btw My browser doesn't like the certificate from this site, its issuer 
is unknown.


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

Re: net-boot installer question

[John Summerfield]

John Smith wrote:
On Sun, 2004-06-27 at 01:04, Joey Hess wrote: 
 

John Smith wrote:
   

just started using the debian-installer, got my first few
systems installed with sarge through PXE and net-boot.
	Wanted to find out more about the inner workings and 
modify the first boot of the kernel. According to the instructions
vmlinuz is booted with init=/linuxrc so I uncompressed and mounted
the initrd.gz and expected to find a /linuxrc. I didn't. Is this
not valid anymore or am I on the wrong track?
 

Those must be old instructions (which ones?), init has been /sbin/init
for a while.
   

Used those:
http://wiki.debian.net/index.cgi?DebianInstallerNetbootPXE
 

Since I'm PXE-booting, I thought I'd cast my eye over it. I don't know 
what version of d-i I'm using, likely it's beta4 or later (depending on 
what was on the ISOs I picked up a while ago).

I'm using this:
label 4
   kernel images/sarge/vmlinuz
   append initrd=images/sarge/net-initrd.gz
with apparent success - I've not let it go far as I want to fiddle with 
it first,

What I really want to say though is this:
The requested URL /debian-boot/2004/debian-boot-200406/msg0.html' 
was not found on this server.

Since I haven't a clue where it really is, I'll leave it to someone else.
_I_ do have a linuxrc, but it's pretty short.
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Hacking on d-i

[John Summerfield]

Andrew Pollock wrote:
On Sun, Jun 27, 2004 at 01:00:32PM +0800, John Summerfield wrote:
 

I'd like to make a few minor changes for my own use - specify my local 
mirror, maybe partition layout and such.

I started out by unpacking the initrd to see what's in it, and was 
pretty happy to see debian-installer is a shell script: there's almost 
no limit to the magic one can do there on a good day:-)

There's no clue there as to what to do, but I did try the alternative 
f/e (bogl) and discovered it does nothing useful.

A bit of  finding and grepping and I discovered where the list of 
mirrors is. I've _not_ found a sensible way to change that list, nor  to 
specify my preferred value in any kind of configuration file.

Next, I went looking for the d-i home page. I tried 
www.debian.org/debian-installer - no, not there. Finally, Google gave up 
some clues and I finished up at https://alioth.debian.org/projects/d-i/
   

Have you been to http://www.debian.org/devel/debian-installer ?
 

No, my searches didn't show that one. I was looking for pages including 
the term "home  page" and that one doesn't.

 

There, I find "This project has no visible documents ."
No subprojects have been set up, or you cannot view them.
   

What you need to do is checkout a copy of the d-i subversion repository.
See http://www.nl.debian.org/devel/debian-installer/svn
 

Well I don't have subversion, but I saw this link:
daily subversion repository tarball 
<http://d-i.alioth.debian.org/svn/svnrepo.tar.gz>

It points to an empty compressed tarball:
[EMAIL PROTECTED]:~$ lynx -head -dump 
http://d-i.alioth.debian.org/svn/svnrepo.tar.gz
HTTP/1.0 200 OK
Accept-Ranges: bytes
Date: Sun, 27 Jun 2004 10:24:26 GMT
Content-Length: 45
Content-Type: application/x-tar
Server: Apache/1.3.26 (Unix) Debian GNU/Linux PHP/4.1.2 mod_ssl/2.8.9 
OpenSSL/0
.9.6g
Last-Modified: Sat, 26 Jun 2004 11:14:02 GMT
ETag: "9180d1-2d-40dd5a7a"
Content-Encoding: gzip
Via: 1.1 netcache (NetCache NetApp/5.5R4D2)

I'm on dialup, I was hoping to estimate what my download time would be.
I have a machine on ADSL I can hack on, but it runs woody and must stay 
on woody.

There's a lot of documentation, some of it quite enlightening within a
directory of the source tree.
 

I've read some of the docs: so far what I've seen is more involved than 
I'd hoped for,

btw The faq has wrong info on Anaconda: Anaconda is multiplatform - 
hands _all_ the kit IBM sells including IA64 and AMD-64 machines. Does 
network installs (and boots) - I've done them many times. Gives you a 
GUI installer if you want, even on S/390. Gives both text-based and GUI 
options. In 2000 it supported installing on Sun and Alpha hardware.

I'm still reading
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Hacking on d-i

[John Summerfield]

Colin Watson wrote:
On Sun, Jun 27, 2004 at 09:01:36PM +0800, John Summerfield wrote:
 

Andrew Pollock wrote:
   

What you need to do is checkout a copy of the d-i subversion repository.
See http://www.nl.debian.org/devel/debian-installer/svn
 

Well I don't have subversion, but I saw this link:
   

You can get hold of Subversion packages for woody here:
 http://people.debian.org/~adconrad/
 

I have a machine on ADSL I can hack on, but it runs woody and must stay 
on woody.
   

Use an unstable chroot, if you can.
 

It's a bit short of disk, and it hangs when I try to use the 120 Gbyte 
WD drive I put in it:-((

The most serious problem I want to circumvent is the default internet 
site it installs from. I _can_ get round that one with a transparent 
proxy firewall rule and configuring Apache as a proxy: I did that once 
before (for RHL) and it worked a treat.

I was hoping d-i waa a bit easier to configure than it seems.

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

Re: Automatic Installations

[John Summerfield]

Brian Sutherland wrote:
Hello all,
I have to do a project requiring an auto-installation and auto
configuration of many machines. I have looked at fai and
autoinstall but am looking for other possibilities as well. Can
d-i do this? or would i be better off taking some pieces of it
and using them in fai?
Some references of where I could start would be great.
 

You could also take a look at Anaconda at progeny.com.
Anaconda does a fine job on Red Hat Linux, and Progeny has ported it to 
Debian.

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

Suggestion for choose-mirror

[John Summerfield]

Could choose-mirror be changed so that if certain environment variables 
are set then it uses values from those rather than prompt for them?

II suggest that the proxy information be specified thus:
http_proxy=http://example.com:3128/
The reason I suggest http_proxy over any others is that it's already 
used for that purpose by other commonly-used programs such as lynx, wget.

Of course, for an ftp install then ftp_proxy would be appropriate.
I don't care so much what others are, but maybe
source=
 http://www.example.com/debian/
 ftp://ftp.example.com/pub/debian
 nfs://nfs.example.com/var/local/mirror/debian
 usb://
 cd://
and whatever else is needed.
While these are more than almost anyone will type in at tbe boot prompt 
(more than one or twice, at least), it's not hard to supply the info  
via pxelinux, syslinux, grub etc configurations.

This would go some way to automating the install prcedure, especially if 
the idea's adapted to other packages.

Any thoughts?
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Suggestion for choose-mirror

[John Summerfield]

Colin Watson wrote:
On Thu, Jul 15, 2004 at 08:40:52PM +0800, John Summerfield wrote:
 

Could choose-mirror be changed so that if certain environment variables 
are set then it uses values from those rather than prompt for them?

II suggest that the proxy information be specified thus:
http_proxy=http://example.com:3128/
   

I expect that you can already do this using
mirror/http/proxy=http://example.com:3128/ at the kernel boot prompt.
 

This would go some way to automating the install prcedure, especially if 
the idea's adapted to other packages.
   

Since d-i uses debconf, most questions are pre-answerable like this.
 

Where are the Qs and As documented?
It would be pretty trivial to front-end d-i to get a text file 
containing some shell script to set these.


--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Bug#257118: marked as done (debian-installer: update www.d.o/devel/debian-installer)

[John Summerfield]

Joey Hess wrote:
Geert Stappers wrote:
 

Currently is the most recent News entry dated 30 april 2004
and no signs of TC1. ( my apology for typing previous RC1 )
I have no clue why Test Candidate 1 doesn't has
a entry in webwml/english/devel/debian-installer/News/2004
   

Because it wasn't a general user quality release..
 

What do you want users to actually test? I'd have thought a 
three-month-old beta a little long in the tooth for testing.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Problem formatting existing partitions

[John Summerfield]

Frans Pop wrote:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On Thursday 15 July 2004 09:14, John wrote:
 

I'm trying to install to a disk with existing (ext3) parititions.
I chose to use the existing partitions, to reformat them as ext3.
The format failed as the partitions have existing journals.
   

That should not cause the formatting to fail. I do this all the time.
Could it be the failure occurs because the partitions are mounted in
some way?
 

It transpired they were mounted, but that's not what the message said.
If not please file a full installation report using [1] and attach the
files syslog, messages and partman from /var/log directory (if possible).
That should contain all the information we need to track down the problem.
 

That is not possible. I retried the partitioning with cfdisk, and _then_ 
the reformatting failed because the partitions were mounted (I had been 
surprised for find Linux partitions on the disk and had been investigating).

One I unmounted the parttitions the formatting proceded normally and 
first part of the install has completed. Atm I've got Knoppix booted so 
I can inspect the system (and take a copy) before finishing the install.

All the evidence is gone.

On a side note, can we have symlinks in /dev for those who like the 
disks to appear in the traditional place and to munge them with more 
user-friendly tools such as fdisk?
   

I'm afraid you'll have to start fdisk with /dev/discs/...
 

btw Is there any reason that LAN-booting expects I want to do a network 
install? I could, for example, install off USB on a system that won't 
boot USB.
   

The reason is we expect 99,9% of people using the netboot boot images to
install off the network :-)
You could try to place the hd-image boot images instead of the netboot ones
on your boot server. I think they should contain the functionality to use
an iso on e.g. an usb stick.
[1] http://www.nl.debian.org/devel/debian-installer/report-template
Cheers,
FJP
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.1 (GNU/Linux)
iD8DBQFA9skYgm/Kwh6ICoQRAsg+AJ0WijnfM6uwS5zjstF7tj+vcqvVgQCdGw9K
ccMh6iMH29rVlB18acuWdU0=
=cy1f
-END PGP SIGNATURE-
 


--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Problem formatting existing partitions

[John Summerfield]

Osamu Aoki wrote:
On Thu, Jul 15, 2004 at 03:14:47PM +0800, John wrote:
...
 

On a side note, can we have symlinks in /dev for those who like the 
disks to appear in the traditional place and to munge them with more 
user-friendly tools such as fdisk?
   

I know the feelings :) 

But situation is not as bad as you think.  d-i shell has [TAB] code
enabled.
 

And mightily pleased I was to find that so:-)
fdisk /dev/[TAB][TAB]
will list all device names and the rest is easy guess by the same 
[TAB][TAB] expansion.
 

I'd still like the symlinks:-)
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#250865: Set automatically bootable flag [was: Bug#250865: Install repor]

[John Summerfield]

Anton Zinoviev wrote:
On Thu, May 27, 2004 at 12:23:10PM -0300, Martin Michlmayr wrote:
 

reassign 250865 partman
retitle 250865 warn if not partition is bootable
severity 250865 minor
thanks
partman should also set the bootable flag automatically if there's
none already.
   

Only on i386 I guess.
What if both the root and /boot partitions are logical and not primary?
Parted (and partman) allow bootable logical partitions but I have no
idea whether bootable logical partitions are usefull for anything.
 

Most boot managers don't requre bootable partitions for anything.
In my environment only Darwin uses the bootable flag, and atm I can't 
boot darwin at all on my dual-boot system.

OS/2 can boot from a logical partition: what the OS/2 bootmananger does 
with active flags, if anything, I don't know. I sispect nothing as 
otherwise the code in the MBR would get confused.

For those who don't know, OS/2 bootmanager has its own partition.
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/

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

Re: Suggestion for choose-mirror

[John Summerfield]

Frans Pop wrote:
On Friday 16 July 2004 00:26, John Summerfield wrote:
 

Where are the Qs and As documented?
   

http://wiki.debian.net/DebianInstallerFAQ
 

I don't see there any of the debconf Qs and As to which I was referring. 
Such as what the Q is that asks about mirrors and how to automatically 
answer it. Colin's answered that one, but there are more.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

iptables

[John Summerfield]

I just installed Sarge off the local mirror.
Is there any point in installing ipchains?
If you think it valuable for use with 2.4 kernels, how about with 2.6?
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

ssh

[John Summerfield]

Is there any good reason ssh is _not_ installed?
I did the absolute minimum install because I didn't like the choices 
offered.

I found myself without any way of getting into the box from outside, or 
connecting to remote sites (in my case, the other side of the desk).

I have (in development) a script that installs the stuff I want and sets 
it up the way I want, but getting it into the box is harder than I like.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Open ports

[John Summerfield]

After my minimal install, I find some ports open to the public. I don't 
think they should be open by default  to anyone.

They're all in the inetd package, any my preference is to have all inet 
services disabled by default.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

smartmontools

[John Summerfield]

I've fallen in love with smartmontools. I think it should be on every 
Linux box that has disk drives.

For those unfamiliar with it, it monitors the health of disk drives 
(SCSI and ATA, SATA support coming), and can warn of impending failure 
in time to get your data off the failing drive.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: ssh

[John Summerfield]

Clive Menzies wrote:
On (16/07/04 15:11), Kenshi Muto wrote:
 

To: John Summerfield <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
From: Kenshi Muto <[EMAIL PROTECTED]>
Date: Fri, 16 Jul 2004 15:11:35 +0900
Subject: Re: ssh
At 16 Jul 04 05:00:22 GMT,
John Summerfield wrote:
   

Is there any good reason ssh is _not_ installed?
 

ssh is "standard" priority. If you don't skip tasksel, ssh will be
installed as default.
   

Interesting that telnet is included in the bare install (before running
tasksel, deselect or aptitude) - given the security risk of using
telnet, it would make sense to remove it and perhaps include ssh
instead?
 

The telnet client is no security risk, and is very useful for diagnosing mail problem, 
configuring billion routers (well, some of them) etc.
I didn't run tasksel as last time I did, it pulled in great gobs of stuff I didn't 
want.
However, I would like ssh (server and client) in the minimum install so as to provide 
a means of setting up headless boxes.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: ssh

[John Summerfield]

Frans Pop wrote:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On Friday 16 July 2004 07:00, John Summerfield wrote:
 

Is there any good reason ssh is _not_ installed?
   

There is a ssh-client component for d-i in the making
that will allow you to use ssh while the installation
is running (e.g. to get logfiles of the new system).
 

The problem I had was after the first boot. I regard what happens _before_ then a 
separate problem.
What do folk on zSeries do?

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Bug#257118: marked as done (debian-installer: update www.d.o/devel/debian-installer)

[John Summerfield]

Joey Hess wrote:
John Summerfield wrote:
 

What do you want users to actually test? I'd have thought a 
three-month-old beta a little long in the tooth for testing.
   

I think the web site is perfectly clear on this, so I'll just quote it.
If you just need an installation that works, we recommend you use beta4
of the installer, after checking its errata.
 

The text is clear enough. I simply didn't believe it because, as I said, 
it's very old

If you'd like something newer to help us test a future release of
the installer, or if beta4 does not support your hardware, try one
of the daily built images.
 

Why not send these out in Sid?
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: squashfs compressed file system update?

[John Summerfield]

On Wed, 19 Feb 2003, Glenn McGrath wrote:

> On Tue, 18 Feb 2003 10:08:45 -0600 (CST)
> Drew Scott Daniels <[EMAIL PROTECTED]> wrote:
> 
> > Would squashfs work? I doubt this is an optimal compression, but it
> > might be better than alternatives. Currently the best free solution
> > might be based on PPMd which has been packaged into Debian.
> > 
> 
> Im not confident about squashfs, i think the best solution is
> initrd.romfs.gz for normal installs, but squashfs or cramfs might be
> good for lowmem installs, i havent looked into squashfs yet, not sure
> when i will get time.
> 
> I have looked at PPMd, ive read thats good but it took me a while to
> workout how to use it to get maximum ocmpression, here is a comparison.
> 
> 8921088 Packages
> 2375680 Packages.gz
> 1814528 Packages.bz2
> 1335296 Packages.pmd
> 
> That was done with "PPMd e -o16 -m256 -r2 Packages", it surprised me
> that the -m option (amount of memory to use) effects the compression
> ratio.
> 
> Thats a pretty improvement over bz2.

Wouldn't it be more meaningful to compress an initrd file?


-- 
Cheers
John Summerfield
 


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

Re: Image file too large for low memory

[John Summerfield]

On 18 Feb 2003, Owen B. Mehegan wrote:

> Sure -
> 
> I added these two lines near the top of the file:
> 
> allow booting;
> allow bootp;
> 
> And then added this section at the end:
> 
>  host laptop {
>filename "/tftpboot/pxelinux.0";
>server-name "pointer";
>next-server pointer;
>hardware ethernet 00:B0:D0:A4:2B:88; 
>  }

These re PXE bits I've picked up too:
# PXE stuff
# Definition of PXE-specific options
# Code 1: Multicast IP address of bootfile
# Code 2: UDP port that client should monitor for MTFTP responses
# Code 3: UDP port that MTFTP servers are using to listen for MTFTP requests
# Code 4: Number of secondes a client must listen for activity before trying
# to start a new MTFTP transfer
# Code 5: Number of secondes a client must listen before trying to restart
# a MTFTP transfer
option space PXE;
option PXE.mtftp-ipcode 1 = ip-address;
option PXE.mtftp-cport code 2 = unsigned integer 16;
option PXE.mtftp-sport code 3 = unsigned integer 16;
option PXE.mtftp-tmout code 4 = unsigned integer 8;
option PXE.mtftp-delay code 5 = unsigned integer 8;


and
   class "pxeclients"
{
match if substring (option vendor-class-identifier, 0, 
9) = "PXEClient";
option vendor-class-identifier "PXEClient";
# At least one of the vendor-specific option must be set.   We set
# the MCAST IP address to 0.0.0.0 to tell the bootrom to still use
# TFTP (address 0.0.0.0 is considered as "no address")
#   option PXE.mtftp-ip 192.168.0.1; vendor-option-space 
PXE;
option PXE.mtftp-ip 0.0.0.0; vendor-option-space PXE;
filename "/PXE/pxelinux.0";
}



And there's this for the boot roms available from rom-o-matic:
# Filched from http://clic.mandrakesoft.com/documentation/pxe/ch04.html
  class "Etherboot"
{
 match if substring (option vendor-class-identifier, 
0, 13) = "Etherboot-5.0";
 option vendor-encapsulated-options 
3c:09:45:74:68:65:72:62:6f:6f:74:ff;
 option vendor-class-identifier "Etherboot-5.0";
 vendor-option-space PXE;
     option PXE.mtftp-ip 0.0.0.0;
 filename "/tftpboot.ser";
}
 
-- 
Cheers
John Summerfield
 


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

Bug#181812: busybox-cvs: find applet doesn't return the same files then busybox

[John Summerfield]

On Thu, 20 Feb 2003, Thorsten Sauter wrote:

> Package: busybox-cvs
> Version: 0.60.99.cvs20030114-1 (not installed)
> Severity: normal
> 
> Hi,
> 
> the find applet in busybox-cvs differs to the one in busybox.
> I think the both should work exactly.
> 
> $ find /dev/discs/*/disc  (busybox)
> /dev/discs/disc0/disc
> $ find /dev/discs/*/disc  (busybox-cvs)
> find: /dev/discs/*/disc: No such file or directory
> 
> I'm using the find util from the busybox-cvs udeb package for the
> debian-installer to detect the available harddrives in system. The old
> busybox-cvs (don't know the version anymore) support the command above.
> 
> Do you think it's possible to reinclude/support the command?

The second looks like standard shell behaviour when nothing matches the wildcard. If 
there should be a match, then your shell has the problem, not the find command.




-- 
Cheers
John Summerfield
 



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

Re: Bug#181812: busybox-cvs: find applet doesn't return the samefiles then busybox

[John Summerfield]

On Fri, 21 Feb 2003, Thorsten Sauter wrote:

> On Thu, Feb 20, 2003 at 02:48:05PM -0800, Matt Kraai wrote:
> > The shell performs glob expansion, not the command.  Are you sure that
> > /dev/discs/disc0/disc existed when you tested busybox-cvs?
> 
> true. sorry about this, it's maybe an ash problem and not a bug in the
> find applet, but it concerns to busybox-cvs.
> 
> The directory /dev/discs/disc0 exist, and the file /dev/discs/disc0/disc
> exist also. But /dev/discs/disc0 is an symbolic link into /dev/ide/
> I can simply try this with booting the d-i netimage and the don't select
> busybox-cvs in anna and the other time select it.

Does the target for the symlink exist? Its lack also causes that symptom.

-- 
Cheers
John Summerfield
 


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

Network install of Woody

[John Summerfield]

The documentation for using bootroms (such as PXE) is incomplete. See  the (ia386) 
install guide:
4.5.5 Move TFTP Images Into Place which links to 11.2.3, and contains the text, "NOT 
YET WRITTEN."

5.5 Booting from TFTP
This talks about floppies, confusing since I don't plan on using one (unless I
lack a working bootrom in the NIC).

4.5.6 Installing with TFTP and NFS Root
Doesn't really give any idea about how to do it. How do I create the NFS root to 
export? It happens I'm running Red Hat Linux, so Debian tools won't work. Maybe if 
it's on a CD somewhere

11.2.3 Description of Installation System Files
This mentions several files called tftpboot.img, but I recall someone on another list 
found it impossible to boot such large files because at that time the system's in real 
mode, can address no more than one Mbyte of RAM, and of that, only 640K is available 
to programs.

pxelinux can load a kernel and an initrd, so it doesn't look like those are the right 
files. Possibly, I could use a "rescue" image and a "root image."


I have the inclination to try to get it working. Is there someone who can work with me?

What I suggest is someone create corrected text (don't worry about formatting), and I 
will try it out. When the text checks out, maybe others can try it, but it should go 
to the maintainer for inclusion.

I already have working dhcp, tftp and a couple of machines that do PXE-boot, and I 
also have an Etherboot ROM-on-a-floppy that boots of the network, so most of the 
work's done.

I don't plan to visit other ways of handing out IP addresses etc. ISC's DHCPD works 
and is available (AFAIK) to all Linux users.

btw Do not reply directly to me. Mail to this address not from an approved source gets 
treated as spam.
If you think the discussion would clutter the list too much, I can contact you using 
another address.





-- 
Cheers
John Summerfield
 


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

Re: Network install of Woody

[John Summerfield]

On Thu, 20 Feb 2003, Chris Tillman wrote:

> On Fri, Feb 21, 2003 at 11:58:23AM +0800, John Summerfield wrote:
> > The documentation for using bootroms (such as PXE) is incomplete. See  the (ia386) 
>install guide:
> > 4.5.5 Move TFTP Images Into Place which links to 11.2.3, and contains the text, 
>"NOT YET WRITTEN."
> > 
> > 5.5 Booting from TFTP
> > This talks about floppies, confusing since I don't plan on using one (unless I
> > lack a working bootrom in the NIC).
> > 
> > 4.5.6 Installing with TFTP and NFS Root
> > Doesn't really give any idea about how to do it. How do I create the NFS root to 
>export? It happens I'm running Red Hat Linux, so Debian tools won't work. Maybe if 
>it's on a CD somewhere
> > 
> > 11.2.3 Description of Installation System Files
> > This mentions several files called tftpboot.img, but I recall someone on another 
>list found it impossible to boot such large files because at that time the system's 
>in real mode, can address no more than one Mbyte of RAM, and of that, only 640K is 
>available to programs.
> > 
> > pxelinux can load a kernel and an initrd, so it doesn't look like those are the 
>right files. Possibly, I could use a "rescue" image and a "root image."
> > 
> > 
> > I have the inclination to try to get it working. Is there someone who can work 
>with me?
> > 
> > What I suggest is someone create corrected text (don't worry about formatting), 
>and I will try it out. When the text checks out, maybe others can try it, but it 
>should go to the maintainer for inclusion.
> 
> Well, I'd suggest the opposite. We included that text based on one
> person's experience, if you can tell us how to improve the text, 
> based on your experience, we'd be happy to change it. There seems
> to be a lot of interest, but not too much concrete input. Maybe
> one of the issues is there are so many ways to get it to work?
 
If I were to write it (assuming I knew enough, which at present I don't), you
could probably point out errors, but not the inadequacies.

At present, I in my ignorance am well-equipped to locate the latter. If I wait
until I have worked through it and found a path that works, then it will become
clear enough that I will no longer be able to indentify clearly what's missing,
what facts I have assumed but which might not be known to others.


 

-- 
Cheers
John Summerfield
 


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

Re: debian-installer status 2003-02-21

[John Summerfield]

On 21 Feb 2003, Tollef Fog Heen wrote:

> - - automated installs and testing.  If somebody could give automated
>   installations and how to get those working better a shot, it'd be
>   cool.  Getting the report tool (which should fill out the report
>   template available at http://raw.no/d-i/report-template.html)
>   written would make it easier for testers to give us good reports.
> - - fix the ext3 problem; mkfs.ext3 gives an error about «file not
>   found» (which is really that it can't find /etc/mtab).  mkswap
>   should also be added.

Is there a quick-start guide on how to integrate D-I with Woody and do
this? Automatically is the only way I want to install Debian.

I've become quite adept at installing Red Hat Linux using kickstart, and
I've written my own (crude) Debian installer, so I've had time to
develop thoughts on what ought to happen.



-- 
Cheers
John Summerfield
 


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

Bug#181975: di-utils-mkfs: failure with unpartitioned hard drive

[John Summerfield]

On Sat, 22 Feb 2003, Oskar Liljeblad wrote:

> On Friday, February 21, 2003 at 15:00, Matt Kraai wrote:
> > 
> > > The fix is either to tell the user that there are no partitions,
> > > or (IMHO the preferred way for experts) allow the user to create
> > > a file system on the whole disk. Theoretical (untested) fix
> > > below.
> > 
> > Why would you prefer not to partition the drive?
> 
> IMHO the PC partition table format is obscure and a remnant of
> the past. If you plan on using all of the hard drive space for
> storage you won't need partitions either. I've been using
> several secondary hard drives (for storage only) without
> partition tables for quite some time now.
> 
> Seriously though, is this idea stupid?

I think it reasonable for removable media such as MO disks; the format is
called "superfloppy." For regular fixed disks, I think it more likely than not
to cause confusion, and any saving of disk space is hardly significant.


> Am I the only one who does this?
> 
> I don't know if it is possible to boot from a hard drive without
> a partition table on PC platforms. If there isn't there is no
> point whatsoever in allowing users to install debian on
> a hard drive without partitions.

I have installed LILO in both MBR and partitions, sometimes on the same system.
It works well, and I expect GRUB to do so too. I think if a superfloppy formt
disk doesn't boot, a bug report is in order.

OTOH the maintainer might respond, "unsupported.";-}


-- 
Cheers
John Summerfield
 



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

Re: Network install of Woody

[John Summerfield]

On Fri, 21 Feb 2003, Eduard Bloch wrote:

> #include 
> * John Summerfield [Fri, Feb 21 2003, 02:10:57PM]:
> 
> > > to be a lot of interest, but not too much concrete input. Maybe
> > > one of the issues is there are so many ways to get it to work?
> >  
> > If I were to write it (assuming I knew enough, which at present I don't), you
> > could probably point out errors, but not the inadequacies.
> > 
> > At present, I in my ignorance am well-equipped to locate the latter. If I wait
> > until I have worked through it and found a path that works, then it will become
> > clear enough that I will no longer be able to indentify clearly what's missing,
> > what facts I have assumed but which might not be known to others.
> 
> What about making notices on your way? Later, you can review them and
> remember what caused the problems and how another person may solve the
> problem easier.

In my experience, and the experience of others, good documentation can only be
written by someone who knows and understands the topic well. This is why
authors such as Colleen McCullouch (she spent 13 years researching before
writing "First Man in Rome") spend so much time in research.


While I could muddle my way through, at the end of it I would not have as good
a document as I would expect others here would produce. I see my my greatest
value in this as a checker; at present I won't make the same assumptions
someone to whom it's a trivial undertaking to make.  


-- 
Cheers
John Summerfield
Please, no off-list mail. It won't be read, it will be handled as spam.

 


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

Re: Bug#182349: anna: Should reload packages file if something wentwrong last time

[John Summerfield]

On Mon, 24 Feb 2003, Thomas Viehmann wrote:

> [Sorry, forgot the bug cc in my first post.]
> 
> Matt Kraai wrote:
> [anna reloading package files]
> > What problem this fix?
> One gets invalid packages files every so often (e.g. during mirror update).
> 

I get messages that my "Package files are corrupt" sometimes wrt my
local (Woody) mirror which I've constructed from ISO images. However,
there don't seem to be any adverse consequences, so I've ignored it.

Should I report this as a bug, if so, against what?



-- 
Cheers
John Summerfield
Please, no off-list mail. It won't be read, it will be handled as spam.

 


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

Re: Running debian-installer within bochs and uml

[John Summerfield]

On Mon, 24 Feb 2003, Thomas Viehmann wrote:

> Hi.
> 
> Because I don't have a cdrw burner nor a desire to reboot my computer with the
> debian-installer mirror for installing, I've been trying to run debian-installer
> within a bochs simulation session and uml.
> Bochs is somewhat slow, but the alternative plex86 doesn't seem to support
> networking. user-mode-linux doesn't support the mapping block devices as much
> AFAIK, so one has to use the ubd devices.

Check the UML documentation. I don't have it to hand at the moment, but
I believe there is a commandline option to map drives to hd{a,b,c} etc.

BTW UML would provide a reason to not partition drives.

> 
> 

-- 
Cheers
John Summerfield
Please, no off-list mail. It won't be read, it will be handled as spam.

 


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

Re: d-i web pages (Was: ports-status page)

[John Summerfield]

On 10 Mar 2003, Petter Reinholdtsen wrote:

> [Martin Sjögren]
> > You should probably add a link to Tollef's "getting started" page:
> > http://raw.no/d-i/getting_started.html
> 
> Would it be a good idea to have a set of web pages available somewhere
> on the web, and maintained using CVS?  I get the impression that the
> d-i information is spread across web pages all over the place, and
> that it is hard for new users to find all the relevant information.

Sounds an excellent idea to me, especially if you want strangers testing
d-i.

> 

-- 
Cheers
John Summerfield
Please, no off-list mail. It won't be read, it will be handled as spam.

 


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

Re: cdebconf upload

[John Summerfield]

On Mon, 10 Mar 2003, Chris Tillman wrote:

> On Mon, Mar 10, 2003 at 12:54:47PM +0100, Petter Reinholdtsen wrote:
> > 
> > [Martin Sjögren]
> > > I want to make a cdebconf upload to fix the problem of bogl being
> > > included in cdebconf-udeb.
> > 
> > Yes, please do.  The new cdebconf packages breaks my d-i build because
> > the floppies are full.
> 
> The floppies-are-full problem was a major PITA in boot-floppies,
> and is starting to come up quite often already for d-i.
> 
> Would it be worthwhile to standardize on 1 kernel floppy, 1 root
> floppy, with strictly minimal stuff on the root system just to be
> able to access floppies/CDs and start the installer, and then put off
> any remaining floppy involvement until after boot is complete - to
> load any other needed modules for specific cases as needed.

Until recently, RHL has installed off one floppy, for most users. Up to
and including RH 7.3, I could boot a floppy (a standard floppy with any
luck), and that floppy would load the kernel & initrd, identify the NIC
& load the driver, find the network and install.

At the syslinux prompt, one would enter "linux ks="


If at all possible, Debian likewise whould install from a single floppy.
A single-floppy install can be automated much better than one that
requires two.

If the installer absolutely cannot be run from a 1.4 Mbyte floppy, then
consider a one-floppy boot that will install from a network. (I am
assuming a 2.8 Mbyte floppy is fine).

The one-floppy boot that can install from a network might use etherboot,
grub or similar to get the kernel/init from a LAN. Not especially
useful (maybe) for the average punter, but wonderful for mass installs.


 

-- 
Cheers
John Summerfield
Please, no off-list mail. It won't be read, it will be handled as spam.

 


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

Re: cdebconf upload

[John Summerfield]

On Tue, 11 Mar 2003, Erik Andersen wrote:

> On Mon Mar 10, 2003 at 11:00:32PM -0500, Joey Hess wrote:
> > - switch to a smaller libc, such as uclibc
> 
> The only downside to this is that I currently do not support all
> the architectures that are supported by Debian.  Adding support
> for a new arch to uClibc is really not very hard, and most of the
> needed bits can be directly swiped from glibc, but it does take
> someone that is familiar with the architecture and willing to put
> in a bit of time.
> 
> Debian arches not currently supported by uClibc:
> hppa, hurd, ia64, s390
> 
> Working but need some additional polish and ldso support:
> alpha, sparc, m68k 
> 
> Fully supported:
> arm, i386, mips, mipsel, powerpc, sh
 
 
 Does d-i need a trim libc for all these architectures? For S/390, for
 example, if you can get a bootable kernel/initrd in place, you can get
 as bulky an installer as you could wish too.



-- 
Cheers
John Summerfield
Please, no off-list mail. It won't be read, it will be handled as spam.

 


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

Re: [PROPOSAL] 2.4.27 as default 2.4 kernel for sarge

[John Summerfield]

John Summerfield wrote:
Sven Luther wrote:
fwiw I noticed something very like this between 2.2 and 2.4 when 2.4 
was new: 2.2 was faster on my Pentium system. I think it was a 
earlier version of the same chipset.

Here are results on 2.6.7-1-k7:
/dev/hda:
Timing buffered disk reads:  108 MB in  3.02 seconds =  35.78 MB/sec
/dev/hdg:
Timing buffered disk reads:  120 MB in  3.01 seconds =  39.93 MB/sec
kowari:/etc#
Both drives are WD120 Gb (different models), hdg is on an Abit 
hotrod66 (promise chips).
  

And how much you would get in 2.4 ?
 

Sorry.
kowari:~# hdparm -t /dev/hd{a,g}{,}
/dev/hda:
Timing buffered disk reads:  152 MB in  3.02 seconds =  50.33 MB/sec
/dev/hda:
Timing buffered disk reads:  152 MB in  3.02 seconds =  50.33 MB/sec
/dev/hdg:
Timing buffered disk reads:  120 MB in  3.02 seconds =  39.74 MB/sec
/dev/hdg:  Timing buffered disk reads:  120 MB in  3.00 seconds =  
40.00 MB/sec
kowari:~# uname -r
2.4.26-1-k7
kowari:~#

I have further results, and I've just filed a bug report (confirmation 
not in yet so no no).
kowari:~# hdparm -t /dev/hd{a,g}{,}

/dev/hda:
Timing buffered disk reads:  112 MB in  3.05 seconds =  36.73 MB/sec
/dev/hda:
Timing buffered disk reads:  110 MB in  3.00 seconds =  36.64 MB/sec
/dev/hdg:
Timing buffered disk reads:  118 MB in  3.01 seconds =  39.21 MB/sec
/dev/hdg:
Timing buffered disk reads:  120 MB in  3.01 seconds =  39.83 MB/sec
kowari:~# uname -r
2.6.8-1-k7
kowari:~#
It's a self-built kernel, but without modification.
I neglected to include a modules listing in the bug report. Here are 
those myst likely to be involved:
ide_generic 1472  0
sis551316776  1
hpt366 22788  2
ide_disk   19264  5 hpt366
ide_core  138776  5 
usb_storage,ide_generic,sis5513,hpt366,ide_disk
sd_mod 21632  0
ata_piix8132  0
libata 41348  1 ata_piix
scsi_mod  124620  3 usb_storage,sd_mod,libata
unix   28592  170

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#268434: no usb in 2.6 at install time

[John Summerfield]

Michael Stone wrote:
Package: debian-installer
When booting from an rc1 cd on i386 the installer loads usbcore and
usbkbd, but not uhci_hcd. This makes it hard to install using a usb
keyboard. I'd suggest trying to load uhci_hcd, ohci_hcd, and ehci_hcd
during the startup phase.
Mike Stone
Does this mean installing to USB drive does not work? I was trying just 
that the other day (using Fedora Core 2 because I had the disks and 
didn't want to destroy the existing disk).

There are also USB network cards, all may need to be running.
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/

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

Re: RE/FW: Please stop sending me emails

[John Summerfield]

Martin Schulze wrote:
peter green wrote:
 

can someone please remove this guy from the list before his bloody
autoresponders drive us all crazy
   

He's removed from all lists and his mail address has been blocked
from sending to any list.  A mail to listmaster@ or some listmasters
personally should have been faster.
 

Just as I was about to come up with a procmail filter to send all his 
mail to [EMAIL PROTECTED] and/or brianb there. Hopefuly Brian Borowski will 
offer F some counselling.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [PROPOSAL] 2.4.27 as default 2.4 kernel for sarge

[John Summerfield]

Steve Langasek wrote:
We depend on the experts (the kernel team) for the information we need
in order to make good decisions -- or better, to help *you* make good
decisions.
Based on this thread and other discussions, I understand that the
current 2.4.26 packages are unsuitable for release because of security
bugs, and it looks like the new 2.4.27 packages are closer to being in a
releasable state than 2.4.26 is.  Compared with this, the NEW delays are
minor; I'm happy to stick out my neck to get the 2.4.27 packages
processed faster if I know the effort won't be wasted.
 

As a user, I'd prefer to wait a little longer so the experts can get it 
right; recent history suggest we'll live a long time with the results, 
so better a little more pain now if it means a better future.


--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [PROPOSAL] 2.4.27 as default 2.4 kernel for sarge

[John Summerfield]

kowari:~# hdparm -t /dev/hd{a,g}{,}
/dev/hda:
Timing buffered disk reads:  152 MB in  3.02 seconds =  50.33 MB/sec
/dev/hda:
Timing buffered disk reads:  152 MB in  3.02 seconds =  50.33 MB/sec
/dev/hdg:
Timing buffered disk reads:  120 MB in  3.02 seconds =  39.74 MB/sec
/dev/hdg:  Timing buffered disk reads:  120 MB in  3.00 seconds =  
40.00 MB/sec
kowari:~# uname -r
2.4.26-1-k7
kowari:~#
 

I have further results, and I've just filed a bug report (confirmation 
not in yet so no no).
kowari:~# hdparm -t /dev/hd{a,g}{,}

/dev/hda:
Timing buffered disk reads:  112 MB in  3.05 seconds =  36.73 MB/sec
/dev/hda:
Timing buffered disk reads:  110 MB in  3.00 seconds =  36.64 MB/sec
/dev/hdg:
Timing buffered disk reads:  118 MB in  3.01 seconds =  39.21 MB/sec
/dev/hdg:
Timing buffered disk reads:  120 MB in  3.01 seconds =  39.83 MB/sec
kowari:~# uname -r
2.6.8-1-k7
kowari:~#
It's a self-built kernel, but without modification.
I neglected to include a modules listing in the bug report. Here are 
those myst likely to be involved:
ide_generic 1472  0
sis551316776  1
hpt366 22788  2
ide_disk   19264  5 hpt366
ide_core  138776  5 
usb_storage,ide_generic,sis5513,hpt366,ide_disk
sd_mod 21632  0
ata_piix8132  0
libata 41348  1 ata_piix
scsi_mod  124620  3 usb_storage,sd_mod,libata
unix   28592  170
   

I wonder if your ide chipset got correctly detected (aparently), and if dma
and other such parameters are identic.
 

The sis5513 is plausible, I know it is used for earlier SiS chipsets.
Here are the kernel messages I think relevant:
PCI: Probing PCI hardware (bus 00)
Uncovering SIS963 that hid as a SIS503 (compatible=0)
Enabling SiS 96x SMBus.
ACPI: PCI Interrupt Routing Table [\_SB_.PCI0._PRT]
and
libata version 1.02 loaded.
Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
SIS5513: IDE controller at PCI slot :00:02.5
SIS5513: chipset revision 0
SIS5513: not 100% native mode: will probe irqs later
SIS5513: SiS 962/963 MuTIOL IDE UDMA133 controller
   ide0: BM-DMA at 0xf000-0xf007, BIOS settings: hda:DMA, hdb:pio
   ide1: BM-DMA at 0xf008-0xf00f, BIOS settings: hdc:pio, hdd:pio
hda: WDC WD1200BB-00DWA0, ATA DISK drive
Using anticipatory io scheduler
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
hda: max request size: 1024KiB
hda: Host Protected Area detected.
   current capacity is 234439535 sectors (120033 MB)
   native  capacity is 234441648 sectors (120034 MB)
hda: 234439535 sectors (120033 MB) w/2048KiB Cache, CHS=16383/255/63, 
UDMA(100)
/dev/ide/host0/bus0/target0/lun0: p1 p2 p3 p4 < p5 >
HPT366: IDE controller at PCI slot :00:09.0

I'll paste the full set at the end.
Also, i suppose that this may be the result of the libataization of said
driver, could that be possible ? 
 

I presume you're directing this at "not me?"
Here is the full set of kernel messages:
er 11: 16384 bytes)
Detected 1407.145 MHz processor.
Using pmtmr for high-res timesource
Console: colour dummy device 80x25
Dentry cache hash table entries: 131072 (order: 7, 524288 bytes)
Inode-cache hash table entries: 65536 (order: 6, 262144 bytes)
Memory: 511428k/524224k available (1527k kernel code, 12044k reserved, 
694k data, 148k init, 0k highmem)
Checking if this processor honours the WP bit even in supervisor mode... Ok.
Calibrating delay loop... 2793.47 BogoMIPS
Security Scaffold v1.0.0 initialized
Mount-cache hash table entries: 512 (order: 0, 4096 bytes)
CPU: After generic identify, caps: 0183fbff c1c7fbff  
CPU: After vendor identify, caps:  0183fbff c1c7fbff  
CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
CPU: L2 Cache: 256K (64 bytes/line)
CPU: After all inits, caps:0183fbff c1c7fbff  0020
Intel machine check architecture supported.
Intel machine check reporting enabled on CPU#0.
CPU: AMD Athlon(tm) processor stepping 04
Enabling fast FPU save and restore... done.
Checking 'hlt' instruction... OK.
enabled ExtINT on CPU#0
ESR value before enabling vector: 
ESR value after enabling vector: 
ENABLING IO-APIC IRQs
init IO_APIC IRQs
IO-APIC (apicid-pin) 2-0, 2-16, 2-17, 2-18, 2-19, 2-20, 2-21, 2-22, 
2-23 not connected.
..TIMER: vector=0x31 pin1=2 pin2=-1
Using local APIC timer interrupts.
calibrating APIC timer ...
. CPU clock speed is 1406.0582 MHz.
. host bus clock speed is 267.0920 MHz.
checking if image is initramfs...it isn't (ungzip failed); looks like an 
initrd
Freeing initrd memory: 4492k freed
NET: Registered protocol family 16
PCI: PCI BIOS revision 2.10 entry at 0xfa600, last bus=1
PCI: Using configuration type 1
mtrr: v2.0 (20020519)
ACPI: Subsystem revision 20040326
ACPI: Interpreter enabled
ACPI: Using IOAPIC for interrupt routing
ACPI: PCI Root Bridge [PCI0] (00:00)
PCI: Probing PCI hardware (bus 0

Re: RE/FW: Please stop sending me emails

[John Summerfield]

Frank Carmickle wrote:
I apologize.  But I did remove myself twice from this list in the last few
days.  Then I removed the whitelist entry for this list.  I thought that I
was unsubscribed when I removed the entry but no such luck.  Someone
should look in to this problem.  Neither the web form or the message I
sent removed me from the list.
 

Frankie's mail is still broken. I tried to send him this message:
I believe the author of ASK would like you to discuss the problem with you.
with this result:
This message was created automatically by mail delivery software (Exim).
A message that you sent could not be delivered to one or more of its
recipients. This is a permanent error. The following address(es) failed:
 pipe to |/usr/bin/ask.py --loglevel=5 --logfile=/usr/friends/frankiec/ask.log 
--home=/usr/friends/frankiec
   generated by [EMAIL PROTECTED]
   Child process of address_pipe transport returned 100 from command:
   /usr/bin/ask.py
The following text was generated during the delivery attempt:
-- pipe to |/usr/bin/ask.py --loglevel=5 --logfile=/usr/friends/frankiec/ask.log 
--home=/usr/friends/frankiec
  generated by [EMAIL PROTECTED] --
ERROR: /usr/friends/frankiec/.askrc does not exist or is unreadable. Exiting...
Attention:
The system could not deliver your message due to a technical problem.
Information about the problem has been recorded locally for analysis.
--- Problem Details ---
Traceback (most recent call last):
 File "/usr/bin/ask.py", line 65, in ?
   config = askconfig.AskConfig(sys.argv)
 File "/usr/lib/ask/askconfig.py", line 91, in __init__
   self.__read_config(self.home + "/.askrc")
 File "/usr/lib/ask/askconfig.py", line 104, in __read_config
   sys.exit(self.RET_PROCMAIL_CONTINUE)
SystemExit: 0
---
-- This is a copy of the message, including all the headers. --

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: RE/FW: Please stop sending me emails

[John Summerfield]

peter green wrote:
it may have been caused by admin action on his side
during the height of the problem i sent a rather strongly worded email to
[EMAIL PROTECTED] about the issue and it may be that they have taken action to
disable his broken mail system
 

I also send mail to abuse@ several relevant places, including a real 
name (in case our friend was abuse) that I obtained with whois.

However, this error arose through broken Python code (which may be ASK); 
I'd expect administrative action might result in ASK being banned for 
the moment.

The author thinks the problem should not happen (though it's worth 
noting that the version in use is betaware),  and ask whether F changed. 
Unless F's mail works, he's not going to find out.

I'm cc'ing abuse so he knows there's an ongoing problem. Doubtless Abuse 
can get all the details he wants from the archive. I've still got a few 
hundred originals he can have if he needs more.


--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Apology re mail probs

[John Summerfield]

The author of the program has asked me to offer his apologies. Full 
details attached (unless there's an attachment stripper in effect).

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--- Begin Message ---
Hi John,

Now it seems clear to me what happened. 

As I said, ASK knows how to avoid mail-loops with the challenges. If a number
of emails is received from the same sender in a given amount of time, ASK
will refuse to send further challenges to that address in a while. This is in
place and works pretty well.

ASK also has the concepts of "lists". It works with a "whitelist" a
"blacklist" and an "ignorelist". The difference between black and ignore is
that black will send back an email saying "Please stop sending me emails".

It looks like Frank got some spam coming from the list and added the list
address to his blacklist without getting out of the list first.  This caused
the nastygram sent by his blacklist to cycle over and over, causing the
problem.

I have to admit I never conceived this possibility. The blacklist is to
be used sparingly and very cautiously. In any case, I'll be removing
this feature from ASK for the future versions. I think ignoring spam is
good enough.

Hopefully Frank will read this email and remove the list address from
his blacklist.

Please distribute my apologies to the people on the list. I'll make sure this
feature gets removed from ASK.

Regards,
Paga

On Sun, Aug 29, 2004 at 01:04:05PM +0800, John wrote:
> Marco Paganini wrote:
> >Hi John,
> >
> >
> >>How many would you like? I've got 500 in my trash folder maked "unread" 
> >>so they're easy to find. Those are the ones that arrived after I cleaned 
> >>out the first batch and read legitimate email.
> >
> >
> >Can you send me some 10 or 20 of them? I need to see why this is
> >happening.
> >
> 
> 40?



-- 
Marco Paganini  | UNIX / Linux / Networking
[EMAIL PROTECTED]   | PGP: http://www.paganini.net/pgp/
http://www.paganini.net | Magnus Frater te spectat...

--- End Message ---

Re: [PROPOSAL] 2.4.27 as default 2.4 kernel for sarge

[John Summerfield]

Sven Luther wrote:
It's a self-built kernel, but without modification.
I neglected to include a modules listing in the bug report. Here are 
those myst likely to be involved:
ide_generic 1472  0
sis551316776  1
hpt366 22788  2
ide_disk   19264  5 hpt366
ide_core  138776  5 
usb_storage,ide_generic,sis5513,hpt366,ide_disk
sd_mod 21632  0
ata_piix8132  0
libata 41348  1 ata_piix
scsi_mod  124620  3 usb_storage,sd_mod,libata
unix   28592  170
 

   

I wonder if your ide chipset got correctly detected (aparently), and if dma
and other such parameters are identic.

 

The sis5513 is plausible, I know it is used for earlier SiS chipsets.
Here are the kernel messages I think relevant:
PCI: Probing PCI hardware (bus 00)
Uncovering SIS963 that hid as a SIS503 (compatible=0)
Enabling SiS 96x SMBus.
ACPI: PCI Interrupt Routing Table [\_SB_.PCI0._PRT]
   

I asked because another guy (with a piix chipset though), was claiming that
his chipset was not detected, and thus that dma was not activated.
 

I clean forgot to explicitly affirm that it has DMA on:
kowari:~# hdparm /dev/hda
/dev/hda:
multcount=  0 (off)
IO_support   =  0 (default 16-bit)
unmaskirq=  0 (off)
using_dma=  1 (on)
keepsettings =  0 (off)
readonly =  0 (off)
readahead= 256 (on)
geometry = 16383/255/63, sectors = 234439535, start = 0
kowari:~#
I prefer u1 c1 but generally those make little difference.
kowari:~# hdparm -t /dev/hda
/dev/hda:
Timing buffered disk reads:   92 MB in  3.04 seconds =  30.23 MB/sec
kowari:~# hdparm -t /dev/hda{,,}
/dev/hda:
Timing buffered disk reads:  112 MB in  3.01 seconds =  37.17 MB/sec
/dev/hda:
Timing buffered disk reads:  108 MB in  3.05 seconds =  35.43 MB/sec
/dev/hda:
Timing buffered disk reads:  108 MB in  3.02 seconds =  35.73 MB/sec
kowari:~#
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [PROPOSAL] 2.4.27 as default 2.4 kernel for sarge

[John Summerfield]

Sven Luther wrote:
It's a self-built kernel, but without modification.
I neglected to include a modules listing in the bug report. Here are 
those myst likely to be involved:
ide_generic 1472  0
sis551316776  1
hpt366 22788  2
ide_disk   19264  5 hpt366
ide_core  138776  5 
usb_storage,ide_generic,sis5513,hpt366,ide_disk
sd_mod 21632  0
ata_piix8132  0
libata 41348  1 ata_piix
scsi_mod  124620  3 usb_storage,sd_mod,libata
unix   28592  170
 

   

I wonder if your ide chipset got correctly detected (aparently), and if dma
and other such parameters are identic.

 

The sis5513 is plausible, I know it is used for earlier SiS chipsets.
Here are the kernel messages I think relevant:
PCI: Probing PCI hardware (bus 00)
Uncovering SIS963 that hid as a SIS503 (compatible=0)
Enabling SiS 96x SMBus.
ACPI: PCI Interrupt Routing Table [\_SB_.PCI0._PRT]
   

I asked because another guy (with a piix chipset though), was claiming that
his chipset was not detected, and thus that dma was not activated.
 

I clean forgot to explicitly affirm that it has DMA on:
kowari:~# hdparm /dev/hda
/dev/hda:
multcount=  0 (off)
IO_support   =  0 (default 16-bit)
unmaskirq=  0 (off)
using_dma=  1 (on)
keepsettings =  0 (off)
readonly =  0 (off)
readahead= 256 (on)
geometry = 16383/255/63, sectors = 234439535, start = 0
kowari:~#
I prefer u1 c1 but generally those make little difference.
kowari:~# hdparm -t /dev/hda
/dev/hda:
Timing buffered disk reads:   92 MB in  3.04 seconds =  30.23 MB/sec
kowari:~# hdparm -t /dev/hda{,,}
/dev/hda:
Timing buffered disk reads:  112 MB in  3.01 seconds =  37.17 MB/sec
/dev/hda:
Timing buffered disk reads:  108 MB in  3.05 seconds =  35.43 MB/sec
/dev/hda:
Timing buffered disk reads:  108 MB in  3.02 seconds =  35.73 MB/sec
kowari:~#
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [PROPOSAL] 2.4.27 as default 2.4 kernel for sarge

[John Summerfield]

Sven Luther wrote:
It's a self-built kernel, but without modification.
I neglected to include a modules listing in the bug report. Here are 
those myst likely to be involved:
ide_generic 1472  0
sis551316776  1
hpt366 22788  2
ide_disk   19264  5 hpt366
ide_core  138776  5 
usb_storage,ide_generic,sis5513,hpt366,ide_disk
sd_mod 21632  0
ata_piix8132  0
libata 41348  1 ata_piix
scsi_mod  124620  3 usb_storage,sd_mod,libata
unix   28592  170
 

   

I wonder if your ide chipset got correctly detected (aparently), and if dma
and other such parameters are identic.

 

The sis5513 is plausible, I know it is used for earlier SiS chipsets.
Here are the kernel messages I think relevant:
PCI: Probing PCI hardware (bus 00)
Uncovering SIS963 that hid as a SIS503 (compatible=0)
Enabling SiS 96x SMBus.
ACPI: PCI Interrupt Routing Table [\_SB_.PCI0._PRT]
   

I asked because another guy (with a piix chipset though), was claiming that
his chipset was not detected, and thus that dma was not activated.
 

I clean forgot to explicitly affirm that it has DMA on:
kowari:~# hdparm /dev/hda
/dev/hda:
multcount=  0 (off)
IO_support   =  0 (default 16-bit)
unmaskirq=  0 (off)
using_dma=  1 (on)
keepsettings =  0 (off)
readonly =  0 (off)
readahead= 256 (on)
geometry = 16383/255/63, sectors = 234439535, start = 0
kowari:~#
I prefer u1 c1 but generally those make little difference.
kowari:~# hdparm -t /dev/hda
/dev/hda:
Timing buffered disk reads:   92 MB in  3.04 seconds =  30.23 MB/sec
kowari:~# hdparm -t /dev/hda{,,}
/dev/hda:
Timing buffered disk reads:  112 MB in  3.01 seconds =  37.17 MB/sec
/dev/hda:
Timing buffered disk reads:  108 MB in  3.05 seconds =  35.43 MB/sec
/dev/hda:
Timing buffered disk reads:  108 MB in  3.02 seconds =  35.73 MB/sec
kowari:~#
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: enabling CONFIG_IDEDMA_ONLYDISK

[John Summerfield]

Frederik Schueler wrote:
Hi,
On Sun, Aug 29, 2004 at 12:00:07PM +0200, Christoph Hellwig wrote:
 

We have tons of reports for broken CDs in the BTS, and I noticed that
unlike the commercial distibution vendors we don't have
CONFIG_IDEDMA_ONLYDISK set ibn our install kernels.  I'd rather play
safe and slow in this case.
   

I suggest setting this only on the generic installer kernels, so users
can install the system and activate DMA for their drives later when
installing the optimized kernel.
 

I would prefer installing hdparm, with a default config file that turns 
on appropriate options for /dev/hda which surely is not a CD drive, and 
with comments describing how to set options for other drives.

It should be run at boot time with a script in /etc/init.d, but not run 
by default.

Ensuring the installer knows what to do by sending email to root is a 
sound final touch.

As a user who installs his own software, find there are many things I 
should know to do but don't and that extends to software that is really 
useful and that is often not installed, but I think should be. In this 
category I include
hdparm
smartmontools
lmsensors

The first two are essential for anyone who has disk drives, the third 
can warn you you're cooking your computer.


--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

d-i and USB drives

[John Summerfield]

I'm playing with one of my toys: I have a Laptop drive, and I'm booting 
all the installers I can lay my hands on to see what can't find it.

d-i does (I think it's the April beta), but calling it "SCSI" is going 
to cause confusion, esp with any new to Linux.

I like the fact you give some info about the drives: I've just failed 
the FreeBSD installer because it gives me the choice of "ad0" and "da0," 
with no explanation.

However, I expect a n00b would be confused with either.


--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Kernel probs with ATA disk

[John Summerfield]

I've just submitted two kernel bug reports that have the potential to be 
release-critical:
2.6.7 kernel can't detect my (new) ATA drive on my Pentium III system. 
See #270198
2.4.26-1-686 drives the ATA drive very slowly as it doesn't allow me to 
turn DMA on. See #270199

I've not tried d-i on this system, it's my server (including boot 
server), but I thought you'd like to know there is a problem. Or two.

I have the same kernel on other system using (maybe not precisely) the 
same chipset and other drives wthout these problems.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Kernel probs with ATA disk

[John Summerfield]

Joey Hess wrote:
([EMAIL PROTECTED] may be an amusing reply-to, but it is not appropriate
for this list. A great many posters on this list are not subscribers and
I CC anyone I do not know is a subscriber.)
 

A great number of folk CC me without regard for what they know about me. 
It stands.

John Summerfield wrote:
 

I've just submitted two kernel bug reports that have the potential to be 
release-critical:
2.6.7 kernel can't detect my (new) ATA drive on my Pentium III system. 
See #270198
2.4.26-1-686 drives the ATA drive very slowly as it doesn't allow me to 
turn DMA on. See #270199
   

None of these are kernels that we're using in the current version of the
installer. I suggest you try the newer kernel versions.
 

2.6.7 latest I can find for Sarge.
Package kernel-image-2.6 is a virtual package provided by:
 kernel-image-2.6.3-1-686 2.6.3-2
 kernel-image-2.6.7-1-k7-smp 2.6.7-2
 kernel-image-2.6.7-1-k7 2.6.7-2
 kernel-image-2.6.7-1-686-smp 2.6.7-2
 kernel-image-2.6.7-1-686 2.6.7-2
 kernel-image-2.6.7-1-386 2.6.7-2
 kernel-image-2.6.5-1-k7-smp 2.6.5-4
 kernel-image-2.6.5-1-k7 2.6.5-4
 kernel-image-2.6.5-1-686-smp 2.6.5-4
 kernel-image-2.6.5-1-686 2.6.5-4
 kernel-image-2.6.5-1-386 2.6.5-4
You should explicitly select one to install.
E: Package kernel-image-2.6 has no installation candidate
ns:~# apt-get install kernel-image-2.6.7-1-686
Reading Package Lists... Done
Building Dependency Tree... Done
kernel-image-2.6.7-1-686 is already the newest version.
0 upgraded, 0 newly installed, 0 to remove and 37 not upgraded.
I'm working on kernel-image-2.4.27-1-686


--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Kernel probs with ATA disk

[John Summerfield]

Joey Hess wrote:
John Summerfield wrote:
 

A great number of folk CC me without regard for what they know about me. 
It stands.
   

And you don't even set a proper Mail-Followup-To.
 

Settle down Joey. AFAIK my MUA doesn't offer the facility. It does allow 
me to make a choice that reduces the amount of unwanted mail I get, so I 
did it. You're the first to make a fuss over it.

fwiw 2.4.27-1-686 also sucks:
ns:~# hdparm -d1 /dev/hda
/dev/hda:
setting using_dma to 1 (on)
HDIO_SET_DMA failed: Operation not permitted
using_dma=  0 (off)
ns:~# hdparm -t /dev/hda{,}
/dev/hda:
Timing buffered disk reads:   12 MB in  3.46 seconds =   3.47 MB/sec
/dev/hda:
Timing buffered disk reads:   12 MB in  3.45 seconds =   3.48 MB/sec
ns:~#
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Boot grub on floppy vs on MBR

[John Summerfield]

Haines Brown wrote:
I'm running debian sarge and trying to make a grub boot floppy for a
new cross-installation of sarge from my old disk (sda, sarge,
kernel 2.4.18-bf2.4) to a new scsi disk added to the scsi bus (sdc,
sarge, kernel 2.6.7-1). The sdc1 root partition is flagged as
boot. I'm using reiserfs on both hard disks, but not the floppy.
At this point, I created a grub boot floppy in preparation for
installing grub on sdc. That floppy boots sda OK, but only leads to a
kernel panic when I try to boot sdc, even though the floppy stanza 
is the same as the stanza used to boot sdc from the grub bootloader
the MBR of sda (except it's hd0 for sda1):

   kernel /vmlinuz ro root=/dev/sdc1
   root (hd2,0)
Kernel panic: 

   VFS: Cannot open root device "sdc1" or 08:21
   Please append correct "root=" option
   Kernel panic: VFS: Unable to mount rootfs on 08:21
Apparently grub can't see /dev/sdc1 even though from sda I can do:  
 

Yer joking.  How did it get there if grub can't read it?
Better look to see why the kernel doesn't comprehend the partition.

--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Two Macedonias

[John Summerfield]

Konstantinos Margaritis wrote:
On ÎÎÎ 23 ÎÎÏ 2004 11:44, Christian Perrier wrote:
 

2) use the names 'Republic Macedonia' and 'Greek Macedonia'.
   

I like that one, but, as Steve Langasek said, in the namespace of
autonomous territories, there is no collision at all.
 

Well, in my opinion, if we decide changing FYROM to something in
iso-codes (and thus, as a direct consequence, in debian-installer),
we should use "Macedonia".
And, well, I'm pretty sure that most Greek users indeed share this
advice. I'm already sure of this for one of them..:)
   

Hm, I can't say I'm overjoyed with this choice. If this needs to be 
changed, I'd prefer the distiction to be made, 'Republic of 
Macedonia' would be far better and avoid unwanted comments from both 
sides.

Konstantinos
 

I think the style in Australia is "The former Republic of Macedonia." 
I've always though it peculiar, but I think our Greeks are happy with 
it, and the second bigest Greek city in the world is in Australia.

Macedonia is a town in Victoria:-)
--
Cheers
John
-- spambait
[EMAIL PROTECTED]  [EMAIL PROTECTED]
Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]