Hi everybody

I've googled a lot about these problems, but have found not solution
(yet) I hope you can help me.

I Installed Solaris 10 U2 on and old computer (PIII @ 866, 512MB RAM,
40GB HDD). Then, installed gnome and KDE, both from blastwave. Problems
I'm facing are:

1. GNOME doesn't work correctly (clock applet crashes on startup, and
the desktop takes about 5 minutes to load)
2. KDE does not show on dtlogin (and trying to launch it by hand
complains about not finding xsetroot, and other stuff and fails)
3. How to use GDM instead of dtlogin? I disabled dtlogin, but no idea
how to enable GDM (I see a gdm in svcs -a, but I think it's the SUN one,
not the blastwave one)

Any hint/link/help, is greatly appreciated. Thanks.

German Andres

--- In solarisx86@yahoogroups.com, "Germán Andrés Pulido F." <g-
pulido@...> wrote:
>
> Hi everybody
>
> I've googled a lot about these problems, but have found not solution
> (yet) I hope you can help me.
>
> I Installed Solaris 10 U2 on and old computer (PIII @ 866, 512MB
RAM,
> 40GB HDD). Then, installed gnome and KDE, both from blastwave.
Problems
> I'm facing are:
>
> 1. GNOME doesn't work correctly (clock applet crashes on startup,
and
> the desktop takes about 5 minutes to load)
> 2. KDE does not show on dtlogin (and trying to launch it by hand
> complains about not finding xsetroot, and other stuff and fails)
> 3. How to use GDM instead of dtlogin? I disabled dtlogin, but no
idea
> how to enable GDM (I see a gdm in svcs -a, but I think it's the SUN
one,
> not the blastwave one)
>
> Any hint/link/help, is greatly appreciated. Thanks.
>
> German Andres
>
-----------------------

Well, as for GNOME you are in the state of flux with it being updated.
The updated packages are in www.blastwave.org/testing. Mainly, update
evolution-ds,gnome-panel,and control panel.

As for KDE, it works very well. I'll look into some of those issues
you mentioned.

The GDM commands are /opt/csw/sbin/gdm*. I think I include a link to
the gdm manual and some docs are included to walk you through this.
Brian Cameron at Sun is one of the experts (and a maintainer) of GDM
(see http://blogs.sun.com/yippi/entry/gdm_configuration_overhaul).

Let me know what else you see....
~ Ken Mays @ EarthLink

I'm running S10X86 on an Intel D865PERL motherboard with 4GB of RAM
installed.  This board supports up to 4GB but also maps some of it's own
memory space in to the top 256MB of the 4GB space.

I see this warning when I boot and find the stuff below in
/var/adm/messages.  I'm assuming that this has to do with the
unaccessible memory space.

How do I tell S10 to only use the first 3840MB of RAM at boot time?

pci_autoconfig: [ID 771164 kern.info] NOTICE: reprogram pci device
[1/0/1] (pci1002,2003)
rootnex: [ID 349649 kern.info] isa0 at root
pci_autoconfig: [ID 516285 kern.warning] WARNING: failed to program mem
space [1/0/1] BAR@0x10 length 0x4000000
pci_autoconfig: [ID 771164 kern.info] NOTICE: reprogram pci device
[1/0/1] (pci1002,2003)
pcplusmp: [ID 658230 kern.info] NOTICE: apic: local nmi: 1 0 0 1
pci_autoconfig: [ID 516285 kern.warning] WARNING: failed to program mem
space [1/0/1] BAR@0x10 length 0x4000000
pcplusmp: [ID 658230 kern.info] NOTICE: apic: local nmi: 2 0 0 1
pcplusmp: [ID 658230 kern.info] NOTICE: apic: local nmi: 1 0 0 1

Thanks,
Frank

Thanks a lot for your help Ken. I have currently simlinked
/opt/csw/etc/gdm/gdm.conf to /etc/X11/gdm/gdm.conf (the first one is a
sinlink to the second one) and now it seems to take customizations
fine. (done with the option of modify GDM configuration, given in the
"Actions" GDM Menu). Now, I have no clue where may I add KDE and GNOME
CSW to GDM (I also have no idea if it is using all of its files, or is
using part of its files and part of the SUN GDM). Anyway, this seems
to be getting better everytime :)

Thanks for the help.

--- In solarisx86@yahoogroups.com, "maybird1776" <maybird1776@...> wrote:
>
> --- In solarisx86@yahoogroups.com, "Germán Andrés Pulido F." <g-
> pulido@> wrote:
> >
> > Hi everybody
> >
> > I've googled a lot about these problems, but have found not solution
> > (yet) I hope you can help me.
> >
> > I Installed Solaris 10 U2 on and old computer (PIII @ 866, 512MB
> RAM,
> > 40GB HDD). Then, installed gnome and KDE, both from blastwave.
> Problems
> > I'm facing are:
> >
> > 1. GNOME doesn't work correctly (clock applet crashes on startup,
> and
> > the desktop takes about 5 minutes to load)
> > 2. KDE does not show on dtlogin (and trying to launch it by hand
> > complains about not finding xsetroot, and other stuff and fails)
> > 3. How to use GDM instead of dtlogin? I disabled dtlogin, but no
> idea
> > how to enable GDM (I see a gdm in svcs -a, but I think it's the SUN
> one,
> > not the blastwave one)
> >
> > Any hint/link/help, is greatly appreciated. Thanks.
> >
> > German Andres
> >
> -----------------------
>
> Well, as for GNOME you are in the state of flux with it being updated.
> The updated packages are in www.blastwave.org/testing. Mainly, update
> evolution-ds,gnome-panel,and control panel.
>
> As for KDE, it works very well. I'll look into some of those issues
> you mentioned.
>
> The GDM commands are /opt/csw/sbin/gdm*. I think I include a link to
> the gdm manual and some docs are included to walk you through this.
> Brian Cameron at Sun is one of the experts (and a maintainer) of GDM
> (see http://blogs.sun.com/yippi/entry/gdm_configuration_overhaul).
>
> Let me know what else you see....
> ~ Ken Mays @ EarthLink
>

I bumped into this recently with the last Solaris 10 update or two.
When installing and setting things up Solaris defaults to Xorg server.

If I leave it just as it is with the initial install it will auto
configure when I am docked and when I am undocked.  I have a Dell
Latitude D810.  The native screen resolution is 1920 x 1200 when
undocked.  My docked configuration is 1600 x 1200 with the LCD monitor
I use.

When I tried changing to Xsun I ran into problems having to use
kdmconfig every time I would dock or undock.  Now by just leaving it
with Xorg it takes care of this for me.

NOTE - If you do a reconfiguration like sys-unconfig simply choose
Xorg again but do not create a configuration file like it states.
This resets it back to Xorg using the system default configured during
the initial installation.

v/r
Jim Johansen

James:

You are discovering what should be made clear in the documentation:
Xorg is the replacement for Xsun, it is the default, and Xsun should only
be chosen for a rapidly diminishing set of conditions, like old
hardware, or the need
for an Xsun-specific extension..

Is there some problem or surprise implied in this message?

Stuart

X group






jamesdonjohansen wrote:
>
> I bumped into this recently with the last Solaris 10 update or two.
> When installing and setting things up Solaris defaults to Xorg server.
>
> If I leave it just as it is with the initial install it will auto
> configure when I am docked and when I am undocked. I have a Dell
> Latitude D810. The native screen resolution is 1920 x 1200 when
> undocked. My docked configuration is 1600 x 1200 with the LCD monitor
> I use.
>
> When I tried changing to Xsun I ran into problems having to use
> kdmconfig every time I would dock or undock. Now by just leaving it
> with Xorg it takes care of this for me.
>
> NOTE - If you do a reconfiguration like sys-unconfig simply choose
> Xorg again but do not create a configuration file like it states.
> This resets it back to Xorg using the system default configured during
> the initial installation.
>
> v/r
> Jim Johansen
>

--- In solarisx86@yahoogroups.com, Stuart Kreitman
<Stuart.Kreitman@...> wrote:
>
> James:
>
> You are discovering what should be made clear in the documentation:
> Xorg is the replacement for Xsun, it is the default, and Xsun should
only
> be chosen for a rapidly diminishing set of conditions, like old
> hardware, or the need
> for an Xsun-specific extension..
>
> Is there some problem or surprise implied in this message?
>
> Stuart
>
> X group

  No surprise at all.  I have an idea.  Let's get rid of Xsun
on the distribution completely and make it simple.  Hmm, Xsun
might be required for SunRay but that is a long lost separate
consolidation altogether.  If true maybe the Sunray team can
just take over the Xsun support and handle it with their own
resources.  Is it really necessary to keep Xsun around for
old obscure video drivers?  If so what are these drivers and
why are they important?

---Bob

palowoda wrote:

> I use to think that one of Sun's biggest headaches was to
>get content drivers into Solaris x86.  After reading this
>article:
>http://www.cs.auckland.ac.nz/~pgut001/pubs/vista_cost.txt
>if half the claims are true it's going to make it more
>difficult. I know the content revocation is in external devices
>but I wasn't aware hardware OEM where pressed to build
>product designs in the motherboard.  They are not required.
>Does this mean Sun's hardware might become an advantage
>in the marketplace not following Vista based hardware design
>specifications?  How will Sun respond to selling Opteron based
>systems into the Vista marketplace?  How will Sun's project
>Dream work with Vista technology?  Is DRM taken seriously
>by Sun or just one of those buddy pet projects?  More important
>does Sun and Sun's customers even really care about DRM to
>give it the time of day?
>
>
>
Those of us who want to download and watch legal copies of TV programs
on Solaris do.  The UK's channel 4 site only supports windoze due to DRM
issues.

Ian

I have never experienced this error, not on linux, not on FreeBSD. I
also have NO clue at all where this error could have come from. But
anyway, my computer wouldn't boot today. Had to run fsck manually.
So I did.. The / (root) partition had irreparable cylinder group; header
problem; program terminated. A whole list of msgs.. CG131 Bad CG magic
number (0x0 should be 0x90255) and lots more. The system was dead. I
knew of no way of repairing my root partition. Sh*t.

A spare disk and some ufsdumping / restoring later I have my root
partition back. After a newfs all is as new and works OK.

Remains my question where these CG errors could have come from. I always
thought the UFS filesystem was rather safe.

Does somebody know of any clues on this issue?

--
http://nagual.nl/ -- PGP/GnuPG key: F86289CE
++ Running FreeBSD 6.1 ++ Solaris 10 6/06 ++

Hi,
My Solaris 10 x86 SunRay server is very slow at starting Gnome apps
after I added some CSW packages.
Using truss on the gcalctool it seems to be a font problem.
Full truss output:-
http://brake.homeunix.org/~rich/gcalctool.log.txt

  0.8728 open("/home/rich/.fonts.cache-1", O_RDONLY)     = 19
  0.8728 fstat64(19, 0x08040DA0)                         = 0
  0.8728 fstat64(19, 0x08040CE0)                         = 0
  0.8729 ioctl(19, TCGETA, 0x08040D74)                   Err#25 ENOTTY
  0.8778 read(19, " " / u s r / o p e n w i".., 8192)    = 8192
  0.8783 read(19, " > ^ 1 ! ! ! K ? &     z".., 8192)    = 8192

and is still playing with fonts at:-

15.5196 xstat(2, "/usr/sfw/lib/libfontconfig.so.1", 0x08044CA0) Err#2
ENOENT
15.5197 xstat(2, "/usr/X11/lib/libfontconfig.so.1", 0x08044CA0) Err#2
ENOENT
15.5197 xstat(2, "/usr/openwin/lib/libfontconfig.so.1", 0x08044CA0)
Err#2 ENOENT
15.5198 xstat(2, "/usr/openwin/lib/libfontconfig.so.1", 0x08044CA0)
Err#2 ENOENT
15.5198 xstat(2, "/usr/openwin/sfw/lib/libfontconfig.so.1",
0x08044CA0) Err#2 ENOENT
15.5198 xstat(2, "/lib/libfontconfig.so.1", 0x08044CA0) Err#2 ENOENT
15.6063 open("/usr/openwin/lib/X11/fonts/TrueType/VeraIt.ttf",
O_RDONLY) = 19


Nearly 15 seconds later

The version of Solaris is :-
rich: cat /etc/release
                         Solaris 10 6/06 s10x_u2wos_09a X86
            Copyright 2006 Sun Microsystems, Inc.  All Rights Reserved.
                         Use is subject to license terms.
                              Assembled 09 June 2006
rich: prtdiag -v
System Configuration: Sun Microsystems Sun Fire V20z
BIOS Configuration: Phoenix Technologies Ltd. V1.35.2.2   04/25/2006
BMC Configuration: IPMI 1.5 (KCS: Keyboard Controller Style)

==== Processor Sockets ====================================

Version                          Location Tag
-------------------------------- --------------------------
AMD                              CPU0-Socket 940
AMD                              CPU1-Socket 940
rich: psrinfo -v
Status of virtual processor 0 as of: 01/02/2007 14:01:18
   on-line since 12/16/2006 13:25:28.
   The i386 processor operates at 1791 MHz,
         and has an i387 compatible floating point processor.
Status of virtual processor 1 as of: 01/02/2007 14:01:18
   on-line since 12/16/2006 13:25:34.
   The i386 processor operates at 1791 MHz,
         and has an i387 compatible floating point processor.

The CSW packages added ore updated :-

    PKGINST:  CSWbdb4
   INSTDATE:  Dec 19 2006 17:41
    PKGINST:  CSWbdb43
   INSTDATE:  Dec 19 2006 17:39
    PKGINST:  CSWbonobo2
   INSTDATE:  Dec 19 2006 17:39
    PKGINST:  CSWbzip2
   INSTDATE:  Dec 19 2006 17:39
    PKGINST:  CSWfam
   INSTDATE:  Dec 19 2006 17:41
    PKGINST:  CSWfconfig
   INSTDATE:  Dec 19 2006 17:39
    PKGINST:  CSWftype2
   INSTDATE:  Dec 19 2006 17:39
    PKGINST:  CSWgail
   INSTDATE:  Dec 19 2006 17:41
    PKGINST:  CSWgcc3corert
   INSTDATE:  Dec 19 2006 17:41
    PKGINST:  CSWgcc3g++rt
   INSTDATE:  Dec 19 2006 17:41
    PKGINST:  CSWgconf2
   INSTDATE:  Dec 19 2006 17:41
    PKGINST:  CSWgcrypt
   INSTDATE:  Dec 19 2006 17:40
    PKGINST:  CSWglib2
   INSTDATE:  Dec 19 2006 17:39
    PKGINST:  CSWgnomekeyring
   INSTDATE:  Dec 19 2006 17:41
    PKGINST:  CSWgnomevfs2
   INSTDATE:  Dec 19 2006 17:41
    PKGINST:  CSWgnucash
   INSTDATE:  Dec 19 2006 17:42
    PKGINST:  CSWgnucashcommon
   INSTDATE:  Dec 19 2006 17:38
    PKGINST:  CSWgnutls
   INSTDATE:  Dec 19 2006 17:40
    PKGINST:  CSWgpgerr
   INSTDATE:  Dec 19 2006 17:40
    PKGINST:  CSWgsed
   INSTDATE:  Dec 19 2006 17:39
    PKGINST:  CSWgtk2
   INSTDATE:  Dec 19 2006 17:40
    PKGINST:  CSWgtkhtml31
   INSTDATE:  Dec 19 2006 17:42
    PKGINST:  CSWguile
   INSTDATE:  Dec 19 2006 17:42
    PKGINST:  CSWguilelib12
   INSTDATE:  Dec 19 2006 17:42
    PKGINST:  CSWgwrap
   INSTDATE:  Dec 19 2006 17:42
    PKGINST:  CSWkrb5lib
   INSTDATE:  Dec 19 2006 17:41
    PKGINST:  CSWlibart
   INSTDATE:  Dec 19 2006 17:41
    PKGINST:  CSWlibatk
   INSTDATE:  Dec 19 2006 17:39
    PKGINST:  CSWlibbonoboui
   INSTDATE:  Dec 19 2006 17:42
    PKGINST:  CSWlibcairo
   INSTDATE:  Dec 27 2006 21:11
    PKGINST:  CSWlibcups
   INSTDATE:  Dec 19 2006 17:40
    PKGINST:  CSWlibglade2
   INSTDATE:  Dec 19 2006 17:41
    PKGINST:  CSWlibgnome
   INSTDATE:  Dec 19 2006 17:42
    PKGINST:  CSWlibgnomecanvas
   INSTDATE:  Dec 19 2006 17:41
    PKGINST:  CSWlibgnomecups
   INSTDATE:  Dec 19 2006 17:42
    PKGINST:  CSWlibgnomeprint
   INSTDATE:  Dec 19 2006 17:42
    PKGINST:  CSWlibgnomeprintui
   INSTDATE:  Dec 19 2006 17:42
    PKGINST:  CSWlibgnomeui
   INSTDATE:  Dec 19 2006 17:42
    PKGINST:  CSWlibgsf
   INSTDATE:  Dec 19 2006 17:42
    PKGINST:  CSWlibidl
   INSTDATE:  Dec 19 2006 17:39
    PKGINST:  CSWlibnet
   INSTDATE:  Dec 19 2006 17:39
    PKGINST:  CSWlibofx
   INSTDATE:  Dec 19 2006 17:42
    PKGINST:  CSWlibogg
    PKGINST:  CSWlibpopt
   INSTDATE:  Dec 19 2006 17:39
    PKGINST:  CSWlibtool
   INSTDATE:  Dec 19 2006 17:42
    PKGINST:  CSWlibxft2
   INSTDATE:  Dec 19 2006 17:40
    PKGINST:  CSWlibxml
    PKGINST:  CSWlibxml2
   INSTDATE:  Dec 19 2006 17:39
    PKGINST:  CSWlibxrender
   INSTDATE:  Dec 19 2006 17:40
    PKGINST:  CSWoldaprt
   INSTDATE:  Dec 19 2006 17:41
    PKGINST:  CSWopencdk
   INSTDATE:  Dec 19 2006 17:40
    PKGINST:  CSWopensp
   INSTDATE:  Dec 19 2006 17:42
    PKGINST:  CSWorbit2
   INSTDATE:  Dec 19 2006 17:39
    PKGINST:  CSWpango
   INSTDATE:  Dec 19 2006 17:40
    PKGINST:  CSWperl
   INSTDATE:  Dec 19 2006 17:39
    PKGINST:  CSWpmdatemanip
   INSTDATE:  Dec 19 2006 17:39
    PKGINST:  CSWpng
   INSTDATE:  Dec 19 2006 17:34
    PKGINST:  CSWreadline
   INSTDATE:  Dec 19 2006 17:40
    PKGINST:  CSWrenderdev
   INSTDATE:  Dec 19 2006 17:39
    PKGINST:  CSWsambalib
   INSTDATE:  Dec 19 2006 17:41
    PKGINST:  CSWsasl
   INSTDATE:  Dec 19 2006 17:41
    PKGINST:  CSWslib
   INSTDATE:  Dec 19 2006 17:42
    PKGINST:  CSWtexinfo
   INSTDATE:  Dec 19 2006 17:39

Why is Gnome now using the file /home/rich/.fonts.cache-1 each time an
application is started and why are font files being searched etc.
On my slow PC runnig Solaris nv b54 gcalctool starts in about 2
seconds and does not do all the font stuff.
Cheers
Richard.

On Mon, 1 Jan 2007, dick hoogendijk wrote:

> I have never experienced this error, not on linux, not on FreeBSD. I
> also have NO clue at all where this error could have come from. But
> anyway, my computer wouldn't boot today. Had to run fsck manually.
> So I did.. The / (root) partition had irreparable cylinder group; header
> problem; program terminated. A whole list of msgs.. CG131 Bad CG magic
> number (0x0 should be 0x90255) and lots more. The system was dead. I
> knew of no way of repairing my root partition. Sh*t.
>
> A spare disk and some ufsdumping / restoring later I have my root
> partition back. After a newfs all is as new and works OK.
>
> Remains my question where these CG errors could have come from. I always
> thought the UFS filesystem was rather safe.
>
> Does somebody know of any clues on this issue?

It's very likely that you simply have a hard disk that is about to die or
some other hardware related issue(s).  The fact that you were able to get
the disk back online does not indicate that the underlying issue has been
resolved.  A couple of items worth checking:

a) that you have some air movement over the disk CHA (Cylinder Head
Assembly)
b) burn a copy of the Ultimate Boot CD Rom and run memtestx86 for a couple
of hours
c) verify that you don't have a sticky fan in the box - including inside
the power supply
d) verify that you don't have an accumulation of dust between the CPU fan
and the CPU heatsink (you'll have to remove the CPU fan to inspect).
e) replace any fans that are 3 years old (or older)
f) if the system is not on a UPS, it may have been affected by a power
related event

BTW: you can see the source code for the above error.  Use the OpenGrok
source code browser at opensolaris.org.  It's probably:

http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/cmd/fs.d/ufs/fsck/\
utilities.c#1926

Regards,

Al Hopper  Logical Approach Inc, Plano, TX.  al@...
            Voice: 972.379.2133 Fax: 972.379.2134  Timezone: US CDT
OpenSolaris.Org Community Advisory Board (CAB) Member - Apr 2005
              OpenSolaris Governing Board (OGB) Member - Feb 2006

Frank Tarczynski wrote:
> I'm running S10X86 on an Intel D865PERL motherboard with 4GB of RAM
> installed.  This board supports up to 4GB but also maps some of it's own
> memory space in to the top 256MB of the 4GB space.
>
> I see this warning when I boot and find the stuff below in
> /var/adm/messages.  I'm assuming that this has to do with the
> unaccessible memory space.
>
> How do I tell S10 to only use the first 3840MB of RAM at boot time?
>
> pci_autoconfig: [ID 771164 kern.info] NOTICE: reprogram pci device
> [1/0/1] (pci1002,2003)
> rootnex: [ID 349649 kern.info] isa0 at root
> pci_autoconfig: [ID 516285 kern.warning] WARNING: failed to program mem
> space [1/0/1] BAR@0x10 length 0x4000000
> pci_autoconfig: [ID 771164 kern.info] NOTICE: reprogram pci device
> [1/0/1] (pci1002,2003)
> pcplusmp: [ID 658230 kern.info] NOTICE: apic: local nmi: 1 0 0 1
> pci_autoconfig: [ID 516285 kern.warning] WARNING: failed to program mem
> space [1/0/1] BAR@0x10 length 0x4000000
> pcplusmp: [ID 658230 kern.info] NOTICE: apic: local nmi: 2 0 0 1
> pcplusmp: [ID 658230 kern.info] NOTICE: apic: local nmi: 1 0 0 1
>
> Thanks,
> Frank

I finally had a chance to look this up.

This appears to be mentioned in several bugs.  This one is still open:

6412370 Warning errors at start up for USB ECHI host controller.

But it is possibly a duplicate of:

6354996 pci_boot.c incorrectly accounts for system resources in PCI
configuration

The fix for 6354996 is in build 52 of Solaris Express.  So you might want to
verify if booting with Solaris Express resolves this problem.  If it does
you can ask for the fix to be backported to Solaris 10 via an escalation,
assuming you have a Support contract.

Mike

palowoda wrote:
>  No surprise at all.  I have an idea.  Let's get rid of Xsun
> on the distribution completely and make it simple.

Working on it - have to give one year prior notice first, which
should happen in the next S10 update release:
http://www.opensolaris.org/os/community/arc/caselog/2006/641/

> Hmm, Xsun
> might be required for SunRay but that is a long lost separate
> consolidation altogether.  If true maybe the Sunray team can
> just take over the Xsun support and handle it with their own
> resources.  Is it really necessary to keep Xsun around for
> old obscure video drivers?  If so what are these drivers and
> why are they important?

SPARC graphics & Sun Ray are the primary devices stuck on Xsun
now that agpgart has finally shipped in Solaris 10, so that
Intel 810/815 chipsets can use Xorg.

--
	 -Alan Coopersmith-           alan.coopersmith@...
	  Sun Microsystems, Inc. - X Window System Engineering

This didn't take long:

http://www.infoworld.com/article/06/12/29/HNdrmhacked_1.html?source=NLC-SEC2007-\
01-01

--- In solarisx86@yahoogroups.com, Alan Coopersmith
<alan.coopersmith@...> wrote:
>
> palowoda wrote:
> >  No surprise at all.  I have an idea.  Let's get rid of Xsun
> > on the distribution completely and make it simple.
>
> Working on it - have to give one year prior notice first, which
> should happen in the next S10 update release:
> http://www.opensolaris.org/os/community/arc/caselog/2006/641/

  Uggh the one year notice.  I know it's a policy but I have to
wonder (other than the Sunray team) who the notice is intended
to reach or affect?  Maybe there is nobody with third party
apps that are dependent on Xsun for x86/x64 to begin with.
It does by time though for support to tell users not to
use Xsun.  Good thing?

  By the way had to stop and think about fast-tracking an EOF. :-)

>
> > Hmm, Xsun
> > might be required for SunRay but that is a long lost separate
> > consolidation altogether.  If true maybe the Sunray team can
> > just take over the Xsun support and handle it with their own
> > resources.  Is it really necessary to keep Xsun around for
> > old obscure video drivers?  If so what are these drivers and
> > why are they important?
>
> SPARC graphics & Sun Ray are the primary devices stuck on Xsun
> now that agpgart has finally shipped in Solaris 10, so that
> Intel 810/815 chipsets can use Xorg.
>

  Yes the limited resources put into the UltraSparc desktop is
becoming a Solaris liability in general.  Been that way for
sometime now.  Unfortunately that SPARC graphics group doesn't
have a good business plan to turn it around.  Or at least one
that is evident.

  We now bring you back to our regular programming channel.

---Bob

Mike Riley wrote:
>
>
> Ian Collins wrote:
>  > dick hoogendijk wrote:
>  >
>  >> On Thu, 2006-12-21 at 22:47, Ian Collins wrote:
>  >>
>  >>
>  >>
>  >>> It is and it worked for me with a branded zone.
>  >>>
>  >>> This build is also worth having for the desktop removable media
>  >>> integration, just plug a drive in and up pops a file browser window.
>  >>> Well done to all involved with this.
>  >>>
>  >>>
>  >> I wonder, if I go for Nevada build54, will I be able to upgrade to
>  >> solaris-11 once it's been released?
>  >>
>  >>
>  >>
>  > Considering there isn't a release date for Solaris11, I wouldn't worry
>  > too much about that.
>  >
>  > Ian
>
> But that is an interesting question. In the past I believe we have always
> supported and tested for upgrading from a Beta release. I wonder, will they
> conduct tests for every version of Solaris Express we released?
>
> I doubt it, since that would be a huge test matrix. But it is an
> interesting thought.
>

Interesting, but completely infeasible due to the test matrix; no way
you'd get a cost-benefit analysis that would support it.  We already
have a number of cases where upgrades (especially the Live variety) are
known to be buggy across various transition points in the Nevada build
series.

If you're going to hop on the Solaris Express train, you need to plan on
upgrading every couple of months to stay current enough for the latest
build to upgrade correctly, since that's about as far back as most
projects test.

Dave

---- dick hoogendijk <dick@...> wrote:
> I have never experienced this error, not on linux, not on FreeBSD. I
> also have NO clue at all where this error could have come from. But
> anyway, my computer wouldn't boot today. Had to run fsck manually.
> So I did.. The / (root) partition had irreparable cylinder group; header
> problem; program terminated. A whole list of msgs.. CG131 Bad CG magic
> number (0x0 should be 0x90255) and lots more. The system was dead. I
> knew of no way of repairing my root partition. Sh*t.

So I have no idea if you're running 9, 10, or nevada.  If on 9 or 10, it's
possible you're back reved on the fsck patch.  If you're up to date on
patches, then possibly consider filing a bug report.

> A spare disk and some ufsdumping / restoring later I have my root
> partition back. After a newfs all is as new and works OK.

  Been there done that.

> Remains my question where these CG errors could have come from. I always
> thought the UFS filesystem was rather safe.

Well, UFS for the most part should be safe.  However, I had a 250GB disk that
ran for about 18 months and then started having problems.  I'd see problems
like yours, but managed to get there in enough time to back all the important
stuff up.  Eventually, I started having problems like yours.  The BIOS started
reporting a "SMART" disk bad or failing, replace soon message, and I couldn't
get a valid fsck to go on a 210GB slice.

I did a couple of format/analyze on that disk connected to another system,
along with installing linux and doing a low leverl on it and then put it on a
windows box and did a slow NTFS format on it.  One more format/analyze
and all the HW errors appeared to go away, and the SMART errors also went
away.  I'm aware that this disk is probably going to continue to degredate, but
I wanted to see what the effect would be on a failing disk.  As expected, the
HW facilities were able to cope with the remapping of bad sectors, and the
disk appears, for intensive purposes, to not be failing anymore.

Ben

The good news is that the Nvidia driver software keeps getting better and
better.  With ref to:
http://tech.groups.yahoo.com/group/solarisx86/message/38373 that Ultra 20
system was replaced with a homebrew system after the U20 motherboard died
[1].  So the dual LCDs, which had been configured to run with
seperate XServers (one for each screen), was reconfigured to run with
TwinView and Xinerama on the new system running Update 3 and (the latest
Nvidia driver software) NVIDIA-Solaris-x86-1.0-9746.run, by executing:

nvidia-xconfig -a --separate-x-screens --xinerama
--twinview-orientation="LeftOf"

This produced a working (/etc/X11) xorg.conf file!  The only changes that
were made was to add 'Option "TargetRefresh" "60.0"' for the 30" monitor
and to hard code the required native screen resolutions.  The resulting
config is available (for the curious) at:

http://solx86.logical-approach.com/Dell30Iiyama21.TwinView.xorg.conf.20070101

Kudos to John Martin and all who continue to persue perfection with the
Nvidia related software.

[1] Many of the components were transferred to the new box.  I'll write up
this "story" later....

PS: A useful enhancement to nvidia-xconfig, would be to include the
command line args, the userid, the hostname and a timestamp of when the
config was generated, as (expanded) comments to the resulting xorg.conf
file.  Similar to how the comments are generated in a sendmail.cf file
after running the sendmail m4 command.  Also - it would be useful to
maintain more than one (say 3) backup of the xorg.conf file ... just in
case.

Regards,

Al Hopper  Logical Approach Inc, Plano, TX.  al@...
            Voice: 972.379.2133 Fax: 972.379.2134  Timezone: US CDT
OpenSolaris.Org Community Advisory Board (CAB) Member - Apr 2005
              OpenSolaris Governing Board (OGB) Member - Feb 2006

Mike Riley a écrit :
> This didn't take long:
>
>
http://www.infoworld.com/article/06/12/29/HNdrmhacked_1.html?source=NLC-SEC2007-\
01-01

The article I've read in the IHT was far less optimistic. They said that
one disc had been copied, and that didn't meant all of them could.
And that's why Blu Ray includes technology to update the copy protection
scheme by putting automatic updates on discs.
Also, AFAIK, both technologies can revoke keys from compromised players
as soon as they become known. Kill-switches. Unlike CSS, this is not static.

Laurent

I installed 10 11/06 on my mail & web server and now I don't get any response
from apache2 or sendmail from any external site. Internal sites connect just
fine. I also run "netservices open" but still no response to external
query's.All other services including ssh & dns  work fine and the firewall is
properly configured to pass ports 80 & 25 to the external facing nic.

Any Idea what's blocking these services they worked fine with 10 6/06?

__
Thanks in advance,
Robert

In message <en2l4c+7228@eGroups.com>, "palowoda" writes:
>Does this mean Sun's hardware might become an advantage=20
>in the marketplace not following Vista based hardware design
>specifications?  How will Sun respond to selling Opteron based
>systems into the Vista marketplace?  How will Sun's project=20
>Dream work with Vista technology?  Is DRM taken seriously=20
>by Sun or just one of those buddy pet projects?  More important
>does Sun and Sun's customers even really care about DRM to
>give it the time of day?=20=20

All these questions and more are going to be answered during
Jonathan Schwartz's keynote at CES next week.
<URL:http://blogs.sun.com/jonathan/entry/ces>

Whoops, never mind, Anil Gadre and company don't have a booth
at CES.
<URL:https://myces2007.bdmetrics.com/Portal/browseresults.aspx?blet=S,T>

Its all going to be explained in the JDS/Solaris track at JavaOne
in May.
<URL:http://www.moscone.com/site/do/event/view?nav.type=0&nav.filter=0705&nav.ba\
se=0701&id=179>

John
groenveld@...

Laurent Blume wrote:
> Mike Riley a écrit :
>> This didn't take long:
>>
>>
http://www.infoworld.com/article/06/12/29/HNdrmhacked_1.html?source=NLC-SEC2007-\
01-01
>
> The article I've read in the IHT was far less optimistic. They said that
> one disc had been copied, and that didn't meant all of them could.
> And that's why Blu Ray includes technology to update the copy protection
> scheme by putting automatic updates on discs.
> Also, AFAIK, both technologies can revoke keys from compromised players
> as soon as they become known. Kill-switches. Unlike CSS, this is not static.
>
> Laurent

If they were to disable player in that manner they would run afoul of fair
use laws.  If person A broke the key for a player from Pioneer, but person B
was using the same player with licensed software, why should person B suffer
for what person A did?  There would be far more people out there using
licensed player software than non-licensed versions, most likely.

Mike

rwhartzell@... wrote:
> I installed 10 11/06 on my mail & web server and now I don't get any response
from apache2 or sendmail from any external site. Internal sites connect just
fine. I also run "netservices open" but still no response to external
query's.All other services including ssh & dns  work fine and the firewall is
properly configured to pass ports 80 & 25 to the external facing nic.
>
> Any Idea what's blocking these services they worked fine with 10 6/06?
>
> __
> Thanks in advance,
> Robert

I just upgraded my system last night and noticed there is a question about
if you want external systems to be able to access your system.  I think that
was their way of asking if you wanted the system made secure.

Try this command as root:

# /usr/sbin/netservices open

Mike

John D Groenveld wrote:
> In message <en2l4c+7228@eGroups.com>, "palowoda" writes:
>> Does this mean Sun's hardware might become an advantage=20
>> in the marketplace not following Vista based hardware design
>> specifications?  How will Sun respond to selling Opteron based
>> systems into the Vista marketplace?  How will Sun's project=20
>> Dream work with Vista technology?  Is DRM taken seriously=20
>> by Sun or just one of those buddy pet projects?  More important
>> does Sun and Sun's customers even really care about DRM to
>> give it the time of day?=20=20
>
> All these questions and more are going to be answered during
> Jonathan Schwartz's keynote at CES next week.
> <URL:http://blogs.sun.com/jonathan/entry/ces>
>
> Whoops, never mind, Anil Gadre and company don't have a booth
> at CES.
> <URL:https://myces2007.bdmetrics.com/Portal/browseresults.aspx?blet=S,T>


Yeah, but I will be there...

Mike

--- In solarisx86@yahoogroups.com, Mike Riley <michael.riley@...> wrote:
>
> rwhartzell@... wrote:
> > I installed 10 11/06 on my mail & web server and now I don't get any
response from apache2 or sendmail from any external site. Internal sites
connect just fine. I also run "netservices open" but still no response
to external query's.All other services including ssh & dns  work fine
and the firewall is properly configured to pass ports 80 & 25 to the
external facing nic.
> >
> > Any Idea what's blocking these services they worked fine with 10
6/06?
> >
> > __
> > Thanks in advance,
> > Robert
>
> I just upgraded my system last night and noticed there is a question
about
> if you want external systems to be able to access your system.  I
think that
> was their way of asking if you wanted the system made secure.
>
> Try this command as root:
>
> # /usr/sbin/netservices open
>
> Mike
>

I tried that and also
# svccfg -s svc:/network/smtp:sendmail setprop config/local_only = false
# svcadm refresh svc:/network/smtp:sendmail
still no luck
--
Robert

--- In solarisx86@yahoogroups.com, "rwhartzell" <rwhartzell@...>
> I tried that and also
> # svccfg -s svc:/network/smtp:sendmail setprop config/local_only = false
> # svcadm refresh svc:/network/smtp:sendmail
> still no luck
> --
> Robert
>

  Even if you manually start the http server on your localhost
and connect to the local host you it doesn't work?

---Bob

rwhartzell@... wrote:

>I installed 10 11/06 on my mail & web server and now I don't get any response
from apache2 or sendmail from any external site. Internal sites connect just
fine. I also run "netservices open" but still no response to external
query's.All other services including ssh & dns  work fine and the firewall is
properly configured to pass ports 80 & 25 to the external facing nic.
>
>Any Idea what's blocking these services they worked fine with 10 6/06?
>
>
>
Do you see anything in your Apache access log?

You could also try snooping port 80 while attempting to connect from
outside your firewall.

Ian

--- In solarisx86@yahoogroups.com, Mike Riley <michael.riley@...> wrote:
>
> Laurent Blume wrote:
> > Mike Riley a écrit :
> >> This didn't take long:
> >>
> >>
http://www.infoworld.com/article/06/12/29/HNdrmhacked_1.html?source=NLC-SEC2007-\
01-01
> >
> > The article I've read in the IHT was far less optimistic. They
said that
> > one disc had been copied, and that didn't meant all of them could.
> > And that's why Blu Ray includes technology to update the copy
protection
> > scheme by putting automatic updates on discs.
> > Also, AFAIK, both technologies can revoke keys from compromised
players
> > as soon as they become known. Kill-switches. Unlike CSS, this is
not static.
> >
> > Laurent
>
> If they were to disable player in that manner they would run afoul
of fair
> use laws.  If person A broke the key for a player from Pioneer, but
person B
> was using the same player with licensed software, why should person
B suffer
> for what person A did?  There would be far more people out there using
> licensed player software than non-licensed versions, most likely.
>

  I was thinking the same thing. What happens when a hacker war
starts up and keys are being revoked all the time?  Well
I guess if you have to take it back and get serviced somebody
will make some money on the service call, or lose money.

---Bob

Mike Riley wrote:

>John D Groenveld wrote:
>
>
>>In message <en2l4c+7228@eGroups.com>, "palowoda" writes:
>>
>>
>>>Does this mean Sun's hardware might become an advantage=20
>>>in the marketplace not following Vista based hardware design
>>>specifications?  How will Sun respond to selling Opteron based
>>>systems into the Vista marketplace?  How will Sun's project=20
>>>Dream work with Vista technology?  Is DRM taken seriously=20
>>>by Sun or just one of those buddy pet projects?  More important
>>>does Sun and Sun's customers even really care about DRM to
>>>give it the time of day?=20=20
>>>
>>>
>>All these questions and more are going to be answered during
>>Jonathan Schwartz's keynote at CES next week.
>><URL:http://blogs.sun.com/jonathan/entry/ces>
>>
>>Whoops, never mind, Anil Gadre and company don't have a booth
>>at CES.
>><URL:https://myces2007.bdmetrics.com/Portal/browseresults.aspx?blet=S,T>
>>
>>
>
>
>Yeah, but I will be there...
>
>
>
With a fold away one man booth??

Ian

Happy new year !


Just downloaded Oracle10G for Solaris x86 but fitting it into a CD is
proving to be difficult.

The Oracle instalation files are more than 730MB in size

I made an ISO image file "mkisofs -o /oracle10g.iso -r /opt/database"
the resultant iso file was 749 MB in size which is too large for an
ordinary CD.


mkisofs with out the -r switch will give an 640MB iso file which when
burnt into a CD results in some file names being incomplete . eg
the executable file "runinstaller" becomes "runistal"



I also tried using cdrecord with -overburn option but the iso file is
just proving to be too big.

"cdrecord -v dev=1,0,0 -dao -eject -overburn /oracle10g.iso"


Blocks total: 359849 Blocks current: 359849 Blocks remaining: -24073
./cdrecord: WARNING: Data may not fit on current disk.
./cdrecord: Notice: Overburning active. Trying to write more than the
official disk capacity.
Forcespeed is OFF.
Starting to write CD/DVD at speed 48 in real SAO mode for single
session.
Last chance to quit, starting real write    0 seconds. Operation
starts.
Waiting for reader process to fill input buffer ... input buffer
ready.
BURN-Free is ON.
Turning BURN-Free off
Performing OPC...
Sending CUE sheet...
Writing pregap for track 1 at -150
Starting new track at sector: 0
Track 01:  724 of  749 MB written (fifo  90%) [buf  49%]
10.0x../cdrecord: I/O error. write_g1: scsi sendcmd: no error
CDB:  2A 00 00 05 A8 AC 00 00 1C 00
status: 0x2 (CHECK CONDITION)
Sense Bytes: 70 00 03 00 00 00 00 0A 00 00 00 00 0C 00 00 00 00 00
Sense Key: 0x3 Medium Error, Segment 0
Sense Code: 0x0C Qual 0x00 (write error) Fru 0x0
Sense flags: Blk 0 (not valid)
resid: 57344
cmd finished after 4.764s timeout 200s

write track data: error after 759521280 bytes
./cdrecord: A write error occured.
./cdrecord: Please properly read the error message above.
Writing  time:  442.874s
Average write speed  11.8x.
Min drive buffer fill was 0%
Total of 32 possible drive buffer underruns predicted.
Fixating...
Fixating time:    0.014s
./cdrecord: fifo had 13318 puts and 13246 gets.
./cdrecord: fifo was 2 times empty and 454 times full, min fill was
0%.



Kindly assist