Message-ID: <OFEDE4C783.5CB26C59-ONC1256ACC.00234CC3@hasler.ascom.ch> From: "Thomas Schmid" <Thomas.Schmid@ascom.ch> Date: Wed, 19 Sep 2001 09:30:13 +0200 Subject: [suse-sparc] Feedback on SuSE-7.3 on Sparc (Beta 1)
Hi,
I have tried SuSE 7.3 Beta 1 on a Sparcstation 5, with 64MB RAM, 2 x 1GB
SCSI-SCA-harddrives,
external SCSI Plextor 8/20 CD-R CD-ROM / -Writer und einem alten Toshiba
4x SCSI CD-ROM:
It is a lengthy report, if you want to know more details, I can try to get
them....
Machine:
========
CPU Info:
cpu : Fujitsu MB86904
fpu : Lsi Logic/Meiko L64804 or compatible
promlib : Version 3 Revision 2
prom : 2.15
type : sun4m
BogoMips : 84.78
MMU type : Fujitsu Swift
ESP SCSI Info:
Sparc ESP Host Adapter:
PROM node ffd3d234
PROM name esp
ESP Model FAS100A
DMA Revision Rev 2
Live Targets [ 1 3 6 ]
Target # config3 Sync Capabilities Disconnect Wide
1 00000003 [2f,04] yes no
3 00000003 [2f,04] yes no
6 00000001 [2f,04] yes no
Harddisk / CD-ROM Info:
<4>esp0: IRQ 36 SCSI ID 7 Clk 40MHz CCF=8 TOut 167 NCR53C9XF(espfast)
<4>ESP: Total of 1 ESP hosts found, 1 actually in use.
<4>scsi0 : Sparc ESP100A-FAST
<4>scsi : 1 host.
<4> Vendor: SEAGATE Model: ST31051W Rev: 0526
<4> Type: Direct-Access ANSI SCSI revision: 02
<4>Detected scsi disk sda at scsi0, channel 0, id 1, lun 0
<4> Vendor: IBM Model: DPES31080SUN1.05 Rev: S80B
<4> Type: Direct-Access ANSI SCSI revision: 02
<4>Detected scsi disk sdb at scsi0, channel 0, id 3, lun 0
<4> Vendor: TOSHIBA Model: CD-ROM XM-5301TA Rev: 0925
<4> Type: CD-ROM ANSI SCSI revision: 02
<4>Detected scsi CD-ROM sr0 at scsi0, channel 0, id 6, lun 0
<4>scsi : detected 1 SCSI cdrom 2 SCSI disks total.
<6>esp0: target 6 asynchronous
Installation #1:
=============
First I tried to install SuSE 7.3 with the following configuration:
- Using both SCSI drives, as SCSI-ID=3 => /dev/hdb as root-drive "/",
SCSI-ID 1 => /dev/hda as "/usr",
installing from the Plextor CD-R SCSI CD-ROM
The installation in yast2 went smooth so far, until it reached the point
of re-formatting and mounting the
hardrives; there it stopped installing with error messages about not being
able to format and mount the
harddrives
Then I tried to install again with the same hw configuration, this time
using yast1. Booting from CD-ROM,
going "manual" and then "install", "yast1", it stopped shortly after with
the error message "An error has
ocurred".
On tty10, the following error message appeared:
<4>scsi0: MEDIUM ERROR on channel 0, id 6, lun 0, CDB: Request Sense 00 00
00 10 00
<4>Info fld=0x1ac, Current sr0b:00: sense key Medium Error
<4>Additional sense indicates Circ unrecovered error
<4>CD-ROM I/O error: dev 0b:00, sector 1712
<4>scsi0: MEDIUM ERROR on channel 0, id 6, lun 0, CDB: Request Sense 00 00
00 10 00
<4>Info fld=0x134, Current sr0b:00: sense key Medium Error
<4>Additional sense indicates Circ unrecovered error
<4>CD-ROM I/O error: dev 0b:00, sector 1232
<4>scsi0: MEDIUM ERROR on channel 0, id 6, lun 0, CDB: Request Sense 00 00
00 10 00
<4>Info fld=0x266, Current sr0b:00: sense key Medium Error
<4>Additional sense indicates L-ec uncorrectable error
<4>CD-ROM I/O error: dev 0b:00, sector 2456
The next try was installing via NFS: The same hw configuration, I
successfully booted from the local CD-ROM,
switched to "manual", started "installation" using "NFS". I mounted the
CD-ROM in an i386-SuSE-7.1 PC,
and then was able to install the "standard" selection from yast1 plus
"kde"-packages successfully.
But rebooting afterwards using kernel 2.2.19, during the boot process I
received error messages from the
SCSI host adapter:
<4>scsi : aborting command due to timeout : pid 21913, scsi0, channel 0,
id 1, lun 0 Read (10) 00 00 14 28 f6 00 00 80 00
<4>esp0: Aborting command
<4>esp0: dumping state
<4>esp0: dma -- cond_reg<a6400310> addr<f06f31fc>
<4>esp0: SW [sreg<11> sstep<04> ireg<18>]
<4>esp0: HW reread [sreg<01> sstep<cc> ireg<00>]
<4>esp0: current command [tgt<01> lun<00> pphase<CLUELESS> cphase<DATAIN>]
<4>esp0: disconnected
<4>SCSI host 0 abort (pid 21913) timed out - resetting
<4>SCSI bus is being reset for host 0 channel 0.
<4>esp0: Resetting scsi bus
<4>esp0: Gross error sreg=40
<4>esp0: SCSI bus reset interrupt
I was able to boot but barely; the process was very slow, due to the
continous resets of the SCSI host adapter.
The whole behaviour is about the same as I have seen with SuSE 7.1 SPARC
using kernel 2.2.18.
The next try was using the kernel 2.4.9, against better advice....it did
not even boot.
The last I did was coying the kernel 2.4.2 from SuSE 7.1 SPARC (running on
another SS5, 110MHz, 96MB RAM) to
this machine, then it did boot ok, and ran fine. No errros from the SCSI
host adapter, no problems accessing the locally
connected SCSI CD-ROM.
But from my experience on the other SS5, about every 2 weeks, the station
looks up with this kernel, leaving me with
the STOP-A / "reset" sequence to revive the SS5....;-)
Summary:
- The install process ran fine with the above exception, which is not the
fault of the installer.
- In "yast1", if you try to install a kernel, it does not offer a 2.2; I
had to take the 2.4 for 32-bit. Wrong label.
Intermezzo #1:
=============
Just to see if anything was wrong with the SCSI sub system (I have the
settings straight out of the SS5, which ran
Solaris 2.x for the last years), I installed NetBSD 1.5 for SPARC. Now
NetBSD is fairly small, and it did only use
1 disk, but I could install it from the local CD-ROM, access hd SCSI-ID 3,
configure the system, run X11 (twm) and
run some apps without any errors. So I think the SCSI subsystem is
basically sound.
Installation #2:
=============
The next step was installing SuSE 7.3 on the same SS5, but just using 1
SCSI hd. The other hd stayed in the machine,
but was not going to be used for SuSE. First I tried again with the local
SCSI-CD-ROM, but the same errors appeared
as in "Installation #1".
OK, booting from the local SCSI CD-ROM, switching to "manual" , starting
the "installation" via "NFS" and using "yast2".
This time the installation was successfull, but lengthy; I installed "KDE
complete" plus "Network Server" on 1 SCSI disk
(SCSI ID 1). During the installation the error of the "esp" SCSI host
adapter appeared once, but the installation continued.
Attached is the "boot.msg" from the whole installation process.
Summary:
- The installation went fine (nice pictures of SuSE employees, I
believe...), I ended up with the SuSE-KDE 2.2 desktop.
- During install, the installer complains about not finding a "swap"
partition, despite one was setup and configured for
formating in the installation step before.
- When the installer is doing the re-formatting and mounting, it does not
display the swap partition, ie:
"Partition /dev/sda2 als formatieren"; quite iritating that the term
"swap" does not appear in the dialog window
- Despite my input about country = Switzerland, the SuSE-KDE desktop is
set to "country = "Germany"
- The SuSE KDE desktop is not very well suited for machines with 256 color
framebuffers, starting other apps let them
with almost no colors (terminal programs a la "dialog" are fun in white on
white...). I was not able to adjust the desktop
settings to get a good result.
-> to circumvent this problem, I logged in console mode, deleted the .kde2
directory and then started the X11 / KDE;
since KDE does not find a setup (?) it creates a standard one, which is
near perfect for machines with 256 color frame-
buffer; could you use that in your install ?
Summary:
=========
I have SuSE 7.3 beta1 now running, suing it as a remote console using VNC
to check on NT-servers; I wonder if there
are any patches to kernel 2.2.19 to correct the errorneous behaviour of
the "esp"-SCSI-driver. Otherwise it looks quite
good, keep up the good work, please ! I'm sorry that my descriptions of
all the installer labels is not 100% correct, I did
not put them down in absolute wording.
Kind regards,
Thomas Schmid
Email: Thomas.Schmid@ascom.ch
(See attached file: boot.msg)
-- To unsubscribe, e-mail: suse-sparc-unsubscribe@suse.com For additional commands, e-mail: suse-sparc-help@suse.com
This archive was generated by hypermail 2.1.0 : Wed Sep 19 2001 - 01:20:00 PDT