Package: debian-installer
Severity: important
S390 boot fails (sometimes just after detecting memory, sometimes after
detecting devices) on z/VM 5.2 and a Flex-ES machine.
Since there are reports of d-i working on z9 boxes, I suspect the issue
is that the kernel is built to exploit later System
On Mar 22, 2007, at 1:09 PM, Kapil Hari Paranjape wrote:
On Thu, 22 Mar 2007, Martin Michlmayr wrote:
You can simply start a 2nd connection via SSH and then open a
terminal.
Perhaps I'm mistaken. I did try this as installer with password
install but I seem to recall that I was logged out.
I tried a few days ago to install Etch on an s390 system, and failed
to partition my DASD because kernel modules appropriate to the
installer kernel level weren't found. I presume this is a known
issue that will be resolved when RC1 rolls out and is just version
skew between testing and
Package: debian-installer
Severity: important
In earlier versions of the Sarge installer, when you selected a qeth
device which was a HiperSockets transport (Layer 3 only) coupled to a VM
guest LAN (or, probably, to a real HiperSockets trans-LPAR connector,
but I don't have any way to test that),
Package: netinst
Version: base-network
Severity: important
Using etch Beta 2: installation goes OK, but network device is not picked
up on reboot. This seems to be because alias eth0 qeth is commented out
in /etc/modules.conf, and although qeth is present in /etc/modules, it is
not loaded at
Package: debian-installer
Severity: minor
Tags: l10n
If you do an S/390 installation (I'm reporting this from an ix86
machine, but the problem only occurs with a Hipersocket installation,
which means S/390), and you have a Hipersocket device, that's
incorrectly spelled Hypersocket in the
On Nov 17, 2004, at 12:25 PM, Joey Hess wrote:
Frans Pop wrote:
There is a very nice workaround for this.
If 'debian-installer/locale=C' is added in parmfile.debian, all
problems
in the current prebaseconfig of countrychooser will be avoided.
I would suggest including in the Release Notes and
On Nov 16, 2004, at 10:57 AM, Eugeniy Meshcheryakov wrote:
After using google.com I found that I need to pass dasd=0120 to zipl
in order
to boot. That worked. It is posible that dasd parameter is missing
because I
did not use d-i for formating dasd and partitioning.
I believe this is the case.
A
the Linux/390 list, something that might help:
Adam Thornton [EMAIL PROTECTED] wrote:
This just in from d-i.
Anyone here know anything about why the kernel doesn't set environment
variables from the command line?
Begin forwarded message:
PPID='167'
Command line environment variables are only passed
On Nov 14, 2004, at 4:48 PM, Joey Hess wrote:
Apparently the s390 kernel does not parse the kernel command line for
environment variables. Here I'm in hercules and I've put foo=bar in
parmfile.debian, which adds it to the kernel command line:
I've forwarded this to the Marist Linux/390 list, which
On Nov 15, 2004, at 11:09 AM, Joey Hess wrote:
Package: s390-dasd
Severity: normal
I see a loop in s390-dasd, it asks me if I want to format device 0120,
I say yet, it formats it and asks me again. Over and over. If I say no,
it just asks me again. The back button has no effect. I've tried to run
Package: Debian-installer
Version: RC1 card decks, whatever's on ftp.debian.org on 20041015
Severity: grave
Justification: renders package unusable
If you try to install, using the Debian-installer RC1 boot card decks,
from ftp.us.debian.org (on 15 Oct 2004), d-i tries to install the s390x
On Wed, 2004-09-08 at 03:49, [EMAIL PROTECTED] wrote:
There are SCSI controller from adaptec - module AIC7xxx.
During detecting hardware I aic7xxx loading very long time. On 2% detecting hardware
fails with message Cnat find debian installer cd
Woody installer pass this stage successfully.
On Thu, 2004-08-26 at 16:16, Sam Liddicott wrote:
I guess these means the debian installer would have to be content to see
some block devices on which to build a file system and not insist so
intently on there being a partition table available
This would be the right answer for FBA under
On Mon, 2004-08-09 at 11:01, Joey Hess wrote:
John Summerfield wrote:
Some of these may apply to otherr arches too. eg kernel version.
Floppies on ?
[...]
We're well aware that the installation manual is not accurate for many
arches other than i386. The only way to get it fixed is for
Using one of today's svn builds, trying to install S/390 in a low memory
configuration (32M) fails: control characters that don't make sense to
the 3270 terminal are spewed across the screen, effectively obscuring
whatever it's trying to tell me, and the installation appears to hang.
I don't have
On Fri, 2004-07-30 at 14:43, Joey Hess wrote:
Adam Thornton wrote:
Using one of today's svn builds, trying to install S/390 in a low memory
configuration (32M) fails
How realistic is that? I thought s390 were big iron?
Yeah, but if you're running under VM, your virtual machine can
As near as I can tell, S/390 still uses partconf, and it and m68k are
the only architectures to do so.
Something I think we should think about--although not for the d-i or
sarge releases, probably--is to get S/390 moved over to partman (if I
understand correctly that partman's functionality is a
Package: installation-reports
INSTALL REPORT
Debian-installer-version: Built from svn 25 July 2004 1900:00 (or so)
-0500
uname -a: Linux di 2.4.26-1-s390 #1 SMP Sun Apr 25 16:48:39 CEST 2004
s390 GNU/Linux
Date: 25 July 2004 21:15 -0500
Method: IPL from virtual reader. Installed unstable from
As I understand it, the run /sbin/reboot or not flag in d-i is kind of
in flux.
It was put there for embedded devices, because reboot repeats the last
boot you did, which will, in those devices, boot from the wrong thing.
This is also the case in the S/390 install: you will have booted (IPLed,
Package: network-console
Version: 0.03
Severity: wishlist
This is just a minor usability issue: after network-console generates
its host keys and tells you that it's ready for you to SSH into the
install system, the user is then told Press Any Key to Continue, and
you're taken, on the console,
The /etc/fstab written for S/390 assumes old-style, static /dev entries:
/dev/dasda1, /dev/dasdb1, and so on.
Unfortunately, the installed system does not have those device nodes,
but instead has devfs: /dev/dasd/address/part1, etc.
The first device (/dev/dasda) works in /etc/fstab and in the
On Fri, 2004-07-23 at 12:23, Thiemo Seufer wrote:
AFAICS the installed system shouldn't mount devfs. It also should have
static /dev entries, managed by makedev.
In that case where in the debian-installer build do I need to put the
script to generate the static /dev entries that the parmfile
On Fri, 2004-07-23 at 13:23, Thiemo Seufer wrote:
Are you sure this doesn't simply hide the static entries by mounting
devfs over it?
Well, no, I'm not. In fact, it probably does, and that's why the very
first /dev/dasda works.
So it looks like the right answer is to rebuild the kernel so as
On Fri, 2004-07-16 at 01:11, Kenshi Muto wrote:
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.
Does not selecting anything in tasksel count as
On Fri, 2004-07-16 at 05:25, John Summerfield wrote:
The problem I had was after the first boot. I regard what happens _before_ then a
separate problem.
What do folk on zSeries do?
Well, I understand that work is progressing on migrating the ssh key and
server from the network-console
On Fri, 2004-07-16 at 09:17, Josselin Mouette wrote:
This could be solved by identifying adapters by their MAC address;
And won't work in the general case for guest LANs under z/VM on S/390:
virtual QDIO devices do not, prior to z/VM 4.4, have unique MAC
addresses, as they're purely Layer 3. In
Package: s390-netdevice
Version: 0.06
Severity: important
Tags: patch
Protocol 1 (Linux) is currently chosen as the default CTC protocol, and
the debconf priority is medium. This means that CTC installation under
VM (usually--if the CTC is defined as vCTCs typically are) would fail,
and CTC
Package: partconf-mkfstab
Version: 0.31
Severity: important
When installing S/390 with debian-installer, I have swap on Partion 1 of
device 0151. The fstab is written to use /dev/dasdb1, but when I IPL, I
get a devfs-style device, /dev/dasd/0151/part1, and there is no
/dev/dasdb1 device node.
Package: s390-netdevice
Version: 0.0.6
Severity: important
Tags: patch
type_text (==iucv) is not the right identifier for doing a modprobe
if the device is IUCV. The iucv module provides the IUCV base
functions, but the netiucv module provides an interface suitable for
treating like a NIC.
On Thu, 2004-07-08 at 16:51, Anibal Monsalve Salazar wrote:
s390
Missing for release of s390: some fixes in zipl-installer, uploads of
the other s390 specific packages.
I have patches, but I don't know if I'm deriving the disk information
the right way.
Adam
---
Package: netcfg-static
Version: 0.72
Severity: important
Tags: patch
IUCV is a point-to-point connection type, but netcfg-static was not
detecting it as such. I have included a patch that fixes this.
-- System Information:
Debian Release: testing/unstable
APT prefers testing
APT policy:
Package: s390-netdevice
Version: 0.05
Severity: important
Tags: patch
I've implemented IUCV support for S/390. It seems to work--I'm doing an
installation with it now.
-- System Information:
Debian Release: testing/unstable
APT prefers testing
APT policy: (700, 'testing')
Architecture: s390
Package: base-installer
Version: 0.090
Severity: important
line 680 of base-installer.postinst should be
if [ $subarch != none ]; then
(the ; then is missing in the downloaded file).
This is the version that is getting downloaded for an unstable build
on S/390 at about 3:00 PM CDT on
On Wed, 2004-07-07 at 15:44, Matt Zimmerman wrote:
Package: di-utils-reboot
Severity: wishlist
Tags: patch
I find it convenient sometimes (especially when using UML) to have d-i halt
the system as its final step, rather than rebooting it.
This will also be very useful for S/390, for the
Yet another tweak to the patch: this one writes module option
information to /etc/modutils/options.
I really don't know where the line
option iucv0 iucv=peer should go, or how that's going to get
automatically moved from the installer system to the installer system.
What's the right answer for
On Fri, Jul 02, 2004 at 02:32:03PM -0400, Martin Michlmayr wrote:
* Adam J. Thornton [EMAIL PROTECTED] [2004-07-02 13:12]:
That is, if you want an s390x install, use a kernel/initrd set for
s390x, and otherwise, use s390. /proc/cpuinfo will tell you the CPU
OK. What about the s390-tape
Package: installation-reports
INSTALL REPORT
Debian-installer-version: svn-up run just before install, and
network-console built at that point; network-console put into
localudebs.
uname -a: Linux di 2.4.26-1-s390 #1 SMP Sun Apr 25 16:48:39 CEST 2004
s390 GNU/Linux
Date: Thursday,
After doing an svn up a couple days ago, I can't build images anymore.
I'm using ftp.us.debian.org as my udeb source. If I have the following
in sources.list.udeb.local, I can't retrieve network-console:
deb ftp://ftp.us.debian.org/debian/ testing main/debian-installer
deb
The first patch I submitted for this did not maintain the order in which
you had attached your DASD, which could lead to problems if you didn't
attach your disks in device address order.
This patch maintains the order in which the DASD was attached. It does
assume four-digit device addresses.
It still doesn't work, but with the new zipl-installer.postinst it comes
a lot closer.
netcfg (actually netcfg-static) still doesn't work the first time. What
you have to do is to select IUCV, which generates an error, and then go
back in from the main menu, select option #3 (network config),
Package: zipl-installer
Severity: grave
Justification: renders package unusable
Tags: sid patch
zipl-installer does not work from d-i installation: /vmlinuz symlink
does not exist, and written parmfile is incorrect.
I have included a patch which addresses these. However, I suspect
there are
On Fri, 2004-06-18 at 15:43, martin f krafft wrote:
Fellow Debianistas,
Why does Debian drop symlinks to vmlinuz and initrd.gz into the root
directory? The FHS does allow it (after all), but it's butt-ugly, if
you ask me.
Because /vmunix is the traditional location of the kernel on Unix and
On Wed, 2004-06-09 at 07:26, Kenshi Muto wrote:
Generally, I believe users will use X Window System or remote shell,
won't use raw console.
This may be true in general; it is definitely not true for S390; almost
everyone will use the console, although that will generally be a tn3270
connection
On Wed, 2004-06-02 at 14:45, David Boyes wrote:
Okay, so the remaining problems are:
- zipl-installer, not available(?)
OK, so this needs to be written/done.
What does this entail? It seems as if it should be a matter of putting
zipl on the base image and just feeding it the right
OK, I've cleared it with management.
There is only one slight wrinkle: I need to get anyone who will have
access to our H70 for doing installer development to sign an NDA.
Before you freak out, let me explain that a little further: it has
nothing whatsoever to do with the Debian work--we'd love
46 matches
Mail list logo
