Debian Bug report logs - #281407
loops forever

Toggle useless messages

Message #5 received at submit@bugs.debian.org (full text, mbox, reply):

From: Joey Hess <joeyh@debian.org>

To: Debian Bug Tracking System <submit@bugs.debian.org>

Subject: loops forever

Date: Mon, 15 Nov 2004 12:09:02 -0500

[Message part 1 (text/plain, inline)]

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
the dasdfmt command by hand, and it succeeds.

~ # cat /proc/dasd/devices 
0120(ECKD) at ( 94:  0) is dasda      : active at blocksize: 4096, 216000 blocks, 843 MB
~ # dmesg |tail
dasd(eckd): /dev/dasda  ( 94:  0),0120@01: volume analysis returned unformatted disk
dasd: waiting for responses...
dasd(eckd): /dev/dasda  ( 94:  0),0120@01: (4kB blks): 864000kB at 48kB/trk compatible disk layout
Partition check:
 /dev/dasd/0120:(nonl)/        : p1
 /dev/dasd/0120:VOL1/  LX0120:
dasd: waiting for responses...
dasd(eckd): /dev/dasda  ( 94:  0),0120@01: (4kB blks): 864000kB at 48kB/trk compatible disk layout
 /dev/dasd/0120:(nonl)/        : p1
 /dev/dasd/0120:VOL1/  LX0120:
~ # cat /proc/subchannels 
Device sch.  Dev Type/Model CU  in use  PIM PAM POM CHPIDs
---------------------------------------------------------------------
000C   0000  3505/01   2821/01          80  80  FF  00000000 00000000
0120   0001  3390/06   3990/C2  yes     80  80  FF  01000000 00000000
0A00   0002  3088/01   3088/08  yes     80  80  FF  0A000000 00000000
0A01   0003  3088/01   3088/08  yes     80  80  FF  0A000000 00000000

I'm installing in hercules, configured as follows:

CPUSERIAL	002623
CPUMODEL	3090
MAINSIZE	64
CODEPAGE	default
XPNDSIZE	0
#CNSLPORT	3270
#HTTPPORT	8081 noauth userid password
#HTTPROOT	/usr/share/hercules/
NUMCPU		1
LOADPARM	0120....
OSTAILOR	LINUX
PANRATE		SLOW
ARCHMODE	ESA/390

000C	3505	/home/joey/kernel.debian /home/joey/parmfile.debian /home/joey/initrd.debian autopad eof
0120	3390	/home/joey/html/d-i/state/hercules-hercules-common.img
0A00	3088	CTCI /dev/net/tun 1500 192.168.42.1 192.168.42.2 255.255.255.0
0A01	3088	CTCI /dev/net/tun 1500 192.168.42.1 192.168.42.2 255.255.255.0

My dasd file was created with the command dasdinit file 3390-1 root 200.

I'm installing with the udebs from testing.

-- 
see shy jo

[signature.asc (application/pgp-signature, inline)]

Message #10 received at 281407@bugs.debian.org (full text, mbox, reply):

From: Joey Hess <joeyh@debian.org>

To: 281407@bugs.debian.org

Subject: notes

Date: Mon, 15 Nov 2004 13:24:17 -0500

[Message part 1 (text/plain, inline)]

- Seems to be caused by a bad state transition, untested fix in svn.
- Workaround is to ssh in again to the shell, remove the postinst script
  that is running and kill it, when it begins to loop.

-- 
see shy jo

[signature.asc (application/pgp-signature, inline)]

Message #15 received at 281407@bugs.debian.org (full text, mbox, reply):

From: Adam Thornton <athornton@sinenomine.net>

To: Joey Hess <joeyh@debian.org>, 281407@bugs.debian.org

Subject: Re: Bug#281407: loops forever

Date: Mon, 15 Nov 2004 12:29:08 -0600

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
> the dasdfmt command by hand, and it succeeds.
>
That's odd.  I've seen dasdfmt refuse to touch virgin DASD before--I 
usually either CMS format it, or use ICKDSF to format the first track 
of it--but in that case a manual dasdfmt would also fail.  And if you 
created your DASD with Hercules dasdinit, then it should already have a 
CDL-compatible format on it (since dasdinit does z/OS formatting).  
Still, let's try a manual initialization with ICKDSF and see if 
anything changes.  I believe that ickdsf.ipl is shipped with Hercules.

Note: the following procedure will *destroy* your 120 disk; you may 
want to try it on a new volume rather than anything you've put time 
into.

Try defining ickdsf.ipl to Hercules as a 3420 device.  So in your 
Hercules.cnf, put

0580    3420    ickdsf.ipl

I *thought* ICKDSF would work from the HMC, but it isn't for me, so 
define a 3270 console

001F  3270

And uncomment CNSLPORT 3270

Then start hercules, connect a 3270 emulator to port 3270, and IPL 580.

Go to your 3270 console.

Hit Enter.

You'll be prompted to hit Clear.  Do so.

You should see something like

"ICK030E DEFINE INPUT DEVICE....."

Type "CONSOLE"  (the dot passes the command to the Hercules 
HMC-equivalent).

CONSOLE should be your output device too.

Then Initialize the volume:

INIT UNITADDR(0120) NOVERIFY

And type "U" to confirm.

This should initialize the DASD volume to prepare it for herc.

Then at the herc console, type STOP, and then IPL 000C (btw: cool!  I 
always used TDF for herc installs; it's nice to know that Hercules will 
do card reader IPLs too!).

See if you get any farther.

If *not*, do the same thing, but after the INIT, try

CPVOL FORMAT UNIT(0120) NOVERIFY

and type "U" to confirm, before stopping the processor.

I'm testing this myself, but on a dual-processer P3-833 that's already 
running a lot of work, it may take me a while.

Adam

Message #20 received at 281407@bugs.debian.org (full text, mbox, reply):

From: Joey Hess <joeyh@debian.org>

To: Adam Thornton <athornton@sinenomine.net>

Cc: 281407@bugs.debian.org

Subject: Re: Bug#281407: loops forever

Date: Mon, 15 Nov 2004 14:01:41 -0500

Adam Thornton wrote:
> That's odd.  I've seen dasdfmt refuse to touch virgin DASD before--I 
> usually either CMS format it, or use ICKDSF to format the first track 
> of it--but in that case a manual dasdfmt would also fail.  And if you 
> created your DASD with Hercules dasdinit, then it should already have a 
> CDL-compatible format on it (since dasdinit does z/OS formatting).  
> Still, let's try a manual initialization with ICKDSF and see if 
> anything changes.  I believe that ickdsf.ipl is shipped with Hercules.
> 
> Note: the following procedure will *destroy* your 120 disk; you may 
> want to try it on a new volume rather than anything you've put time 
> into.

Since I've found and I think fixed the loop, I have not tried this
procedure yet. However I don't know why everyone hasn't been seeing the
loop, I assume I'm hitting some unusual case in s390-dasd.

> Then at the herc console, type STOP, and then IPL 000C (btw: cool!  I 
> always used TDF for herc installs; it's nice to know that Hercules will 
> do card reader IPLs too!).

Can you give me the recipe for booting d-i from tape in hercules?

> I'm testing this myself, but on a dual-processer P3-833 that's already 
> running a lot of work, it may take me a while.

Yeah, that's my situation, I'm installing debian, so I can build, test,
and release my s390-dasd loop fix.

-- 
see shy jo

Message #25 received at 281407@bugs.debian.org (full text, mbox, reply):

From: Joey Hess <joeyh@debian.org>

To: 281407@bugs.debian.org

Subject: FWD: Re: Bug#281407: loops forever

Date: Mon, 15 Nov 2004 18:16:03 -0500

[Message part 1 (text/plain, inline)]

----- Forwarded message from Adam Thornton <athornton@sinenomine.net> -----

From: Adam Thornton <athornton@sinenomine.net>
Date: Mon, 15 Nov 2004 16:20:03 -0600
To: Joey Hess <joeyh@debian.org>
Subject: Re: Bug#281407: loops forever
X-Mailer: Apple Mail (2.619)

Yep.

The loop happens if you have exactly one DASD device.

This probably hasn't been reported before because anyone who runs under 
VM probably has at least a 190 (system) and 191 (user A-disk) disk 
defined as well, and for Hercules I usually define at least a swap 
device as well as the root disk.

My Linux guests usually have a 191 for CMS configuration (I use CMS as 
a smart BIOS, basically), a VDISK for swap, and the root disk, as well 
as links to CMS system programs and tools.

Adam


----- End forwarded message -----

-- 
see shy jo

[signature.asc (application/pgp-signature, inline)]

Message #32 received at 281407-done@bugs.debian.org (full text, mbox, reply):

From: Joey Hess <joeyh@debian.org>

To: 281407-done@bugs.debian.org

Subject: FWD: Fixed in NMU of s390-dasd 0.0.9

Date: Tue, 16 Nov 2004 00:18:16 -0500

[Message part 1 (text/plain, inline)]

NAN

----- Forwarded message from Joey Hess <joeyh@debian.org> -----

From: Joey Hess <joeyh@debian.org>
Date: Mon, 15 Nov 2004 23:32:02 -0500
To: control@bugs.debian.org
Cc: Joey Hess <joeyh@debian.org>,
	Debian Install System Team <debian-boot@lists.debian.org>
Subject: Fixed in NMU of s390-dasd 0.0.9

tag 281407 + fixed

quit

This message was generated automatically in response to a
non-maintainer upload.  The .changes file follows.

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Format: 1.7
Date: Mon, 15 Nov 2004 12:35:50 -0500
Source: s390-dasd
Binary: s390-dasd
Architecture: source s390
Version: 0.0.9
Distribution: unstable
Urgency: low
Maintainer: Debian Install System Team <debian-boot@lists.debian.org>
Changed-By: Joey Hess <joeyh@debian.org>
Description: 
 s390-dasd  - Configure DASD (udeb)
Closes: 281407
Changes: 
 s390-dasd (0.0.9) unstable; urgency=low
 .
   * Joey Hess
     - Fix state transition from confirm to avoid loop. Closes: #281407
   * Updated translations:
     - Bulgarian (bg.po) by Ognyan Kulev
     - Catalan (ca.po) by Jordi Mallach
     - Czech (cs.po) by Miroslav Kure
     - Welsh (cy.po) by Dafydd Harries
     - Danish (da.po) by Claus Hindsgaul
     - German (de.po) by Dennis Stampfer
     - Greek (el.po) by Greek Translation Team
     - Spanish (es.po) by Javier Fernandez-Sanguino Peña
     - Basque (eu.po) by Piarres Beobide Egaña
     - Finnish (fi.po) by Tapio Lehtonen
     - French (fr.po) by French Team
     - Hebrew (he.po) by Lior Kaplan
     - Croatian (hr.po) by Krunoslav Gernhard
     - Hungarian (hu.po) by VEROK Istvan
     - Indonesian (id.po) by Debian Indonesia Team
     - Italian (it.po) by Giuseppe Sacco
     - Japanese (ja.po) by Kenshi Muto
     - Korean (ko.po) by Changwoo Ryu
     - Lithuanian (lt.po) by Kęstutis Biliūnas
     - Latvian (lv.po) by Aigars Mahinovs
     - Norwegian (nb.po) by Bjorn Steensrud
     - Dutch (nl.po) by Bart Cornelis
     - Norwegian (nn.po) by Håvard Korsvoll
     - Polish (pl.po) by Bartosz Fenski
     - Portuguese (Brazil) (pt_BR.po) by André Luís Lopes
     - Romanian (ro.po) by Eddy Petrisor
     - Russian (ru.po) by Russian L10N Team
     - Slovak (sk.po) by Peter KLFMANiK Mann
     - Slovenian (sl.po) by Jure Čuhalev
     - Albanian (sq.po) by Elian Myftiu
     - Swedish (sv.po) by Per Olofsson
     - Turkish (tr.po) by Recai Oktaş
     - Ukrainian (uk.po) by Eugeniy Meshcheryakov
     - Simplified Chinese (zh_CN.po) by Carlos Z.F. Liu
     - Traditional Chinese (zh_TW.po) by Tetralet
Files: 
 d61545be7394c546b22d407ff0a0c4b3 653 debian-installer standard s390-dasd_0.0.9.dsc
 d4cc81d3247c1a43bfedfa398e9abb8a 32579 debian-installer standard s390-dasd_0.0.9.tar.gz
 14cb01b91460e797bda158836a4f0fc9 24678 debian-installer standard s390-dasd_0.0.9_s390.udeb
package-type: udeb

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (GNU/Linux)

iD8DBQFBmXVq2tp5zXiKP0wRArRcAJoCWw7cC0sSsOHwjNnG27S4T7FGhQCgnWVA
zy4Ihy22WKatoXkkWhvHjb0=
=BJcI
-----END PGP SIGNATURE-----


----- End forwarded message -----

-- 
see shy jo

[signature.asc (application/pgp-signature, inline)]

Send a report that this bug log contains spam.