headers
rwager1 | 1 Nov 02:40
Picon

Re: SATA, ZFS & disk errors

Hello Laurent,

Did you resolve this problem? I have had similar problems with SATA
and ZFS, although using the AHCI driver. Basically, I removed the
disks in question and checked them on another system (Windows) using
the disk manufacturer's (Seagate) disk check software. Turned out that
of the 6 Seagates I had ordered, 5 were bad. Cost me a great deal of
time because I was reluctant to believe that I had several bad drives,
it "had to be" something in Solaris or the AHCI driver. 

Rick

--- In solarisx86 <at> yahoogroups.com, "laurentelanor" <laurent@...> wrote:
>
> Hi guys,
> 
> It's seems I'm unlucky with SATA disks: there was a single, but
> uncorrectable read error on a given block of one of my 4 disk RAIDZ2
> volume last night.
> 
> Since then, again, all I/O stopped to that volume, which is not nice,
> and a reboot failed.
> 
> I'll have a busy evening, obviously. Is there a way to know if this
> comes from the si3124 driver or ZFS itself?
> 
> This is S10U4, not very much patched, running 32 bit and since there's
> no patch for it yet, with a slightly modified version of the si3124
> driver.
>
(Continue reading)

Permalink | Reply | 3 comments |
headers
Apostolos Syropoulos | 1 Nov 11:02
Picon
Favicon

Re: xVM platform supports HVM


Today I played a little bit with Xen. I created an hvn file
(WINXP.hvm), which is almost identical to the one presented at

http://www.opensolaris.org/os/community/xen/docs/HVMdomains.htm

Almost identical because the disks section is the following:

disk = [ 'file:/usr/local/tmp/winxp.img,hdc,w', 
         'file:/usr/local/tmp/diskD.img,hdd,w' ]

File winxp.img is a virtual disk created with qemu that
worked just fine with qemu. Now,  I entered the following
command 

xm create WINXP.hvm 

and after a while a window with Windows XP showed up. 
The average memory consumption was 1,8 GiB (i.e., 97.6%)
and the average CPU load was around 40-50% for the first 
core and 80% for the second core. After it installed
a number of drives, I rebooted Windows and I left for a while.
When I returned my system was dead! I had to reboot it and to
do fsck etc. In other words, my first experience with Xen was 
more than disappointing. It is slow, really slow, and compared
to qemu it does not show any real advantage. So why all this
fuss about Xen and processor virtualization? 

A.S.
(Continue reading)

Permalink | Reply | 3 comments |
headers
Francois Dion | 1 Nov 13:42
Picon

Project Indiana

It's available to download.

Didn't want to blog until I made sure I had a good ISO downloaded and
decent bandwidth (350KB/s) until I did  :)

Anyway, links and details here:

http://solarisdesktop.blogspot.com/2007/11/project-indiana.html

BTW, this is GNOME. Select the nimbus theme to get back in part the
look and feel of JDS (you'll have to move stuff around too).

Francois

Please check the Links page before posting:
    http://groups.yahoo.com/group/solarisx86/links
Post message: solarisx86 <at> yahoogroups.com
UNSUBSCRIBE:  solarisx86-unsubscribe <at> yahoogroups.com
Permalink | Reply | 17 comments |
headers
Greg | 1 Nov 15:17
Picon
Favicon

Solaris 10 jumpstart fails on fork failed too many porcesses


On a Dell 650, after completing "install_begin",
I get an error like:

/usr/sbin/ parse_dynamic_clustertoc: fork failed too many processes 

Has anybody encountered this and what is the likely patch to solve?

Thanks,

Greg

Please check the Links page before posting:
    http://groups.yahoo.com/group/solarisx86/links
Post message: solarisx86 <at> yahoogroups.com
UNSUBSCRIBE:  solarisx86-unsubscribe <at> yahoogroups.com
Permalink | Reply | 3 comments |
headers
Mike Riley | 1 Nov 15:53
Picon

Re: Solaris 10 jumpstart fails on fork failed too many porcesses

Greg wrote:
> On a Dell 650, after completing "install_begin",
> I get an error like:
> 
> /usr/sbin/ parse_dynamic_clustertoc: fork failed too many processes 
> 
> Has anybody encountered this and what is the likely patch to solve?

I have not seen that in a very long time.  When I did see it the cause was a 
process being forked over and over again until some limit had been reached. 
  Back then it was usually init that was doing the forking and usually 
because of a problem with the console open failing.

Not sure if that helps.

Mike

Please check the Links page before posting:
    http://groups.yahoo.com/group/solarisx86/links
Post message: solarisx86 <at> yahoogroups.com
UNSUBSCRIBE:  solarisx86-unsubscribe <at> yahoogroups.com
Permalink | Reply | 1 comment |
headers
Greg | 1 Nov 16:36
Picon
Favicon

Re: Solaris 10 jumpstart fails on fork failed too many porcesses


Hi Mike-
Actually, that does help. It would explain other things I'm seeing.
cheers,
Greg
--- In solarisx86 <at> yahoogroups.com, Mike Riley <michael.riley@...> wrote:
>
> Greg wrote:
> > On a Dell 650, after completing "install_begin",
> > I get an error like:
> > 
> > /usr/sbin/ parse_dynamic_clustertoc: fork failed too many processes 
> > 
> > Has anybody encountered this and what is the likely patch to solve?
> 
> I have not seen that in a very long time.  When I did see it the
cause was a 
> process being forked over and over again until some limit had been
reached. 
>   Back then it was usually init that was doing the forking and usually 
> because of a problem with the console open failing.
> 
> Not sure if that helps.
> 
> Mike
>

Please check the Links page before posting:
    http://groups.yahoo.com/group/solarisx86/links
Post message: solarisx86 <at> yahoogroups.com
(Continue reading)

Permalink | Reply | 0 comments |
headers
Mike Riley | 1 Nov 18:29
Picon

New Storage Projects for Open Solaris

These are currently hidden, but wanted to note them for people to watch that 
might be interested:

CIFS Server
http://www.opensolaris.org/os/project/cifs-server

Automatic Data Migration
http://www.opensolaris.org/os/project/adm

Media Management System
http://www.opensolaris.org/os/project/mms

The one that I know something about is the CIFS server, which is targeted to 
replace Samba functionality.  Phase I just went into Nevada this past week.

Mike

Please check the Links page before posting:
    http://groups.yahoo.com/group/solarisx86/links
Post message: solarisx86 <at> yahoogroups.com
UNSUBSCRIBE:  solarisx86-unsubscribe <at> yahoogroups.com
Permalink | Reply | 0 comments |
headers
sol11x86 | 1 Nov 19:21
Picon

Re: Laptop for Sun Solaris OS [need suggation]


 -------------- Original message ----------------------
From: Eric Bruno <eric <at> ebruno.org>
> Al Hopper wrote:
> >
> > On Thu, 25 Oct 2007, John D Groenveld wrote:
> >
> > > In message <867761.31045.qm <at> web44801.mail.sp1.yahoo.com 
> > <mailto:867761.31045.qm%40web44801.mail.sp1.yahoo.com>>, Vishal Gupta 
> > writes:
> > >> Will Sun Solaris 10 Support Intel Core 2 Duo Processor?
> > >
> > > If not Solaris 10, then Solaris Express should.
> > >
> > > Runs fine on my Core 2 Duo based laptop.
> > > <URL:http://www.sun.com/bigadmin/hcl/data/systems/details/2777.html 
> > <http://www.sun.com/bigadmin/hcl/data/systems/details/2777.html>>
> > >
> > > Except for the missing support for the Intel 4965AGN which is
> > > apparently coming RSN.
> > > <URL:http://www.opensolaris.org/jive/post 
> > <http://www.opensolaris.org/jive/post>!reply.jspa?messageID=165468>
> > >
> > > Would it be helpful if the HCL would allow you to query laptops
> > > by processor (or screen size or graphics)?
> >
> > Yes - definately.
> >
> > I know this comment is a little late for this thread, but, it was
> > announced at the OpenSolaris Summit, that Sun was actively evaluating
(Continue reading)

Permalink | Reply | 0 comments |
headers
John Martin | 1 Nov 22:27
Picon

Re: xVM platform supports HVM

Apostolos Syropoulos wrote:
> Today I played a little bit with Xen. I created an hvn file
> (WINXP.hvm), which is almost identical to the one presented at
>
> http://www.opensolaris.org/os/community/xen/docs/HVMdomains.htm
>
> Almost identical because the disks section is the following:
>
> disk = [ 'file:/usr/local/tmp/winxp.img,hdc,w', 
>          'file:/usr/local/tmp/diskD.img,hdd,w' ]
>
> File winxp.img is a virtual disk created with qemu that
> worked just fine with qemu. Now,  I entered the following
> command 
>
> xm create WINXP.hvm 
>
> and after a while a window with Windows XP showed up. 
> The average memory consumption was 1,8 GiB (i.e., 97.6%)
> and the average CPU load was around 40-50% for the first 
> core and 80% for the second core. After it installed
> a number of drives, I rebooted Windows and I left for a while.
> When I returned my system was dead! I had to reboot it and to
> do fsck etc. In other words, my first experience with Xen was 
> more than disappointing. It is slow, really slow, and compared
> to qemu it does not show any real advantage. So why all this
> fuss about Xen and processor virtualization? 
I asked a xVM expert about this.  His reply:

  He doesn't mention if he's using the old build 66 based
(Continue reading)

Permalink | Reply | 0 comments |
headers
Tony Reeves | 1 Nov 23:16
Picon

Re: xVM platform supports HVM

On 11/1/07, Apostolos Syropoulos <asyropoulos <at> yahoo.com> wrote:
>
> Today I played a little bit with Xen. I created an hvn file
> (WINXP.hvm), which is almost identical to the one presented at
>
> http://www.opensolaris.org/os/community/xen/docs/HVMdomains.htm
>
> Almost identical because the disks section is the following:
>
> disk = [ 'file:/usr/local/tmp/winxp.img,hdc,w',
>         'file:/usr/local/tmp/diskD.img,hdd,w' ]
>
> File winxp.img is a virtual disk created with qemu that
> worked just fine with qemu. Now,  I entered the following
> command
>
> xm create WINXP.hvm
>
> and after a while a window with Windows XP showed up.
> The average memory consumption was 1,8 GiB (i.e., 97.6%)
> and the average CPU load was around 40-50% for the first
> core and 80% for the second core. After it installed
> a number of drives, I rebooted Windows and I left for a while.
> When I returned my system was dead! I had to reboot it and to
> do fsck etc. In other words, my first experience with Xen was
> more than disappointing. It is slow, really slow, and compared
> to qemu it does not show any real advantage. So why all this
> fuss about Xen and processor virtualization?
>
(Continue reading)

Permalink | Reply | 0 comments |

Gmane