Re: [osol-discuss] Bootup script

[Krishna Chaitanya]

This worked! Thanks Jim!!
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] when gtk+ update to 2.12 or above

[ilho.kim]

I'm with Opensolaris 2008.11. Could you please check below information ?

configure error.. gtk+-2.0.pc showing binary version 2.l0.0 ..

.
 gmsgfmt... (cached) /usr/bin/gmsgfmt
checking for xgettext... (cached) /usr/gnu/bin/xgettext
checking for pkg-config... /usr/bin/pkg-config
checking pkg-config is at least version 0.9.0... yes
checking for GTK... no
configure: error: GTK+-2.12 is required to compile murrine
ssep...@ssepirot61:~/다운로드/murrine-0.90.3$ cat /etc/release
   OpenSolaris 2008.11 snv_101b_rc2 X86
   Copyright 2008 Sun Microsystems, Inc.  All Rights Reserved.
Use is subject to license terms.
   Assembled 19 November 2008
ssep...@ssepirot61:~/다운로드/murrine-0.90.3$ cat /usr/lib/pkgconfig/gtk+-2.0.pc 
prefix=/usr
exec_prefix=${prefix}
libdir=/usr/lib
includedir=${prefix}/include
target=x11

gtk_binary_version=2.10.0
gtk_host=i386-pc-solaris2.11

Name: GTK+
Description: GTK+ Graphical UI Library (${target} target)
Version: 2.14.3
Requires: gdk-${target}-2.0 atk cairo gio-2.0 pangoft2
Libs: -L${libdir} -lgtk-${target}-2.0 
Cflags: -I${includedir}/gtk-2.0 
ssep...@ssepirot61:~/다운로드/murrine-0.90.3$ 

Thank you for your help.

Regards,
ilho Kim.
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

[osol-discuss] Opensolaris and Solaris 10

[KARTIK VASHISHTA]

Solaris 10 is not a bad place to learn. To, among other things, help the cause 
of Opensolaris ,I created a free root account. YOU may, AS USER 'unix' TELNET 
TO ONE SYSTEM, THEN, FROM THERE, AS "root" you will telnet to any of three 
other systems. The system you will telnet to first is: trainingzone.getmyip.com 
once there, you may telnet to any of these systems: AELLIP , AALOROTOM, 
NAUEEREEFS, CBSH

You may install software, install patches, install zfs filesystems(using 
makefiles).

so, first:
telnet trainingzone.getmyip.com
username: unix (no password, hit enter)
Then, telnet to any/all of: AALOROTOM, NAUEEREEFS, CBSH
example: telnet AALOROTOM  

AND/OR

telnet AELLIP 

username:root
password (hit enter)
username: root (no password, hit enter)

Full training(free): http://www.kartik.com/FreeUNIXTraining.html updates on 
accounts will be made here.

Kartik Vashishta
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

[osol-discuss] Tofu

[sunyong]

Give me some reasons why you dont like tofu and reasons why you like tofu?
I LUV TOFU!teh food and my puppy~ 








[URL=http://www.boxset4less.com/products/Supernatural-Seasons-1-3-DVD-Boxset-DVDS-1523.html]Supernatural
 seasons 1-3 
dvd[/URL]||[URL=http://www.boxset4less.com/products/Smallville-Seasons-1-7-DVD-Boxset-DVDS-1250.html]Smallville
 dvd[/URL]||[url=http://www.boxset4less.com]boxset4less[/url]
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

[osol-discuss] What do you think about no bakes!?

[sunyong]

Ahhh... nothing compares to a no-bake cookie? What do you think of 'em? rofl 








[url=http://www.boxset4less.com]boxset4less[/url]
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

[osol-discuss] japanese food or italian food

[sunyong]

Hi! Do u like japanese food or italian food better??? 







[url=http://www.boxset4less.com]boxset4less[/url]
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] when gtk+ update to 2.12 or above

[Andras Barna]

you are using 2008.05?
you should upgrade


On Tue, Apr 28, 2009 at 4:19 AM, ilho.kim  wrote:
> Dear all,
>
> I'm trying to install GTK+ engine 'murrine' which needs GTK+ version 2.12 or 
> above.
> current opensolaris has GTK+ 2.12. When will we have 2.12 or above version of 
> GTK+?
> or we should go sunfreeware.com ?
>
> Thank you.
> 1hoK.
> --
> This message posted from opensolaris.org
> ___
> opensolaris-discuss mailing list
> opensolaris-discuss@opensolaris.org
>



-- 
Andy
http://blog.sartek.net
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

[osol-discuss] when gtk+ update to 2.12 or above

[ilho.kim]

Dear all,

I'm trying to install GTK+ engine 'murrine' which needs GTK+ version 2.12 or 
above. 
current opensolaris has GTK+ 2.12. When will we have 2.12 or above version of 
GTK+?
or we should go sunfreeware.com ?

Thank you.
1hoK.
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Reading Route table

[James Carlson]

Paul Durrant writes:
> Akhil Jain wrote:
> > 
> > The route command is an administrative command. If there are any C api's 
> > which can directly read routing table (w.o. using system() call) ..
> > 
> 
> man -s 7p route
> 
> what other API do you need?

The big part you'd need (and that isn't documented) is the MIB2
interface that allows you to read the table itself, rather than just
searching for individual matches (with RTM_GET) or getting updates.

Look at /usr/include/inet/mib2.h for details, and see netstat.c for an
example usage of it.  (Or pppd's sys-solaris.c for a much simpler
example.)

Portable applications should avoid doing this, though, so I think the
right initial question is "what are you doing and why do you need to
read this table?"

-- 
James Carlson, Solaris Networking  
Sun Microsystems / 35 Network Drive71.232W   Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757   42.496N   Fax +1 781 442 1677
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Setup VNC on OpenSolaris

[Alan Coopersmith]

Sun didn't "change" VNC - builds 111a and later simply have a bug
in Xvnc, which we've now tracked down and are fixing.

6833541 Xvnc will not start in osol 0906: core in XkbDDXLoadKeymapByNames

-Alan Coopersmith-   alan.coopersm...@sun.com
 Sun Microsystems, Inc. - X Window System Engineering



Tom Chen wrote:
> Jerry,
> 
> You do not need to have graphics card if x-server is enabled, you can
> get same graphics using vnc-viewer through network. To enable x-server,
> we can use ssh or telnet to access the box, modify some files and run
> some scripts, so, graphics card is not needed.
> See http://blog.hellpop.com/2009/01/opensolaris-vnc-server.html,
> however, this method does not work in recent OpenSolaris Express
> releases. I do not know why Sun changed again without alternative
> solution which confused its users.
> 
> Tom
> - Original Message - From: "jerry tan" 
> To: "Tom Chen" 
> Cc: "Halton Huo" ;
> 
> Sent: Monday, April 27, 2009 12:32 AM
> Subject: Re: [osol-discuss] Setup VNC on OpenSolaris
> 
> 
>> Hi, tom,
>> if you have no graphics card, how to start Xserver , and start xvn
>> client to connect it?
>>
>> On 04/21/09 02:31, Tom Chen wrote:
>>> Halton,
>>>
>>> Yes, it works! But unfortunately, some of my servers do not have
>>> graphics card, so, have no video output. Is there any script that I
>>> can use instead?
>>>
>>> Tom
>>> - Original Message - From: "Halton Huo" 
>>> To: "Tom Chen" 
>>> Cc: 
>>> Sent: Monday, April 20, 2009 2:25 AM
>>> Subject: Re: [osol-discuss] Setup VNC on OpenSolaris
>>>
>>>

 On Sat, 2009-04-18 at 19:20 -0700, Tom Chen wrote:
> Thanks, I just tried on OpenSolaris nv112, but somehow, does not
> work sigh. We do not use OpenSolaris 2008.11.
 Hi Tom,

 As I know, OpenSolaris 2009.06 snv_112 is not out. I'm using
 OpenSolaris
 2009.06 snv_111 now. System --> Preferences --> Desktop Sharing (Vino)
 works as I wish on it.

 If it does not work, should be a bug. Please file a bug on
 defect.opensolaris.org for Development/gnome/applications.

 Thanks,
 Halton.


>>>
>>> ___
>>> opensolaris-discuss mailing list
>>> opensolaris-discuss@opensolaris.org
>>
>>
> 
> ___
> opensolaris-discuss mailing list
> opensolaris-discuss@opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] uname -a ... 32bit vs 64bit

[Alexander Eremin]

> Octave Orgeron writes:
> > Now putting all that info into uname would just
> complicate life for everyone and break all sorts of
> stuff. As such, uname is for very high-level info and
> isainfo is for detailed.
> 
> Exactly.  Putting detailed processor information into
> uname is not
> only unnecessary, but it breaks backward
> compatibility by forcing
> people who are depending on uname output (often
> inside of 'configure'
> scripts) to deal with new and unexpected responses
> from the system.
> 
> When we've done this in the past, it's been to add
> support for a new
> processor or system family, and not when we've just
> added support for
> yet another variant of an existing processor.
> 
> Breaking existing code can sometimes be the only
> answer, but it'd
> better be for a really good reason.  I'm not sure
> that merely
> disliking "i386" as shorthand for "all Intel and AMD
> x86 compatible
> CPUs" is enough of a reason to make ./configure fall
> over and die.
May be fears are exaggerated? I have Debian amd64 server with tons of various 
services at my office and uname -p shows me "unknown" (this is bug in Debian 
and Ubuntu) but I successfully make all programs that I needed.  It seems to me 
that anything terrible would not occur if the uname -p produced the x86_64 
output on machine with 64-bit cpu.
Nevertheless, looking through internet I found the same discussions for linux 
and some *bsd and looks like thus not only solaris question. Well, will waiting 
for next generation cpu.


Regards,
Alex
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

[osol-discuss] AMD FX790 chip set

[Mike DeMarco]

Has anyone started using the new AM3 motherboards yet, Looking for any issues 
with compatibility.

Thanks
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Setup VNC on OpenSolaris

[Tom Chen]

Jerry,

You do not need to have graphics card if x-server is enabled, you can get 
same graphics using vnc-viewer through network. To enable x-server, we can 
use ssh or telnet to access the box, modify some files and run some scripts, 
so, graphics card is not needed.
See http://blog.hellpop.com/2009/01/opensolaris-vnc-server.html, however, 
this method does not work in recent OpenSolaris Express releases. I do not 
know why Sun changed again without alternative solution which confused its 
users.


Tom
- Original Message - 
From: "jerry tan" 

To: "Tom Chen" 
Cc: "Halton Huo" ; 
Sent: Monday, April 27, 2009 12:32 AM
Subject: Re: [osol-discuss] Setup VNC on OpenSolaris



Hi, tom,
if you have no graphics card, how to start Xserver , and start xvn client 
to connect it?


On 04/21/09 02:31, Tom Chen wrote:

Halton,

Yes, it works! But unfortunately, some of my servers do not have graphics 
card, so, have no video output. Is there any script that I can use 
instead?


Tom
- Original Message - From: "Halton Huo" 
To: "Tom Chen" 
Cc: 
Sent: Monday, April 20, 2009 2:25 AM
Subject: Re: [osol-discuss] Setup VNC on OpenSolaris




On Sat, 2009-04-18 at 19:20 -0700, Tom Chen wrote:
Thanks, I just tried on OpenSolaris nv112, but somehow, does not 
work sigh. We do not use OpenSolaris 2008.11.

Hi Tom,

As I know, OpenSolaris 2009.06 snv_112 is not out. I'm using OpenSolaris
2009.06 snv_111 now. System --> Preferences --> Desktop Sharing (Vino)
works as I wish on it.

If it does not work, should be a bug. Please file a bug on
defect.opensolaris.org for Development/gnome/applications.

Thanks,
Halton.




___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org





___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] [indiana-discuss] Opensolaris on Ultra 10

[mary ding]

Martin:

BTW, with the fix for bugid 6797441, you can actually copy the fixed 
wanboot binary to /platform/sun4u/wanboot on your current system.  Then 
you can boot from disk instead of CD/DVD to do wanboot install.  This 
make testing easier.



ok boot disk -o prompt -F wanboot - install


mary ding wrote:

Martin:

We are aware of the problem with trying to get AI to work on older sun4u 
boxes. When we did the sparc 2009.01.06 support, we are aware of 
problems with booting sun4u.  We also had s10 updates customers filing 
bugs against wanboot.


I filed the following bugs to track the issues that we found with 
nevada/s10 update wanboot install support.


6797441 - wanboot fails to configure network device when booted from media
6805094 - wanboot install from snv and s10u6/s10u7_04 fails after 
loading miniroot.


With the fix for 6797441, I can finally do wanboot install from media 
and disk on older platforms.


However, AI install for osol works differently and we still had issues 
on older platforms.  We had filed the following bug to track the 
additional issues on older sun4u platforms:


6820873 - Export wanboot arguments so that Automated installer can use them





Martin Bochnig wrote:
On Mon, Apr 27, 2009 at 6:54 PM, Danil Smirnov 
 wrote:

Is anybody here who have successfully installed Opensolaris on Ultra 10?

This article 
http://www.opensolaris.org/jive/thread.jspa?threadID=99516&tstart=0
says that I need OBP level of 4.17 or greater for automated 
Installer. But it looks like

latest OBP version for Ultra 10 is 3.31... :-(

I tried to boot from build 111 by `boot cdrom` command nut it says:
"Bad magic number in disk label
Can't open disk label package
Evaluating: boot cdrom
Can't open boot device"

Could you please advice some other method to install Opensolaris? I 
have Solaris 10 installed on the workstation...




Hello, the AI iso is not meant to be bootable.
Please subscribe to caiman discuss.
Follow Sun's instructions (and run into the next bug on almost all pre
4.17.1 legacy sun4u platforms, which I found and for which I
finally found a fix after 2 weeks of insomnia:
http://opensolaris.org/os/project/caiman/auto_install/Documentation/


Be sure to read my recent posts to caiman-discuss (and PLEASE confirm
that this bug exists, thank you) :

http://www.opensolaris.org/jive/thread.jspa?threadID=101275&tstart=0

Yes, with my patches it now works on the Ultra 10.
p.s. You won't need to mess around with bare metal when setting up the
install and dhcp servers (best in same subnet as client and both
servers on same machine, follow the "simple instructions" in the
guide).
However: No bare metal is needed for the server, I successfully use
VirtualBox for this, with x86/x64 OpenSolaris 2009.06.111 (don't take
111a) as guest, and switch VirtualBox's Networking from NAT to
Bridging.
For convenience I can upload the hdd guest image for you.
Otherwise take the iso's gently provided by Ben Rockwood and Al Hopper
from genunix.org.

If you run into whatever SPARC-specific trouble, always feel free to
ping the me via the lists.

Long live sun4u (and me).


%martin
___
indiana-discuss mailing list
indiana-disc...@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/indiana-discuss


___
indiana-discuss mailing list
indiana-disc...@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/indiana-discuss


___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] [indiana-discuss] Opensolaris on Ultra 10

[mary ding]

Martin:

We are aware of the problem with trying to get AI to work on older sun4u 
boxes. When we did the sparc 2009.01.06 support, we are aware of 
problems with booting sun4u.  We also had s10 updates customers filing 
bugs against wanboot.


I filed the following bugs to track the issues that we found with 
nevada/s10 update wanboot install support.


6797441 - wanboot fails to configure network device when booted from media
6805094 - wanboot install from snv and s10u6/s10u7_04 fails after 
loading miniroot.


With the fix for 6797441, I can finally do wanboot install from media 
and disk on older platforms.


However, AI install for osol works differently and we still had issues 
on older platforms.  We had filed the following bug to track the 
additional issues on older sun4u platforms:


6820873 - Export wanboot arguments so that Automated installer can use them





Martin Bochnig wrote:

On Mon, Apr 27, 2009 at 6:54 PM, Danil Smirnov  wrote:

Is anybody here who have successfully installed Opensolaris on Ultra 10?

This article http://www.opensolaris.org/jive/thread.jspa?threadID=99516&tstart=0
says that I need OBP level of 4.17 or greater for automated Installer. But it 
looks like
latest OBP version for Ultra 10 is 3.31... :-(

I tried to boot from build 111 by `boot cdrom` command nut it says:
"Bad magic number in disk label
Can't open disk label package
Evaluating: boot cdrom
Can't open boot device"

Could you please advice some other method to install Opensolaris? I have 
Solaris 10 installed on the workstation...




Hello, the AI iso is not meant to be bootable.
Please subscribe to caiman discuss.
Follow Sun's instructions (and run into the next bug on almost all pre
4.17.1 legacy sun4u platforms, which I found and for which I
finally found a fix after 2 weeks of insomnia:
http://opensolaris.org/os/project/caiman/auto_install/Documentation/


Be sure to read my recent posts to caiman-discuss (and PLEASE confirm
that this bug exists, thank you) :

http://www.opensolaris.org/jive/thread.jspa?threadID=101275&tstart=0

Yes, with my patches it now works on the Ultra 10.
p.s. You won't need to mess around with bare metal when setting up the
install and dhcp servers (best in same subnet as client and both
servers on same machine, follow the "simple instructions" in the
guide).
However: No bare metal is needed for the server, I successfully use
VirtualBox for this, with x86/x64 OpenSolaris 2009.06.111 (don't take
111a) as guest, and switch VirtualBox's Networking from NAT to
Bridging.
For convenience I can upload the hdd guest image for you.
Otherwise take the iso's gently provided by Ben Rockwood and Al Hopper
from genunix.org.

If you run into whatever SPARC-specific trouble, always feel free to
ping the me via the lists.

Long live sun4u (and me).


%martin
___
indiana-discuss mailing list
indiana-disc...@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/indiana-discuss


___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

[osol-discuss] What FC HBA is supported by comstar?

[Chris Du]

I'm trying to find an FC HBA supported by Comstar. The Comstar project states 
4Gb qlogic/Emulex, and 2 Gb Emulex coming. Do we have a list of known HBA that 
works with Comstar?

Thanks
Chris
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] uname -a ... 32bit vs 64bit

[Dennis Clarke]

> Octave Orgeron writes:
>> Now putting all that info into uname would just complicate life for
>> everyone and break all sorts of stuff. As such, uname is for very
>> high-level info and isainfo is for detailed.
>
> Exactly.  Putting detailed processor information into uname is not
> only unnecessary, but it breaks backward compatibility by forcing
> people who are depending on uname output (often inside of 'configure'
> scripts) to deal with new and unexpected responses from the system.
>
> When we've done this in the past, it's been to add support for a new
> processor or system family, and not when we've just added support for
> yet another variant of an existing processor.
>
> Breaking existing code can sometimes be the only answer, but it'd
> better be for a really good reason.  I'm not sure that merely
> disliking "i386" as shorthand for "all Intel and AMD x86 compatible
> CPUs" is enough of a reason to make ./configure fall over and die.

The only thing that could safely be added to uname is perhaps the OS name
just like GNU uname :

$ uname --version
uname (GNU coreutils) 7.2
Copyright (C) 2009 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later
.
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.

Written by David MacKenzie.
$ uname -a
SunOS mimas 5.8 Generic_117350-61 sun4u sparc SUNW,UltraAX-i2 Solaris

$ /sbin/uname -a
SunOS mimas 5.8 Generic_117350-61 sun4u sparc SUNW,UltraAX-i2

Even so, I don't see what the value is there.

Can anyone tell me what OpenSolaris reports for GNU uname ? Does it say
Solaris or OpenSolaris or something else?

Also, one parameter that I expect to work is the capital X :

$ uname -X
uname: invalid option -- X
Try `uname --help' for more information.

Which doe work with UNIX uname :

$ /sbin/uname -X
System = SunOS
Node = mimas
Release = 5.8
KernelID = Generic_117350-61
Machine = sun4u
BusType = 
Serial = 
Users = 
OEM# = 0
Origin# = 1
NumCPU = 1

Now that I think about it, messing with uname serves no purpose at all.

Dennis Clarke

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] [indiana-discuss] Opensolaris on Ultra 10

[Martin Bochnig]

On Mon, Apr 27, 2009 at 8:52 PM, mary ding  wrote:
> Martin:
>
> BTW, with the fix for bugid 6797441, you can actually copy the fixed wanboot
> binary to /platform/sun4u/wanboot on your current system.  Then you can boot
> from disk instead of CD/DVD to do wanboot install.  This make testing
> easier.
>
>
> ok boot disk -o prompt -F wanboot - install


Yes, that's what I did during my tests: On one machine I had all
editor tabs open (and kept them), then I always copied a bldenv
recompiled wanboot bin over to other box'es disks.

(Some people, e.g.) Bernd Schemmer also once desribed it on Bigadmin:
http://www.sun.com/bigadmin/content/submitted/wanboot_from_disk.html
Actually it doesn't matter too much, on which medium it resides, as
long as the medium can be booted off OBP (well, this limits it a bit
to IDE/SCSI (and FCAL), SATA/SAS disk, CD/DVD, tape, arp/rarp).
Of course with rare exceptions like the SF280R (or in other situations
other boxes), where you get the annoying and feared media related mmu
trap error.
A good OBP-reference is the open src sun4v OBP code from opensparc.net
as well as documentation such as from http://playground.sun.com/1275/,
plus promif and related code and comments inside OS/Net.


Still online, but now I go.
I respond in a few hours, from home.
Mary, thank your for your responses, bug-ids, references.
Looks good now, looks as if we have it working now, on most boxes.


Regards,
%martin
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] uname -a ... 32bit vs 64bit

[James Carlson]

Octave Orgeron writes:
> Now putting all that info into uname would just complicate life for everyone 
> and break all sorts of stuff. As such, uname is for very high-level info and 
> isainfo is for detailed.

Exactly.  Putting detailed processor information into uname is not
only unnecessary, but it breaks backward compatibility by forcing
people who are depending on uname output (often inside of 'configure'
scripts) to deal with new and unexpected responses from the system.

When we've done this in the past, it's been to add support for a new
processor or system family, and not when we've just added support for
yet another variant of an existing processor.

Breaking existing code can sometimes be the only answer, but it'd
better be for a really good reason.  I'm not sure that merely
disliking "i386" as shorthand for "all Intel and AMD x86 compatible
CPUs" is enough of a reason to make ./configure fall over and die.

-- 
James Carlson, Solaris Networking  
Sun Microsystems / 35 Network Drive71.232W   Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757   42.496N   Fax +1 781 442 1677
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] [indiana-discuss] Opensolaris on Ultra 10

[Martin Bochnig]

Mary, thanks for the bugids.

I review them when at home, right now I leave office.
After reviewing them I respond in a detailed way, with the patches and a
 /platform/sun4u/wanboot test binary.

Thank you for the confirmation.


Best,
%martin

On Mon, Apr 27, 2009 at 7:54 PM, mary ding  wrote:
> Martin:
>
> We are aware of the problem with trying to get AI to work on older sun4u
> boxes. When we did the sparc 2009.01.06 support, we are aware of problems
> with booting sun4u.  We also had s10 updates customers filing bugs against
> wanboot.
>
> I filed the following bugs to track the issues that we found with nevada/s10
> update wanboot install support.
>
> 6797441 - wanboot fails to configure network device when booted from media
> 6805094 - wanboot install from snv and s10u6/s10u7_04 fails after loading
> miniroot.
>
> With the fix for 6797441, I can finally do wanboot install from media and
> disk on older platforms.
>
> However, AI install for osol works differently and we still had issues on
> older platforms.  We had filed the following bug to track the additional
> issues on older sun4u platforms:
>
> 6820873 - Export wanboot arguments so that Automated installer can use them
>
>
>
>
>
> Martin Bochnig wrote:
>>
>> On Mon, Apr 27, 2009 at 6:54 PM, Danil Smirnov 
>> wrote:
>>>
>>> Is anybody here who have successfully installed Opensolaris on Ultra 10?
>>>
>>> This article
>>> http://www.opensolaris.org/jive/thread.jspa?threadID=99516&tstart=0
>>> says that I need OBP level of 4.17 or greater for automated Installer.
>>> But it looks like
>>> latest OBP version for Ultra 10 is 3.31... :-(
>>>
>>> I tried to boot from build 111 by `boot cdrom` command nut it says:
>>> "Bad magic number in disk label
>>> Can't open disk label package
>>> Evaluating: boot cdrom
>>> Can't open boot device"
>>>
>>> Could you please advice some other method to install Opensolaris? I have
>>> Solaris 10 installed on the workstation...
>>
>>
>>
>> Hello, the AI iso is not meant to be bootable.
>> Please subscribe to caiman discuss.
>> Follow Sun's instructions (and run into the next bug on almost all pre
>> 4.17.1 legacy sun4u platforms, which I found and for which I
>> finally found a fix after 2 weeks of insomnia:
>> http://opensolaris.org/os/project/caiman/auto_install/Documentation/
>>
>>
>> Be sure to read my recent posts to caiman-discuss (and PLEASE confirm
>> that this bug exists, thank you) :
>>
>> http://www.opensolaris.org/jive/thread.jspa?threadID=101275&tstart=0
>>
>> Yes, with my patches it now works on the Ultra 10.
>> p.s. You won't need to mess around with bare metal when setting up the
>> install and dhcp servers (best in same subnet as client and both
>> servers on same machine, follow the "simple instructions" in the
>> guide).
>> However: No bare metal is needed for the server, I successfully use
>> VirtualBox for this, with x86/x64 OpenSolaris 2009.06.111 (don't take
>> 111a) as guest, and switch VirtualBox's Networking from NAT to
>> Bridging.
>> For convenience I can upload the hdd guest image for you.
>> Otherwise take the iso's gently provided by Ben Rockwood and Al Hopper
>> from genunix.org.
>>
>> If you run into whatever SPARC-specific trouble, always feel free to
>> ping the me via the lists.
>>
>> Long live sun4u (and me).
>>
>>
>> %martin
>> ___
>> indiana-discuss mailing list
>> indiana-disc...@opensolaris.org
>> http://mail.opensolaris.org/mailman/listinfo/indiana-discuss
>
>
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] uname -a ... 32bit vs 64bit

[Octave Orgeron]

Technically, i386 is correct. It's the base processor ISA architecture. Most 
software is still 32 bit, so backwards compatibility is key, as a result the 
lowest common denominator on the x64/x68 platform is i386. This is the same 
result as you would expect on SPARC where uname -p will report "sparc". Even 
though it could an ultrasparc, sparc64, etc. The point being that uname -p will 
report the lowest common platform denominator.

So now.. the question becomes.. why?

Well you wouldn't want software or libraries to barf all over you because the 
lowest common platform denominator was amd64 would you? Last time I looked.. 
few OS's have the kernel and user-land in 64bit like Tru64 Unix. So this is 
important for applications, compilers, installation scripts, etc. When you want 
to compile 64bit apps or even take advantage of CPU specific features and what 
not.. that's where isainfo becomes key to laying out what your processor 
supports. Good example:

$ uname -a
SunOS katana 5.11 snv_101b i86pc i386 i86pc
$ isainfo
amd64 i386
$ isainfo -v
64-bit amd64 applications
ahf sse3 sse2 sse fxsr amd_3dnowx amd_3dnow amd_mmx mmx cmov amd_sysc 
cx8 tsc fpu 
32-bit i386 applications
ahf sse3 sse2 sse fxsr amd_3dnowx amd_3dnow amd_mmx mmx cmov amd_sysc 
cx8 tsc fpu 

Now putting all that info into uname would just complicate life for everyone 
and break all sorts of stuff. As such, uname is for very high-level info and 
isainfo is for detailed.

 *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*
Octave J. Orgeron
Solaris Virtualization Architect and Consultant
Web: http://unixconsole.blogspot.com
E-Mail: unixcons...@yahoo.com
*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*



- Original Message 
From: Alexander Eremin 
To: opensolaris-discuss@opensolaris.org
Sent: Monday, April 27, 2009 11:35:26 AM
Subject: Re: [osol-discuss] uname -a ... 32bit vs 64bit

> 
> 
> Alexander Eremin wrote:
> > Hi,
> > I see now it's a bug 2820 "Uname -m" does not
> report Operating System's mode correctly (32 or 64
> bit).
> 
> It's not supposed to:
> 
> -mPrints the machine hardware
>  name  (class).
> Use  of  this  option  is
> discouraged. Use
> uname -p instead. 
I'm about 'uname -p'. I think it's not  normal when 'uname -p' tell me 'i386' 
on machine with 64bit cpu..

Regards,
Alex
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org



  
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] uname -a ... 32bit vs 64bit

[Martin Bochnig]

On Mon, Apr 27, 2009 at 7:26 PM, Alan Coopersmith
 wrote:
>
> Martin Bochnig wrote:
>> Even on 2009.06_b111:
>> mar...@opensolaris:~$ uname -m
>> i86pc
>> mar...@opensolaris:~$ uname -p
>> i386
>> mar...@opensolaris:~$ uname -a
>> SunOS opensolaris 5.11 snv_111 i86pc i386 i86pc Solaris
>> mar...@opensolaris:~$ arch -k
>> i86pc
>> mar...@opensolaris:~$ isainfo -k
>> amd64
>> --->>
>> Everything but isainfo shows incorrect results (not only o this box, on 
>> _many_).
>> <<---
>
> Those all look correct to me.   Would you want your box to report that it 
> can't
> run i386 binaries and break far more software and scripts?


Yes, exactly.
Please.

Let's look 10 years ahead.
I'm not against the Solaris way (in contrast to LinUX) that bo th
kernels are shipped in the same distro and the right one is chosen
automatically.
Further: Be aware that I am - and always have been - totally
pro- 32bit_userland_libs_all_here___/___$plat64_all_there on the same system.
That's great.

But IMO - as per definition - when you boot into a 32bit kernel you
are not running the same platform, as if you boot the platform's
corresponding 64bit kernel.
Those two (such as sparcv8plus vs. sparcv9 on Ultra I and II systems,
or i386 vs. amd64 on the PC, are _distinct_ platforms.
And as long as arch or uname output dos not reflect this, I consider
it a bug, very objectively do so.

I see the point, but have a different opinion for a few years already,
as written earlier - from time to time.


--
%martin
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] uname -a ... 32bit vs 64bit

[Alexander Eremin]

> 
> 
> Alexander Eremin wrote:
> > Hi,
> > I see now it's a bug 2820 "Uname -m" does not
> report Operating System's mode correctly (32 or 64
> bit).
> 
> It's not supposed to:
> 
> -mPrints the machine hardware
>  name  (class).
> Use  of  this  option  is
> discouraged. Use
> uname -p instead. 
I'm about 'uname -p'. I think it's not  normal when 'uname -p' tell me 'i386' 
on machine with 64bit cpu..

Regards,
Alex
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] uname -a ... 32bit vs 64bit

[Alan Coopersmith]

Martin Bochnig wrote:
> Even on 2009.06_b111:
> mar...@opensolaris:~$ uname -m
> i86pc
> mar...@opensolaris:~$ uname -p
> i386
> mar...@opensolaris:~$ uname -a
> SunOS opensolaris 5.11 snv_111 i86pc i386 i86pc Solaris
> mar...@opensolaris:~$ arch -k
> i86pc
> mar...@opensolaris:~$ isainfo -k
> amd64
> --->>
> Everything but isainfo shows incorrect results (not only o this box, on 
> _many_).
> <<---

Those all look correct to me.   Would you want your box to report that it can't
run i386 binaries and break far more software and scripts?

-- 
-Alan Coopersmith-   alan.coopersm...@sun.com
 Sun Microsystems, Inc. - X Window System Engineering

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Opensolaris on Ultra 10

[Martin Bochnig]

On Mon, Apr 27, 2009 at 6:54 PM, Danil Smirnov  wrote:
> Is anybody here who have successfully installed Opensolaris on Ultra 10?
>
> This article 
> http://www.opensolaris.org/jive/thread.jspa?threadID=99516&tstart=0
> says that I need OBP level of 4.17 or greater for automated Installer. But it 
> looks like
> latest OBP version for Ultra 10 is 3.31... :-(
>
> I tried to boot from build 111 by `boot cdrom` command nut it says:
> "Bad magic number in disk label
> Can't open disk label package
> Evaluating: boot cdrom
> Can't open boot device"
>
> Could you please advice some other method to install Opensolaris? I have 
> Solaris 10 installed on the workstation...



Hello, the AI iso is not meant to be bootable.
Please subscribe to caiman discuss.
Follow Sun's instructions (and run into the next bug on almost all pre
4.17.1 legacy sun4u platforms, which I found and for which I
finally found a fix after 2 weeks of insomnia:
http://opensolaris.org/os/project/caiman/auto_install/Documentation/


Be sure to read my recent posts to caiman-discuss (and PLEASE confirm
that this bug exists, thank you) :

http://www.opensolaris.org/jive/thread.jspa?threadID=101275&tstart=0

Yes, with my patches it now works on the Ultra 10.
p.s. You won't need to mess around with bare metal when setting up the
install and dhcp servers (best in same subnet as client and both
servers on same machine, follow the "simple instructions" in the
guide).
However: No bare metal is needed for the server, I successfully use
VirtualBox for this, with x86/x64 OpenSolaris 2009.06.111 (don't take
111a) as guest, and switch VirtualBox's Networking from NAT to
Bridging.
For convenience I can upload the hdd guest image for you.
Otherwise take the iso's gently provided by Ben Rockwood and Al Hopper
from genunix.org.

If you run into whatever SPARC-specific trouble, always feel free to
ping the me via the lists.

Long live sun4u (and me).


%martin
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] uname -a ... 32bit vs 64bit

[Martin Bochnig]

On Mon, Apr 27, 2009 at 6:53 PM, Martin Bochnig  wrote:
> On Mon, Apr 27, 2009 at 5:45 PM, Alan Coopersmith
>  wrote:
>>
>>
>> Alexander Eremin wrote:
>>> Hi,
>>> I see now it's a bug 2820 "Uname -m" does not report Operating System's 
>>> mode correctly (32 or 64 bit).
>>
>> It's not supposed to:
>>
>>     -m                Prints the machine hardware name  (class).
>>                       Use  of  this  option  is discouraged. Use
>>                       uname -p instead. See NOTES section below.
>>
>>     The arch -k and uname -m commands return equivalent  values;
>>     however,  the use of either of these commands by third party
>>     programs is discouraged, as is the use of the  arch  command
>>     in  general.  To  determine  the  machine's  Instruction Set
>>     Architecture (ISA or processor type), use uname with the  -p
>>     option.
>
>
>
> isainfo -k was always the "recommended" way.



And still!!
Even on 2009.06_b111:

mar...@opensolaris:~$ prtdiag
System Configuration: FUJITSU SIEMENS AMILO Notebook Pa 3515
BIOS Configuration: Phoenix Technologies LTD V1.1310/06/2008

 Processor Sockets 

Version  Location Tag
 --
Athlon X2 QL-62  Socket S1G2

 Memory Device Sockets 

TypeStatus Set Device Locator  Bank Locator
--- -- --- --- 
DDR2in use 1   S1  DIMM1
DDR2in use 2   S2  DIMM2

 On-Board Devices =
ATI RS690M
ESS 1869

 Upgradeable Slots 

ID  StatusType Description
--- -  
11  available PCI  MINI PCI
mar...@opensolaris:~$ psrinfo -v
Status of virtual processor 0 as of: 04/27/2009 18:54:26
  on-line since 04/27/2009 14:37:36.
  The i386 processor operates at 2000 MHz,
and has an i387 compatible floating point processor.
Status of virtual processor 1 as of: 04/27/2009 18:54:26
  on-line since 04/27/2009 14:37:39.
  The i386 processor operates at 2000 MHz,
and has an i387 compatible floating point processor.
mar...@opensolaris:~$ uname -m
i86pc
mar...@opensolaris:~$ uname -p
i386
mar...@opensolaris:~$ uname -a
SunOS opensolaris 5.11 snv_111 i86pc i386 i86pc Solaris
mar...@opensolaris:~$ arch -k
i86pc
mar...@opensolaris:~$ isainfo -k
amd64
mar...@opensolaris:~$


--->>
Everything but isainfo shows incorrect results (not only o this box, on _many_).
<<---


>
> %martin
>
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

[osol-discuss] Opensolaris on Ultra 10

[Danil Smirnov]

Is anybody here who have successfully installed Opensolaris on Ultra 10?

This article http://www.opensolaris.org/jive/thread.jspa?threadID=99516&tstart=0
says that I need OBP level of 4.17 or greater for automated Installer. But it 
looks like
latest OBP version for Ultra 10 is 3.31... :-(

I tried to boot from build 111 by `boot cdrom` command nut it says:
"Bad magic number in disk label
Can't open disk label package
Evaluating: boot cdrom
Can't open boot device"

Could you please advice some other method to install Opensolaris? I have 
Solaris 10 installed on the workstation...
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] uname -a ... 32bit vs 64bit

[Martin Bochnig]

On Mon, Apr 27, 2009 at 5:45 PM, Alan Coopersmith
 wrote:
>
>
> Alexander Eremin wrote:
>> Hi,
>> I see now it's a bug 2820 "Uname -m" does not report Operating System's mode 
>> correctly (32 or 64 bit).
>
> It's not supposed to:
>
>     -m                Prints the machine hardware name  (class).
>                       Use  of  this  option  is discouraged. Use
>                       uname -p instead. See NOTES section below.
>
>     The arch -k and uname -m commands return equivalent  values;
>     however,  the use of either of these commands by third party
>     programs is discouraged, as is the use of the  arch  command
>     in  general.  To  determine  the  machine's  Instruction Set
>     Architecture (ISA or processor type), use uname with the  -p
>     option.



isainfo -k was always the "recommended" way.

%martin
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] uname -a ... 32bit vs 64bit

[Alan Coopersmith]

Alexander Eremin wrote:
> Hi,
> I see now it's a bug 2820 "Uname -m" does not report Operating System's mode 
> correctly (32 or 64 bit).

It's not supposed to:

 -mPrints the machine hardware name  (class).
   Use  of  this  option  is discouraged. Use
   uname -p instead. See NOTES section below.

 The arch -k and uname -m commands return equivalent  values;
 however,  the use of either of these commands by third party
 programs is discouraged, as is the use of the  arch  command
 in  general.  To  determine  the  machine's  Instruction Set
 Architecture (ISA or processor type), use uname with the  -p
 option.

-- 
-Alan Coopersmith-   alan.coopersm...@sun.com
 Sun Microsystems, Inc. - X Window System Engineering

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Java Desktop

[Ché Kristo]

Java Desktop System is just a brand, it always has been GNOME with a Sun theme. 

JDS is Gnome!
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] snv_111a kernel crashes at boot

[John Martin]

Bill Shannon wrote:


It looks like gfxp_private::gfxp_vgatext_detatch() is being called 
twice,

once when gfx_private::gfxp_vgatext_attach() detects an error and cleans
up after itself and again after it returns the DDI_FAILURE error code
to the atiatom module and it calls gfxp_vgatext_detach() again from 
atiatom_detach().

mutex_destroy() gets called twice for the same object.

File a bug.  The gfx_private driver needs to keep state on whether 
the lock

needs to be destroyed in case its detach is called multiple times.

However, the bigger problem will eventually be why the attach function
failed since all it does is basic aperture mapping.


Filing a bug for this is a waste of my time.  If no one else is running
into this, it's not going to be fixed anytime soon.  And if other people
*are* running into it, it will be fixed without *me* filing a bug.

It's less frustrating to just hope that the next build will be better.

At least, that's been my experience so far.

What I'm hoping for is some sort of workaround...


Bugtraq 6834464.   The mutex being destroyed twice by multiple detaches was
added in this change:

 03-Mar-2009 6810939  system should be more open to being interactive 
despite graphical boot



___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] uname -a ... 32bit vs 64bit

[Martin Bochnig]

On Mon, Apr 27, 2009 at 1:18 PM, Alexander Eremin  wrote:
> Hi,
> I see now it's a bug 2820 "Uname -m" does not report Operating System's mode 
> correctly (32 or 64 bit).
> Why not be use SI_ARCHITECTURE_K instead of  SI_ARCHITECTURE for -p flag 
> (-m?) in "uname" code?
> After changing this I tested uname on usual 32bit PC and on my Toshiba A300 
> laptop, which cpu has 64bit support and OpenSolaris loads 64bit kernel on it 
> (but old uname reporting i386):
> # ./new.uname -p    # on 32bit PC
> i386
> # ./new.uname -p   # on Toshiba
> amd64
>
>
> Regards,
> Alexander Eremin


+5

I have the same thing on my local systems for quite some time.
But nobody understood it as a bug so I kept my mouth shut, after my
initial complaints in 2006 as well as a few weeks agin in this thread.


%martin
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] amsn versus ekiga

[Joerg Schilling]

solarg  wrote:

> > amsn is low video quality (you first need to fix a broken library to
> > send video anyway) and no sound.
> > 
> that's not what gilles gravier is saying

What is he saying? Did something change during the past month?

> > ekiga is high quality video but I have not been able to use it behind
> > a NAT firewall on VDSL at home.
> > 
>
> i used ekiga at home (adsl) behind a nat firewall, without problems.

I have a SIP aware DSL router, maybe this is the problem.

Jörg

-- 
 EMail:jo...@schily.isdn.cs.tu-berlin.de (home) Jörg Schilling D-13353 Berlin
   j...@cs.tu-berlin.de(uni)  
   joerg.schill...@fokus.fraunhofer.de (work) Blog: 
http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] uname -a ... 32bit vs 64bit

[Alexander Eremin]

Hi,
I see now it's a bug 2820 "Uname -m" does not report Operating System's mode 
correctly (32 or 64 bit).
Why not be use SI_ARCHITECTURE_K instead of  SI_ARCHITECTURE for -p flag (-m?) 
in "uname" code?
After changing this I tested uname on usual 32bit PC and on my Toshiba A300 
laptop, which cpu has 64bit support and OpenSolaris loads 64bit kernel on it 
(but old uname reporting i386):
# ./new.uname -p# on 32bit PC
i386
# ./new.uname -p   # on Toshiba
amd64


Regards,
Alexander Eremin
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] amsn versus ekiga

[solarg]

Joerg Schilling wrote:

solarg  wrote:


so my guess is that my webcam is "high-end USB-VC webcam"

on amsn website, there is a tool to test webcam:
http://amsn.sourceforge.net/devwiki/tiki-index.php?page=Webcam+In+aMSN#test
but no luck


amsn is low video quality (you first need to fix a broken library to
send video anyway) and no sound.


that's not what gilles gravier is saying


ekiga is high quality video but I have not been able to use it behind
a NAT firewall on VDSL at home.



i used ekiga at home (adsl) behind a nat firewall, without problems.

thanks for reply,

gerard

___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Bootup script

[Jim Klimov]

In traditional Solaris, you have /etc/rc2.d/* scripts which are executed during
networked system initialization. Here's where you put most of the lowlevel
bootups, such as networking or driver initialization, UPS monitoring, checking
for first-reboot-after-upgrade, etc. 

It used to end with /etc/rc2.d/S99dtlogin to launch Xwindows login. So I guess
your pre-login scripts go here.

Then you have /etc/rc3.d which stands for "multiuser networking" - here are 
some additional heavier services such as databases and app servers. They can
be launched after presenting a login prompt (console or graphics).

In Solaris 10 this is interweaved with SMF service framework, where rc2.d and
rc3.d scripts are processed at different "SMF milestones" during boot as they
are automatically "legacy" dependencies for these milestones.

Also note that rc*.d scripts are always executed by /sbin/sh regardless of your
header line, so you can't use all that fancy bash or ksh syntax - unless your 
init
script is a wrapper for another bash script (perhaps also with a "su -" to lower
privileges), that is ;)

//Hope this helps, Jim
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] amsn versus ekiga

[Joerg Schilling]

solarg  wrote:

> so my guess is that my webcam is "high-end USB-VC webcam"
>
> on amsn website, there is a tool to test webcam:
> http://amsn.sourceforge.net/devwiki/tiki-index.php?page=Webcam+In+aMSN#test
> but no luck

amsn is low video quality (you first need to fix a broken library to
send video anyway) and no sound.

ekiga is high quality video but I have not been able to use it behind
a NAT firewall on VDSL at home.

Jörg

-- 
 EMail:jo...@schily.isdn.cs.tu-berlin.de (home) Jörg Schilling D-13353 Berlin
   j...@cs.tu-berlin.de(uni)  
   joerg.schill...@fokus.fraunhofer.de (work) Blog: 
http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

[osol-discuss] amsn versus ekiga

[solarg]

hello all,
after reading this article:
http://blogs.sun.com/gravax/entry/building_amsn_with_audio_and
i gave a try with my laptop os2008.11, and the only thing that doesn't 
work is webcam.
The point is that the same webcam is running well with ekiga, but amsn 
complains that it deosn't found it.

In messages, i got:
Apr 27 12:09:39 delljm usba: [ID 912658 kern.info] USB 2.0 device 
(usb46d,8c3) operating at hi speed (USB 2.x) on USB 2.0 root hub: 
dev...@5, usb_mid1 at bus address 2
Apr 27 12:09:39 delljm genunix: [ID 936769 kern.info] usb_mid1 is 
/p...@0,0/pci1028,1...@1d,7/dev...@5
Apr 27 12:09:39 delljm genunix: [ID 408114 kern.info] 
/p...@0,0/pci1028,1...@1d,7/dev...@5 (usb_mid1) online
Apr 27 12:09:39 delljm usba: [ID 912658 kern.info] USB 2.0 
interface-association (usbia46d,8c3.config1.0) operating at hi speed 
(USB 2.x) on USB 2.0 root hub: interface-associat...@0, usbvc0 at bus 
address 2
Apr 27 12:09:39 delljm genunix: [ID 936769 kern.info] usbvc0 is 
/p...@0,0/pci1028,1...@1d,7/dev...@5/interface-associat...@0
Apr 27 12:09:39 delljm genunix: [ID 408114 kern.info] 
/p...@0,0/pci1028,1...@1d,7/dev...@5/interface-associat...@0 (usbvc0) online


so my guess is that my webcam is "high-end USB-VC webcam"

on amsn website, there is a tool to test webcam:
http://amsn.sourceforge.net/devwiki/tiki-index.php?page=Webcam+In+aMSN#test
but no luck

does anybody can help?

thanks in advance,

gerard


___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Question about (Open)Solaris VM - why is swap required?

[Jim Klimov]

Now, one of my questions remains a mystery: why did the system initially stop 
giving out virtual memory when it hit approx half of its RAM (8Gb of 16Gb)?
Was it just a coincidence?

Jim Laurent suggests that Solaris can be configured without swap whatsoever,
and I thought so too - until I hit these problems above ;)

Is there some tool to check which processes reserved virtual memory and how
much? I played around with "ipcs", but I guess I didn't find the correct 
options 
yet... or this should be seeked elsewhere.

Also, another question arose. I found that /tmp filesystem size can be limited 
by settings in /etc/system, i.e. allowing /tmp to consume no more than 200Mb.
Is it possible to instead "reserve" some virtual memory for filesystems like 
/tmp,
/var/run, /etc/svc/volatile and so on - so they would have some small memory
to work with even when the processes have not enough virtual memory to fork?

Many of the services seem to die, or can't restart after crashing, because they 
can't update some lock file in /var/run or for similar (FS-based) reasons :(

//Jim
-- 
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Re: [osol-discuss] Question about (Open)Solaris VM - why is swap required?

[Jim Klimov]

First of all, thank you all for the links and suggestions. The first two 
comments 
in Jim Laurent's entry do seem like they are about my situation :)

As they write, "It is possible to be out of swap without being out of memory 
... 
because memory allocations count immediately against swap but do not count 
against "physical memory" until the allocated memory (reserved swap) is used".

This makes sense, however according to the tools I have (and among them "top"
because it shows total and "free" RAM and disk swap space, as well as "vmstat"
and now "swap -s") the system never went into on-disk swapping. I guess it 
might "reserve" for memory allocations from the disk swap space, and reduce
the common swap (as total virtual memory) *available for further reservations* -
as seen in "df -k /tmp", "swap -s", "vmstat"'s swap column, etc. However much 
of this reservation was never used, so all needed data and heaps fit into RAM 
and never overflowed to disk, so swap space wasn't "used" indeed.

I think I'll stick with this theory for now ;)

Concerning your questions for further details, here's some system stats from 
when we did another run of the tests (at that moment they spawned about 1200
sun-apache22 httpd prefork processes banging into sun-mysql50 database for
a sample webshop site). Database tuning was rather random, with settings taken
from various suggestions over the Net, so it might well overconsume memory
to improve query caching and reduce on-disk activity.

According to the "free" physical and swap memories in "top", the system went
down to 1.5Gb free RAM and did not use any noticeable disk swap (15Gb/15Gb 
free).

These results were taken at somewhat differents points in time (up to 10-20
minutes apart), but the system was under similar load for a while - so the 
numbers below may not match each other exactly, but they are in the same 
range. Due to the nature of the load, there might be free CPU once in a while
(as captured in top header below).

The moment of "top header" and "vmstat output" is particularly interesting. 
It is at these screenshot moments in the night that my sshd crashed, I guess. 
As we can see from "top", not all memory was consumed, and the load average 
was lower than some time before this moment... 

According to vmstat, the "swap" column was at about 7.3Gb at the moment ssh 
connections were lost. The number of kernel read threads jumped up (they
previously ranged from 0 to 100-150); however pi (page-in), po (page-out), 
sr (scan rate) were zero as far up as my history contained - hence, no on-disk 
swapping nor RAM shortage... right?

The trick is, httpd/mysql tests were running in a local zone with resource 
capping enabled (physical < swap < "real global resources"), and my sshd was 
connected to the global zone with presumably more resources over the local 
zone's limits.

The local zone's /var/adm/messages log file is full of "can't fork" and "swap 
space limit exceeded" reports, while there are almost no such reports in the
global zone's log. Sadly, I hoped resource capping could protect the global zone
as a sort of hypervisor, from runaway local zones...

==
Memstat output:

# echo "::memstat" | mdb -k
Page SummaryPagesMB  %Tot
     
Kernel 741164  5790   36%
Anon   916649  7161   45%
Exec and libs   13069   1021%
Page cache  65140   5083%
Free (cachelist) 6887530%
Free (freelist)299747  2341   15%

Total 2042656 15958
Physical  2012133 15719


==
"swap -s" output:

# swap -s
total: 7978600k bytes allocated + 5322840k reserved = 13301440k used, 10007480k 
available

==
"df -k /tmp" output:

# df -k /tmp
Filesystem size   used  avail capacity  Mounted on
swap   9.7G   200K   9.7G 1%/tmp

==
"top" header:

load averages:  90.2, 148.5, 138.7;up 2+14:45:22   03:04:37
1184 processes: 1123 sleeping, 43 running, 2 zombie, 16 on cpu
CPU states: 29.8% idle, 14.6% user, 55.6% kernel,  0.0% iowait,  0.0% swap
Memory: 16G phys mem, 1590M free mem, 15G total swap, 15G free swap

==
"vmstat" tail:

 kthr  memorypagedisk  faults  cpu
 r b w   swap  free  re  mf pi po fr de sr rm s0 s1 s2   in   sy   cs us sy id
 0 0 0 7263832 1625552 123 155 0 0 0  0  0  0  0  0  0 5730 164041 9080 27 18 54
 0 0 0 7274024 1630608 245 400 0 0 0  0  0  0  0  0  0 4823 53903 7640 12 13 75
 0 0 0 7294872 1640064 126 175 0 0 0  0  0  0  1  1  0 2300 22

Re: [osol-discuss] Setup VNC on OpenSolaris

[jerry tan]

Hi, tom,
if you have no graphics card, how to start Xserver , and start xvn 
client to connect it?


On 04/21/09 02:31, Tom Chen wrote:

Halton,

Yes, it works! But unfortunately, some of my servers do not have 
graphics card, so, have no video output. Is there any script that I 
can use instead?


Tom
- Original Message - From: "Halton Huo" 
To: "Tom Chen" 
Cc: 
Sent: Monday, April 20, 2009 2:25 AM
Subject: Re: [osol-discuss] Setup VNC on OpenSolaris




On Sat, 2009-04-18 at 19:20 -0700, Tom Chen wrote:
Thanks, I just tried on OpenSolaris nv112, but somehow, does not 
work sigh. We do not use OpenSolaris 2008.11.

Hi Tom,

As I know, OpenSolaris 2009.06 snv_112 is not out. I'm using OpenSolaris
2009.06 snv_111 now. System --> Preferences --> Desktop Sharing (Vino)
works as I wish on it.

If it does not work, should be a bug. Please file a bug on
defect.opensolaris.org for Development/gnome/applications.

Thanks,
Halton.




___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org


___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org