Next problem was that the ethernet device wasn't working. I downloaded the
vfe driver code and go
t that resolved. The rest of the setup went like a dream. It took most of a
day, but he left impre
ssed. Once he's ready I'm going to help setup the real machine that he just
can clone from.
I am looking for information about Trusted Solaris Device Drivers.
Specifically, I need to know i f a trusted driver is different from a
standard Solaris driver in terms of Sun's DDI/DDK interfaces.
I have tried posting that sort of question on the Sun Develoepr
Network but got no replies. We
[EMAIL PROTECTED] writes:
On Mon, Sep 18, 2006 at 08:56:20AM -0400, James Carlson wrote:
What are you using on your Linux systems?
Mainly samba (no LDAP). So I need to move all users accounts (passwords)
to new server.
That wasn't the question. In context, the original question was
Now if I copy /etc/(shadow|passwd) from linux to solaris is it
going to work without any problems ?
You don't have to do anything, really.
The /etc/security/policy.conf file only governs the behaviour for newly
encrypted passwords; existing encrypted passwords will be correctly
interpreted,
unfortunately the marvell drivers only work in 32bit mode. :( the installer
exited with an error
on my 64bit install.
Get the skge drivers from SysKonnect and add the appropriate aliases.
(They ship a 32 bit and 64 bit package rather than a combined one, so you
have to do some stuff by
Daniel, thank you very much for this helpful information.
I really thought that files under /proc are files that can be
accessed like normal files - OK, mmap won't work, but that
is not a big deal. Am I the only one that considers
this to be a bug? This is really weird behavior. Looks like I am
So status is readable with the given size, which should be at least 1232.
This works for all pr
ocesses except this one. Any idea?
cat /proc/700/status should work when you use ksh93
No, it will not as long as ksh93 is a 32 bit process:
2656: open64(/proc/2630/status, O_RDONLY)
Felix == Felix Schulte [EMAIL PROTECTED] writes:
Felix On 9/17/06, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
No, it will not as long as ksh93 is a 32 bit process:
Felix If you don't believe me install
Felix
http://www.opensolaris.org/os/project/ksh93-integration/downloads/2006-09-12/
Felix
As I told you: Your system is broken, it works on mine... :)
If it works (cat is allowed to read status files of 64 bit processes),
then it's actually broken.
How many bytes of status are returned?
Casper
___
opensolaris-discuss mailing list
This is the first step of a long (upcoming) journey of Dell's AMD-powered PCs:
http://www.dell.com/content/products/productdetails.aspx/dimen_c521?c=uscs=19l=ens=dhs
Perhaps someone at Sun will try to sell Mr. Michael the vision of
Solaris-based business desktop
s?
But Solaris still doesn't
We run an OpenSuSE/OpenSolaris/OpenOffice study group here in Honolulu and
have a dozen or so all
sorts of AMD-based PCs in our lab. To the best of my memory, all our
Athlon64-based motherboards c
ome with on-board gigabit Marvell NICs. Solaris HCL does show quite a number
of on-board
Did you check the SysKonnect website? They're the company
that actually provides the drivers (SK*) which are presently
integrated with Solaris.
Except those are *really* old.
I thought even Marvell now had the 64 bit driver but I checked
the package and they don't seem to have one which is
I didn't follow the thread, only the most recent message.
However, maybe the open NIC drivers are of help (at least some flavours
of Marvell NICs should be supported).
All Marvell Yukon NICs that I know of appear to be supported by the
skge drivers from SysKonnect.
However, there appears to
Moinak has well summarized a few points, that I did not express so good
(because I took them for g
ranted, for example: That only C++ code is affected, and only when compiled
with SUNWspro).
His summary is a useful addition to this thread.
It is somewhat strange that the runtime.libraries file
That's another thing that is annoying about Solaris - the fact that for some
cards the drivers are
there all along, but because Solaris doesn't have the correct driveralias
entries it won't attach
to the card out of the box.
But there's no way that can be helped; I'm sure Linux suffers a
SUNW breaks the law, because they themselves offer the Distro-JDK for
download and are therefore themselves Distro-JDK distributors:
No, Sun is not bound by its own redistribution licenses so it does not
break the law. It owns the JDK so it can distributes it as it sees
fit.
Casper
On ieee1275 / OBP (i.e. on sparc) you can fake those id's and therefore
avoid a kernel-recompile.
(using nvedit and a script etc.)
On x86 systems, you can probably reprogram the vendor/device ids
(they are programmable in the chips)
But you would only need to do so if the device driver in
[EMAIL PROTECTED] wrote:
SUNW breaks the law, because they themselves offer the Distro-JDK for
download and are therefore themselves Distro-JDK distributors:
No, Sun is not bound by its own redistribution licenses so it does not
break the law. It owns the JDK so it can distributes it as
If the opsol libc is not redistributable how do the actual existing opsol
distros
do it? What is Belenix doing? What is Schillix using?
Did they port the gnu libc to OpenSolaris? Did they port the BSD libc to opsol?
libC not libc.
Casper
___
a boilerplate response could have been made, of the form
Inert name(s) [is |are] entwined with encumbered code, and certain
agreements are in place that
would be repudiated if Sun were to discuss any instant details in a public
forum
We're not allowed to name and shame.
So that's all we can
Thanks for the detailed info.
If v9 infact represents 64-bit then why am I getting 32bit/64bit compilation
conflicts. MySQL conf
ig shows that it is compiled with the v9 flag and from what I understand by
default the compiler wi
ll compile in 64bit (the native architecture, right?)
Default
Thus :
http://polaris.blastwave.org/browser/on/trunk/usr/src/cmd/ssh/ssh/ssh.c
First thing I notice :
* Author: Tatu Ylonen [EMAIL PROTECTED]
* Copyright (c) 1995 Tatu Ylonen [EMAIL PROTECTED], Espoo, Finland
*All rights reserved
And then this :
/*
* Copyright
Some implementations can be encumbered (using arcfour
was difficult
and there was an implied implemnetation key length
limit which we are
now trying to lift)
I'm interested in the limitations of the key lenghts. Is this some sort of
old ITAR imposed limit on exports? Not that it makes
On Tuesday 05 September 2006 05:58 am, [EMAIL PROTECTED] wrote:
I mean, the question is interesting. Can I, for example, port a Linux
kernel driver that is GPL to Solaris, and legally run it ?
I hope to eventually get a positive answer to that ... but not here and
now; back to work :)
Frank,
On Wed, 2006-09-06 at 20:39 +0200, [EMAIL PROTECTED] wrote:
On Tuesday 05 September 2006 05:58 am, [EMAIL PROTECTED] wrote:
I mean, the question is interesting. Can I, for example, port a Linux
kernel driver that is GPL to Solaris, and legally run it ?
I hope to eventually get a positive
[EMAIL PROTECTED] wrote:
On Wed, 2006-09-06 at 20:39 +0200, [EMAIL PROTECTED] wrote:
On Tuesday 05 September 2006 05:58 am, [EMAIL PROTECTED] wrote:
I mean, the question is interesting. Can I, for example, port a Linux
kernel driver that is GPL to Solaris, and legally run it ?
I hope to
DEPDIR=.deps depmode=none /bin/bash ../../../config/depcomp \
/opt/SUNWspro/bin/cc -DHAVE_CONFIG_H -I. -I. -I../../../config
-I/usr/local/mysql/include -xO3 -m
t -D_FORTEC_ -xarch=v9 -xc99=none -I../.. -D_REENTRANT -D_XOPEN_SOURCE=500
-D_GNU_SOURCE -D_REENTR
ANT -g -c -o
Still missing the aes256-cbc ciphers etc etc for various reasons.
Mostly government export controls, not technical.
Import controls but probably not relevant anymore (we ship all the 256 bit
ciphers in a download available to almost everyone)
Auditing ?
Please explain. Do you mean the
You mean its possible to login to a Solaris 10 server by some method and
completely slip under the radar of the audit processes ? Wow .. thats
just totally scary.
If you install OpenSSH or otherwise modify the system entry
points to bypass auditing, yes... Perhaps we should fail setuid if
Import controls but probably not relevant anymore (we ship all
the 256 bit ciphers in a download available to almost everyone)
Is it in the shipping product ?
The crypto packages; no support for OpenSSH but there is support for
openssl.
Casper
___
So you are saying that its possible for a user to do one of the following :
(1) downlod OpenSH source .. build it themselves
(2) get it from SunFreeware
(3) get it from Blastwave
and then access the system in a way that slips under the radar of the BSM
modules? The user can do whatever they
On Tue, 5 Sep 2006, [EMAIL PROTECTED] wrote:
Still missing the aes256-cbc ciphers etc etc for various reasons.
Mostly government export controls, not technical.
Import controls but probably not relevant anymore (we ship all the 256 bit
ciphers in a download available to almost everyone)
Based on the recommendations, I finally installed cc and now I am getting the
following errors whe
n trying to compile sysbench:
You're using this as a C compiler:
checking for gcc... (cached) /opt/SUNWspro/bin/CC
It's a C++ compiler; anyone programming C will deny there's a relationship
You're using this as a C compiler:
checking for gcc... (cached) /opt/SUNWspro/bin/CC
It's a C++ compiler; anyone programming C will deny there's a relationship
between the languages.
Use /opt/SUNWspro/bin/cc.
Even with that I get compile errors when running make:
ANT -g -c -o
former sun employee, member of the opensolaris pilot project they are
using to give the appearance as having inside information that makes
the CDDL and the people behind it look evil.
I know; but *who* is she?
It is interesting how people quote the FSF on the CDDL; they invariably
neglect to
I think a very clear written one to two page rebuttle with absolute
proof or very clear logic behi nd the choice of CDDL released without
much fanfare would be the best approach to extinguishing thi s current
fire, anything more and you're just going to be pouring petrol.
It would be nice if
Essentially take anything from Solaris 8 and run it on Solaris 9 or 10
or Solaris Express or Solaris Express Community Release and just don't
expect kernel modules or drivers to all just work but almost
everything else will.
Uhm, kernel modules and drivers, if DDI/DDK compliant, should also
just
[btw. my idea of hacking intp.c to check for $OS_PERSONALITY isn't as
easy as I'd thought - we're in kernel space at that stage I thnk (still
reading my new copy of the 2nd ed. of the internals book ;-)]
Might be easier to make it into a syscall then.
At exec time we've copied the new
http://www.gnu.org/philosophy/free-sw.html
http://www.fsf.org/licensing/licenses/index_html#GPLIncompatibleLicenses
Common Development and Distribution License (CDDL)
This is a free software license which is not a strong copyleft; it has
some complex restrictio
ns that make it
My concern is not so much the way that the FSF looks at the license even, my
concern is that pockets of the open source community are not accepting our
license. Maybe my concerns are invalid, but if any licensees of CDDL are
facing any type of problem, it can't be good.
I don't see that as
Alan DuBoff [EMAIL PROTECTED] wrote:
On Wednesday 09 August 2006 01:43 am, [EMAIL PROTECTED] wrote:
They use pretty much the same jabs for the Apache 2 license.
Understood, as they do MPL, and others.
But the MPL is unacceptable by private persons as it request a vevue in Santa
Clara CA.
[EMAIL PROTECTED] wrote:
If this is the case, Sun should back it up and make sure that it does stand
on
it's own, and get a statement from the GPL folks (or Debian) that they do
accept it as an open source free license.
Why? OSI is what matters and OSI has made CDDL one of 9 preferred
[Some explanation: for use in a lock-free queue implementation I need a object
'pool' where freed
objects remain available in the pool (as they may still be accessed by other
threads after one thre
ad has freed them) and I was hoping to build this facility using a
umem_cache...]
You don't
I must say that as a previous Debian user for quite a number of years,
Nexenta really kicks! Even though I haven't run a Debian server for
close to 3 years, old habits are hard to break as they say ! There's a
few things I don't have at the moment due to lack of pkg support, but
hopefully that wi
echo `/usr/ucb/whoami` blah/blah/blah/blah.log
Since I figure this should print out the EUID running the script,
which I expect to be root. Well, when run, the log shows the user I
assigned the rights profile to, NOT root as I expected (or rather hoped
for).
Was this a bourne shell script? It
#!/bin/csh
Refuses to run if real and effective uids don't match.
#!/bin/csh -b
Runs set-uid scripts just tine
Casper
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
I have certainly wondered the same.
I know that, at least for some time in the late '90's, that several versions
of
linux shipped with some form of Openwin/open look + OLVWM, so it must be open
to
some extent.
We threw some bits over the fence at some time.
Casper
But why is SUNW so uninterested???
Where is your vision of ^^We strongly believe in One Solaris^^ now?
There is a lot of history there and it is hard to change.
I'm sure they don't want to ship on Xorg on SPARC which only supports
older framebuffers poorly so a lot more work is involved.
Are
(ks93 discuss removed)
But why is SUNW so uninterested???
Where is your vision of ^^We strongly believe in One Solaris^^ now?
There is a lot of history there and it is hard to change.
Migrating from Xsun to Xorg did work on x86.
What lot of history do you mean exactly?
(okay, /dev/fb is
Except NICs you mean?
Including NICs.
Doesn't Grub (at least Grub1) use the etherboot/rom-o-matic NIC drivers?
Not for Solaris as Sun ships it.
The grub that comes with Solaris is loaded in memory using PXE (over the
wire) and then the Grub PXE driver continues to use PXE to bootstrap the
okay.
If (and only if) 3rd parties are involved into the libC* thing, I _would_
understand it.
What I still would not understand - however - is, why the Distros-JDK (on
which SUNW has made so m
uch noise about, back in May'06) has not been built with the open gcc.
A gcc-built JDK would not
I suggest to make /bin/ksh ksh93 from the beginning that you don't
have to deal with any backwards compatibility fuzz later
If you want PowerPC Solaris and SPARC/x86 Solaris be that different,
then yes; if you want scripts to be compatible I suggest not.
(+1 for PowerPC)
Casper
On 7/27/06, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
I suggest to make /bin/ksh ksh93 from the beginning that you don't
have to deal with any backwards compatibility fuzz later
If you want PowerPC Solaris and SPARC/x86 Solaris be that different,
then yes; if you want scripts to be
The issue of backwards compatibility is already addressed very well in
ksh93 itself. Most of the opensolaris distributions - excluding
Solaris itself - are shipping ksh93 as /bin/ksh or are going to ship
it. The ksh integration tree contains a master built switch
specifically for that purpose:
On 7/27/06, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
PS: Somehow I have the feeling that Sun doesn't want to see the
project succeed in replacing ksh88 with ksh93, a feeling which is
based on the open hostilities from Sun personnel and the permanent
delays :(
No, that's not true at all.
Solaris 10 SPARC is currently incompatible with Solaris x86/x64: Xsun
on SPARC vs. Xorg on x86/x64. I haven't heard of too many complaints
because of this compatibility breakage (source code written, compiled
and linked on Solaris 10 Xorg x86/x64 will not compile and link on
Solaris 10 Xsun SPARC
i am specifically referring to:
Xrender
XVideo
XvMC
XRandR
Xcomposite
Certainly the compilation environment should support all; Xrandr
seems to be present on SPARC (certainly the library and client
are).
All of them should compile and run (against Xorg servers) on SPARC.
none of these
this backwards compatibility for backwards compatibility's sake is no
longer a selling point. Linux has proven that backwards compatibility
for its own sake is largely irrelevant (my personal unhappiness about
this incompatibility, grounded in purely philosophical rather than
practical
+1.
A small question, why not ON11? Does it has anything different? Just
curious. :-)
Same reason it's not ON-NV; life goes on after Solaris Nevada.
I suggest the community be named OS-NET; ONNV has the NV disambiguator
which ON lacks.
Casper
___
swap 149192401104 14918136 1%/etc/svc/vol=
atile
swap 15365960 447832 14918128 3%/tmp
swap 14918208 80 14918128 1%/var/run
What we are seeing here is that df reports different sizes for swap, =
which actually
Somehow you are right. But blocks in use accounts against swap, and
swap is used by all those mount points.
Except that it isn't really swap but available virtual memory.
Finally, what is most important to me, the first number (i.e. swap
total) looks broken to me, as a result of the way
Hi all,
I just wanted to understand that Is there any difference in behaviour of
Solaris 10 which has been
upgraded from solaris 9 and a freshly installed Solaris 10?
In one of my customer locations there are two nodes one of them is freshly
installed and other one
is upgraded from Solaris
With SXCR b43 just announced are we able to do a live upgrade from snv_38 to
snv_43 ?
The link for List of known issues at
http://www.opensolaris.org/os/downloads/on/ suggests thats its still, well,
you know, a known issue.
That issue was fixed in b38.
Casper
I see lots of complaints from people about install processes for
Linux, Solaris, BSD, etc. -- but not too many people that are willing
to admit that the Windows install process isn't a very aesthe tic
experience either.
That's likely because few people ever run a barebones windows install.
And
True, but it's still one of those things that bugs me. The recovery
CDs are of course especially evil, since people really like the
polished expiernce they provide, but want it to work for all hardware,
not realising that in many cases it was made specifically for their
system configuration!
Wireless USB finally coming out of the labs
http://eetimes.com/news/latest/showArticle.jhtml?articleID=189600587pgno=1
Interesting, but I thought one of the reasons to have cables was
power?
(Why is Wireless USB needed when you have bluetooth?
Will we have all the bluetooth implementation
I'm a little confused as to what behaviour people want with
PageUp/PageDown in gnome-terminal.
I'm running snv_42.
If I run vim inside a gnome-terminal then PageUp/PageDown moves me
through the document as I'd expect and want.
If I'm at the shell prompt then I need to press Shift PageUp to
Scrolling around in gnome-terminal itself without having
to press shift. I want the terminal to get PgUp, PgDn,
Home, and End. I don't have any applications running inside
a terminal window that I would ever want to send those
events to, and it's twice the effort (at least) to have
to press shift
Could be. At least, I ran frkit, installed acpidrv, and the problem
vanished. Haven't even rebooted yet.
Yes, most probably it then.
The driver does attach immediately when it is installed;
batstat -t will show you the thermal zones configured in the
system.
If there was a thermal interrupt it
Standards you can never have too many :-)
Will we have all the bluetooth implementation security issues all
over again?)
Very likely all those with variations and a bunch of new ones
One good thing, though, is that is uses the same spectrum everywhere.
I have always wondered about Wifi; it
BTW: thermal zone handling was introduced in Nevada build 40 as part of:
6363985 acpica: Metropolis SMB Alerts result in high background system load
Metropolis, of course, is the W2100z
(The thermal event can happen in overtemp conditions and some
other conditions; if the _TMP method is
Why don't we make a general security framework to support all kinds of
devices? Those spec define the similiar functions, such as
authentication, encryption, etc. The abstract layer will reduce much
time for developing similiar protocol.
Because we're limited to the protocol defined for a
On the other hand I never use the pageup/pagedown on the numeric
keypad because all those keys exist elsewhere.
I use them all the time.
I don't even really understand why a full size keyboard like
a Sun type 6/7 even needs the different states.
On MacOS X there is no way to ever change the
[EMAIL PROTECTED] wrote:
Surely changing it strikes me as a wanton change; making it configurable
is fine with me; changing the default is not. Especially since PCs have
a firmware setting which can be controlled; Solaris should not override
that firmware setting.
I'm not asserting we
Or would it perhaps make more sense to have the install program(s) do a
better job of processing input? For the most part, there are very few
valid keys for each input. Instead of echo'ing garbage to the screen,
why not print a message, or beep, or do nothing
Yes; the install program
More than that, it might be desirable to have (for this, but more broadly too)
instructions on how to disable all wireless functionality for those who might
wish
to ensure that they were not inadvertantly allowing it. That applies both to
software
(what drivers to remove; how to make sure they
How did you (generic) install Solaris/NexentaOS (/or BeneliX/SchilliX) in a
dual-booting config
uration? Perferably in an Athlon64 whitebox/laptop.
I believe the simplest way of doing this is:
- create a single Solaris Fdisk partition
- make several different UFS boot
I don't think I'm the first one on the quest for the holy grail, eh, I mean -
a flexible booting s
ystem for your PC's.
Anyway, has anyone set up a nice GRUB menu.lst file for a network booting
(pxegrub) that allows
the users to select between:
1. Boot the installed operating system on
Funny thing, but I always believed tha r=$((3 +1)) notation is for Korn shell
only.
I was codified in the X/Open Unix specifications.
Casper
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
I have several standard-compliant shell scripts.
These scripts use
#!/bin/sh'. However, since /bin/sh on Solaris is not
standard-compliant, these scripts fail.
How very interesting. Would you please list what exactly is it that fails,
best with the snippets
of code that fail?
But
Shawn Walker wrote:
POSIX requires that you get a POSIX shell if you type sh after you
did setup a POSIX compliant PATH
So, there's nothing disturbing about this at all.
Shawn, I would appreciate an advice on the following situation then:
I'm trying to port a software called Asterisk to
Hi thanks for the reply. But what I really want is to derive the very
next IP that the DHCP server
is going to allocate to the next host that sends in a request. This
operation will be performed on the server running the DHCP daemon
itself. Therefore, I cannot do a release / request because the
I am not sure I understand you correctly. Are you suggesting that I
should avoid #! and use explicit invocation of shell interpreter to run
scripts?
If you want to have a bigger chance of executing the script in a
POSIX shell.
This is the only way the standard provides for executing scripts:
if [ -x /usr/xpg4/bin/sh -a $_ != /usr/xpg4/bin/sh ]; then
exec /usr/xpg4/bin/sh $0 $*
Certainly you will need:
exec /usr/xpg4/bin/sh $0 $@
But you can't test on $_ in shell scripts.
Casper
___
opensolaris-discuss mailing list
[EMAIL PROTECTED] wrote:
I can't be sure what make does, but if a script is executable and starts
with : or even just # the shell will fail the exec and will run
the script with $SHELL script.
Indeed, this is one solution! If I remove #! from all scripts,
determine the XPG4-compliant shell
Here's the situation. The software that I've mentioned contains several
scripts to build itself. Those scripts are XPG4 compliant and they
have '#!/bin/sh' as their first line. The scripts use some features
that Solaris's /bin/sh does not grok.
You could make the first line of the
Thanks, Darren. You have seconds. I'll contact you offline to get you set up.
I've not heard any argment why this is different from
visual panels and I would consider this to still be in the
discussing phase.
Two people mentioned visual panels and there was no answer.
Casper
On Wed, 28 Jun 2006, [EMAIL PROTECTED] wrote:
Thanks, Darren. You have seconds. I'll contact you offline to get you set
up.
I've not heard any argment why this is different from
visual panels and I would consider this to still be in the
discussing phase.
Two people mentioned visual
i just did my first nightly build on the 20060616 code-- took just over 5hrs
on my ferrari 4006--
and now i see the 06/06 release of solaris 10 up for download.
so my noobish question is: is there any reason for me to re-install using that
CD/DVD set? i figu
re one reason would be ZFS root
thanks casper! i started with an 05/06 nevada DVD set (snv_38) and updated
that to snv_41 when i
built from the 20060619 sources. i'm downloading the JDS vermillion_43
packages at the moment, so
hopefully with those i'll be as up to date as the nevada version of S10 06/06.
There's no
On 6/27/06, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
thanks casper! i started with an 05/06 nevada DVD set (snv_38) and updated
that to snv_41 when
i
built from the 20060619 sources. i'm downloading the JDS vermillion_43
packages at the moment,
so
hopefully with those i'll be as up
It seems that this is already covered bby the Visual Panels project;
I do not believe we should have two competing projects except in cases
where there are very compelling reasons.
But Visual Panels which is perhaps hindered by its name and this
project seems to overlap fully; and Visual Panels
Hi,
Was browsing through the manpage for ssh_config(4), and this section appears
to me to be somewhat
confusing (or at least contradictory):
[i] UsePrivilegedPort
Specifies whether to use a privileged port for outgoing
connections. The argument must be yes or no.
Casper hit me with my HP calculator .. thus
I think you meant 0.850mm (850 micron), not 850cm :-)
Worse.
bash-3.00$ bc
scale=9
299792458000-- c cm/sec
mm/sec
1*10^9
10
299792458000/10
299.792458000 -- 299 cm at 1 GHz
299.792458000/100
2.997924580 -- 2.99
Edit the path to the kernel that GRUB uses, and remove the amd64 part.
By default, no path is given and it defaults to the best; so
you need to specify kernel/unix on the multiboot line.
Casper
___
opensolaris-discuss mailing list
Steven Sim wrote:
Hello;
I frequently conduct the following operation;
#ufsdump 0f - unmounted ufs raw device | (cd /mnt; ufsrestore xf -)
I have noticed that the above takes place so much slower in single user
mode than in multi user mode.
A newfs operation on a local block device
I've just (live) upgraded from build 36 to build 41 and I have a couple
of observations:
The nVidia driver in the source BE still gets mucked up.
The machine is now a brick. Must be a ZFS change, I see
cannot mount '/export/home1': directory is not empty
No, this is a liveupgrade bug; it
The system concerned is a SunFire 240 with 2 x CPU 1050 Mhz, 2 Gbyte RAM
and 4 internal SCSI disk (c1t0d0,c1t1d0,c1t2d0,c1t3d0)
It's a lab machine.
I booted Solaris 8 2/04 off CDROM and got to single user mode.
Ah, Solaris 8 (what are you doing here, running such old stuff :-)
You'll find
Sorry.. these BFUs,
http://dlc.sun.com/osol/on/downloads/current/on-bfu-20060612.i386.tar.bz2
if I upgrade to this one, does that mean that it'll be all done
automatically for me?
No; because you can't upgrade to bfus and the default install
only applies to *fresh installs*, not upgrades.
Sorry if there's a better place to ask, but is
anyone using Nevada b40 on an x4100?
I'm seeing very long boot times (6+ minutes) which seem to be down
to slow SCSI bus detection, and I can't see the console anymore if I ssh
to the ILOM (although the KVM java app works fine).
This is a BIOS
801 - 900 of 1175 matches
Mail list logo
