[Qemu-devel] Solaris 8 SPARC, Web Start Launcher Crashing

[Qemu-devel] Solaris 8 SPARC, Web Start Launcher Crashing

[P. Wilhelm]

[-- Attachment #1: Type: text/plain, Size: 2572 bytes --]

In late May of 2011, Brian Vandenberg queried this group about failures 
he was having trying to install Solaris 8 SPARC 02/04 on a Qemu VM. The 
interaction seemed to have ended with Blue Swirl indicating that Brian 
should run with "-d in_ascm,int" and that the debug output would likely 
be useful near the end of the output. I did searches on the mailing list 
from Brian's name but did not find another thread where his original was 
followed up on.


Recently and without knowledge of the above, I began trying to to 
install Solaris 8 SPARC 10/01 into a Qemu VM using the git head (as of 
10/25/2014) and openBIOS r1321. After struggling for a while, I found 
that the installations were failing during install when java was called 
to create the sysidcfg file. The error I am seeing is much like some of 
those posted in May 2011 by Brian V.


I am running with the following command line options:

qemu-system-sparc \

-bios openbios-builtin-sparc32.elf \

-hda test.disk \

-cdrom Solaris8Install.iso \

-prom-env 'auto-boot?=false' -m 256 -nographic


The hard drive is a qcow2 image formatted using Solaris 8 install CD 
(booted Qemu in single user mode from CD) 'format' command to create a 
36G Solaris disk.


The initial format and software load appears to work without problem. 
Then after reboot using the hard disk to boot, "Web Start Solaris 
Command Line installation" comes up and says that it will ask about 
system identification information (Network, Name Service, Date and Time, 
etc.). It tells me to "Press Return to continue". I hit return and am 
greeted with:

/sbin/disk0_install[60]: 125 Abort(coredump)


Looking at the /tmp/disk0_install.log file shows that there was a fault 
during execution of "java sysid -nodisplay".


The message from java in the install log was:

signal fault in critical section

signal number: 11, signal code: 1, fault address: 0xeac30000, pc: 
0xef2cd448, sp: 0xeb232c58

libthread panic: fault in libthread critical section : dumping core

stacktrace:

ef2cd43c

ef2cf0d4

ef2c8d98

0


I created a debug log as suggested by Blue Swirl in his response to 
Brian V. in May 2011. I am able to find in the file where there is a 
Data Access Fault when the pc is at the above noted location. However, I 
am uncertain what to do with this information. Is this something where I 
should take a couple of hundred lines of the logfile before the above pc 
and Data Access Fault and include it here? Is there is an RTFM I missed? 
If so, oops, sorry (maybe just point me where I should go?).


Respectfully,
Paul W.


[-- Attachment #2: Type: text/html, Size: 5219 bytes --]

Re: [Qemu-devel] Solaris 8 SPARC, Web Start Launcher Crashing

[Mark Cave-Ayland]

On 25/10/14 20:16, P. Wilhelm wrote:

> In late May of 2011, Brian Vandenberg queried this group about failures
> he was having trying to install Solaris 8 SPARC 02/04 on a Qemu VM. The
> interaction seemed to have ended with Blue Swirl indicating that Brian
> should run with “-d in_ascm,int” and that the debug output would likely
> be useful near the end of the output. I did searches on the mailing list
> from Brian's name but did not find another thread where his original was
> followed up on.
> 
> 
> Recently and without knowledge of the above, I began trying to to
> install Solaris 8 SPARC 10/01 into a Qemu VM using the git head (as of
> 10/25/2014) and openBIOS r1321. After struggling for a while, I found
> that the installations were failing during install when java was called
> to create the sysidcfg file. The error I am seeing is much like some of
> those posted in May 2011 by Brian V.
> 
> 
> I am running with the following command line options:
> 
> qemu-system-sparc \
> 
> -bios openbios-builtin-sparc32.elf \
> 
> -hda test.disk \
> 
> -cdrom Solaris8Install.iso \
> 
> -prom-env 'auto-boot?=false' -m 256 -nographic
> 
> 
> The hard drive is a qcow2 image formatted using Solaris 8 install CD
> (booted Qemu in single user mode from CD) 'format' command to create a
> 36G Solaris disk.
> 
> 
> The initial format and software load appears to work without problem.
> Then after reboot using the hard disk to boot, “Web Start Solaris
> Command Line installation” comes up and says that it will ask about
> system identification information (Network, Name Service, Date and Time,
> etc.). It tells me to “Press Return to continue”. I hit return and am
> greeted with:
> 
> /sbin/disk0_install[60]: 125 Abort(coredump)
> 
> 
> Looking at the /tmp/disk0_install.log file shows that there was a fault
> during execution of “java sysid -nodisplay”.
> 
> 
> The message from java in the install log was:
> 
> signal fault in critical section
> 
> signal number: 11, signal code: 1, fault address: 0xeac30000, pc:
> 0xef2cd448, sp: 0xeb232c58
> 
> libthread panic: fault in libthread critical section : dumping core
> 
> stacktrace:
> 
> ef2cd43c
> 
> ef2cf0d4
> 
> ef2c8d98
> 
> 0
> 
> 
> I created a debug log as suggested by Blue Swirl in his response to
> Brian V. in May 2011. I am able to find in the file where there is a
> Data Access Fault when the pc is at the above noted location. However, I
> am uncertain what to do with this information. Is this something where I
> should take a couple of hundred lines of the logfile before the above pc
> and Data Access Fault and include it here? Is there is an RTFM I missed?
> If so, oops, sorry (maybe just point me where I should go?).

Hi Paul,

Please send me the debug log off-list and I'll have a look and see if
there is anything obvious.


ATB,

Mark.