Actually it works quite well without the changes to "routeadm" suggested
to you.
motie:/home/lw ->pfexec routeadm
Configuration Current Current
Option ConfigurationSystem State
-
Vikash Tulsiyan wrote, On 08/17/09 04:57:
I am writing a script that allows any user(with appropriate permissions) to
install packages and start/stop some services on opensolaris. As far as i know
only Primary Administrator has the priviledge to perform these task. Is my
understanding corre
Try a more qualified string like this if you are running 2009.06
pfexec pkg install pkg:/sunstudio1...@12.1-0.111
Jerome Verstrynge wrote, On 09/01/09 13:58:
Hi,
I am trying to install sunstudio12u1 using:
pfexec pkg install sunstudio12u1
But I get:
Creating Plan -
pkg: the following p
ire:
Pkg SUNWj6dmo: Optional min_version: 0.5.11,5.11-0.101
max version: 0.5.11,5.11-0.101 defined by: pkg:/entire
JVerstry
Leslie H Wood wrote:
Try a more qualified string like this if you are running 2009.06
pfexec pkg install pkg:/sunstudio1...@12.1-0.111
Jerome Verstry
ject).
I am not an expert with system administration, though I have learned a
lot in the past couple of days.
Can you tell me how I can update to the stable 2009.06 release? Or
should I go through Henry Jen?
Thanks,
JVerstry
Leslie H Wood wrote:
Sorry, I should have read the error message pr
For remote use
/usr/dt/bin/ttsession -c dtpad
Naveen sursietty wrote, On 09/11/09 21:18:
I am facing an interesting issue with dtpad. I have NIS ( not NIS+) setup in my LAB.
All machines are installed with Solaris 10. After logging on a particular machine locally. when i tried to open dtpad aft
There is a GUI under System/Administration/Network
or
Make sure there is a correctly populated /etc/resolv.conf for DNS
svcadm disable svc:/network/physical:nwam# Turn off DHCP
svcadm enable svc:/network/physical:default# Turn on static
motie:/home/lxxx ->ls -
Whwnever you change the nsswitch.conf you need to restart the name
service cache to reread the nsswitch.conf.
svcadm restart system/name-service-cache:default
The /etc/resolv.conf should NOT include your router and "man
resolv.conf" explains how it should be populated.
domain my.dot.com
na
Hey vegablue,
I got Eclipse downloaded, and it works great. However, it is in my tmp folder.
One minor issue, in Solaris/OpenSolaris the /tmp directory is wiped at
boot time. So if you want to keep your Eclipse stuff, you will have to
move it.
Without knowing what form the Eclipse data
keithk wrote, On 10/01/09 18:07:
Hi:
I am a newbie with zfs, so I apologize if I did no look deep enough before
posting. I am trying to create a zfs filesystem so I can take snapshots of the
file system. So, I create a file system like this:
rpool/export/vm 44.7G 744G 19.7
It's late and I typoed, that rpool/export/share example should have been
install
zfs set sharenfs='ro,anon=0' rpool/export/install
Leslie H Wood wrote, On 10/01/09 18:51:
keithk wrote, On 10/01/09 18:07:
Hi:
I am a newbie with zfs, so I apologize if I did no look deep
I see this issue all the time but I assumed it is because I have
multiple entries. I am running 2009.06 with SRU5 and the icon always
tells me that "Updates are Available".
motie:/home/lw31248 ->pkg publisher
PUBLISHER TYPE STATUS URI
ipkg-support
http://opensolaris.org/os/discussions/
I see there is a printing discussion alias at the above which may be
able to help and a forum.
http://www.opensolaris.org/jive/forum.jspa?forumID=19
Nibal wrote, On 10/13/09 12:25:
Dear All,
I have sun Solaris 10 server adding all shared
printers as LPR
Nick Andrade wrote, On 10/14/09 12:30:
I completely agree, the download page is VERY UNCLEAR. It assumes way too
much, especially considering all the new tech involved (e.g. I only found out
about Auto Install yesterday after talking to a Sun architect). At the very
least, make it known th
man pingis the RTFM :)
Try ping -s server_name
That Guy wrote, On 10/14/09 15:33:
I'm really, really sorry if this is a "RTFM" situation, as I can't figure out
where to look this up.
In Linux, the name your give the computer shows up on the LAN. e.g. if I call my computer
I hit this problem and I found that some of my firefox
"~/.mozilla/firefox" files have somehow been changed to root
ownership.Changing them back to my userid and group fixed the problem.
awr wrote, On 10/20/09 07:51:
Thanks
The solutions provided by Mozilla:
1) Exit other programs: There
It is in the user directory in "~/.ssh" so for OpenSolaris root user
it will be in:
/root/.ssh
Also check "man ssh"
Will wrote, On 10/27/09 19:33:
I just intsalled amanda, but it is unable to ssh into the system. I need to
make sure that amandabackup is in the ssh authorization file but i
It is so easy since you just add more of the same type to the pool.
For raidz stuff they actually recommend no more than 9 disks in each
raidz for best performance so you would just add multiple sets of raidz
after the first raidz is "created". Check BigAdmin zpool tips.
zpool add pool
Even Solaris has free access to security patches. I assume the problem
is that the easier SRU mechanism for updates bundles both security and
non security updates together. Alan Coopersmith's suggestion looks
like the only recourse.
Tomas Bodzar wrote, On 11/17/09 14:28:
If you want real
You have a typo for "Naming Service Switch"
pfexec cp /etc/nsswitch.conf /etc/nsswitch.conf
svcadm enable svc:/network/dns/client:default
svcadm restart system/name-service-cache:default
I assume you have an appropriate "/etc/resolv.conf" file in place.
evelin wrote, On 11/24/09 23:10:
h
host:/home/test ->pkg list openoffice
NAME (PUBLISHER) VERSION STATE
UFIX
openoffice (opensolaris-ipkg) 3.1.0-0.111 installed
"man pkg" since there are other useful options
Vikash Tulsiyan wrote, On 12/03/09 12:39:
Do
You need MANPATH setup to really use the man pages effectively. They
are never setup by default in Solaris or OpenSolaris.
1. Want to do "man -k searches?
Setup the root MANPATH to include the man directories you want. You
get one by default in OpenSolaris but you can add to it.
r.
Title: Adding USB disk - appears but doesn't appear
Try "rmformat -l"
...Les
Edward Ned Harvey wrote, On 12/10/09 13:29:
I added a USB disk.
If I do “cfgadm -al” it appears as “usb-storage connected
configured ok”
If I do “format” it doesn’t appear at all.
If I do “devfsadm -Cv” i
USB devices do automatically mount these days. I just added a USB stick
and "df" shows it did mount.
desk1:/home/luser ->df -k
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/dsk/c5t0d0p0 499480 728498752 1% /media/NO NAME
Probably due to this servi
I updated from 2009.06 SRU7 to build 129 and my 6 workspaces dropped to
4. Therw is nothing in the preferences option to change this and no
property option. I found nothing to match in Bugster or Bugzilla but I
did find a Fedora match in Google. Is this a known issue? Machine
is an Ul
I see the same on my update to build 129 just done on a Ultra 20. The
icon is just a plain gray blue and when I truss the command I see lots
of python2.6 Err#2.
pfexec truss /usr/bin/packagemanager
Kostas Oikonomou wrote, On 12/18/09 08:42:
Hi,
I've just updated my Toshiba Satellite laptop
side affects is that the workspaces are reduced
and cannot be labeled.
-- Alan
Leslie H Wood wrote:
I updated from 2009.06 SRU7 to build 129 and my 6 workspaces dropped
to 4. Therw is nothing in the preferences option to change this and
no property option. I found nothing to match in Bugs
metacity --replace
Google :)
Leslie H Wood wrote, On 12/18/09 09:05:
I switched back to 2009.06 SRU7 and it is stuck in the same mode and
window behavior. How can you turn it off since there is no man page
for it.
motie:/home/lw ->ps -ef | grep compiz
lw31248 1080 1037 0 08:54
That is only temporary A logout and login restores compiz Is
there a way to disable compiz without running "gnome-cleanup" ?
Thanks...Les
Leslie H Wood wrote, On 12/18/09 09:09:
metacity --replace
Google :)
Leslie H Wood wrote, On 12/18/09 09:05:
I switched back to 20
Alan Coopersmith wrote, On 12/18/09 09:19:
Leslie H Wood wrote:
I switched back to 2009.06 SRU7 and it is stuck in the same mode and
window behavior. How can you turn it off since there is no man page for
it.
Preferences -> Appearance, and set Visual Effects to &q
on by
default. One of the side affects is that the workspaces are reduced
and cannot be labeled.
-- Alan
Leslie H Wood wrote:
I updated from 2009.06 SRU7 to build 129 and my 6 workspaces
dropped to 4. Therw is nothing in the preferences option to change
this and no property option. I fo
0.5.11-0.129
installed -
SUNWgnome-config-editor 0.5.11-0.129
installed -
SUNWjre-config-plugin 0.5.11-0.129
installed -
SUNWlibcompizconfig
Paul
Leslie H Wood wrote:
That must be in a package that is not in the default
Build 111 also known as "2009.06" stable release is probably what should
be used in production especially if you have a support contract. The
support contract gives you access to the SRU1 through SRU7 back ported
fixes to the 2009.06 build.
Builds after 111 (currently at build 130) have fixe
zfs list ?
What does the rpool/swap have ? The complete list would be prefered
B wrote, On 01/04/10 11:19:
Thanks Les
I will stick with 111 on my production notebook and try 130 in a virtualbox if
possible. I heard there were some problems..Downloading now 130
Also on my 111 buil
I think that should be "authorized_keys" and not "authorized_host"
-># ls -al
total 14
drwx-- 2 root other 512 Mar 7 15:25 .
drwxr-xr-x 24 root root 1024 Mar 7 03:33 ..
-rw--- 1 root other 1664 Mar 7 15:34
authorized_
_keys
Thanks,
Sereno
2010/1/7 Leslie H Wood <leslie.w...@sun.com>
I think that should be
"authorized_keys" and not "authorized_host"
-># ls -al
total 14
drwx-- 2 root other 512 Mar 7 15:25 .
drwxr-xr-x
The 2009.06 (build 111b) and 2010.03 stable releases are like RHEL 4, 5 etc
The development builds are like fedora
Quite a few of the development builds are stable but they can have
issues especially when major revisions of items like Xorg are added.
You need to carefully read the release
B wrote, On 01/07/10 15:18:
Thanks, so bottom line stick with 111b on my notebook which I prefer to have a
stable version running and keep 130 on my desktop virtualbox for now?
Did you pickup on the implications of the "beadm list" command I
sent? That it is safe to kick off an upda
ABE's are managed by the beadm command which is pretty simple.
man beadm
Plus any time you run the "pkg image-update" (GUI or CLI) it will
automatically create the update in an ABE. The "pkg image-update"
command has a "--be-name " option so you can define your own name
rather than the
/lw31248
rpool/export/install 2.57G 111G 2.57G /export/install
rpool/export/opt 725M 111G 188M /export/opt
rpool/swap 1023M 111G 842M -
...Les
Leslie H Wood wrote, On 01/07/10 16:51:
ABE's are
managed by the beadm command wh
What do the property settings for sharesmb look like on the other datasets ?
zfs get sharesmb tank/henri/documents tank/henri/music
Henri Maddox wrote, On 01/08/10 08:09:
Hello everybody,
let's say I got the following strukture:
/tank
/tank/henri
/tank/henri/documents
/tank/henri/music
B wrote, On 01/08/10 11:00:
Thanks guys for the input
Any ideas on the easiest way to install an alternate boot environment into the
existing snv_111 that I am using? I would like to be able to boot into dev_130
but have the ability to go back and forth...
I have only install from
B wrote, On 01/08/10 16:10:
Thanks can I just do this:
$ pfexec pkg set-publisher -O http://pkg.opensolaris.org/dev opensolaris.org
$ pfexec pkg image-update
Yes but unless you are happy with the default names for your ABEs
then you may want to use that option to. You will get t
It should be in your path by default and this assume your applications
do not require an older version of Java to work.
motie:/home/lw ->which java
/usr/bin/java
motie:/home/lw ->java -version
java version "1.6.0_13"
Java(TM) SE Runtime Environment (build 1.6.0_13-b03)
Java HotSpot(TM) Server V
B wrote, On 01/12/10 14:36:
I am currently trying image update from 111b to dev_130 on my notebook...I
assume it is working because it has been downloading for quite sometime.
However, my screen went to screensaver mode and will not come out of it while I
am doing image update...Is this norma
Yes it is
ftp://ftp.mozilla.org/pub/calendar/lightning/releases/1.0b1rc1/contrib/
It is working great on my x86 OpenSolaris instance using Thunderbird 3.0
and Lightning 1.0pre and I love the tabs in Thunderbird 3.0
...Les
Alan Steinberg wrote, On 01/12/10 18:17:
This might be due to the new
Thunderbird 3 will only work with Lightning 1.0.So if you have an
earlier version of the Lightning extension, it will get disabled each
time your start Thunderbird. I am not sure what version of Thunderbird
ships with build 130.
I have not had a problem with my ~/.thunderbird files when
Is there a reason you cannot transfer the data over the network ?
Setting up Samba just to copy data seems like work. I have used tar
or rsync which preserves all the ownership ans time stamps on the files.
Michelle Knight wrote, On 01/14/10 15:53:
Well, no rush on this. I decided to
Slight correction:
S10U8 is the first Solaris 10 release to fully support 2TB disks and it
therefore had to go to a 64 bit initial install boot. S10U7 and
earlier used a 32 bit initial boot for the install.
...Les
Bruce wrote, On 01/25/10 11:10:
On Mon, 25 Jan 2010 09:45:58 -0800, B wrote
Giving ZFS the complete disk is the most efficient way for ZFS and then
use datasets for your other requirements. This is the recommended way
per all the guides I have read and it works great. You can add
property controls to each dataset if needed
Filesystem size used avail c
O'Reilly UNIX in a Nutshell or Linux in a Nutshell has pretty useful
sections on sh/ksh/bash shell and scripts.
vattini giacomo wrote, On 01/29/10 13:07:
I need to learn some stuff,ksh scriptings
___
opensolaris-help mailing list
opensolaris-h
/etc/passwd file change your user line to use ksh ?
vattini :x:101:10:Vattini Giacomo:/home/vattini:/bin/ksh
Setup the ~/.profile or ~.kshrc and then logout and login. Probably
best to test it in a new shell window first before you logout. The ksh
should work in a basic fashion without use
You also have the improved vi, vim in OpenSolaris
->which vim
/usr/bin/vim
Richard L. Hamilton wrote, On 01/31/10 09:43:
I think option vi-tabcomplete is pdksh, not real ksh. I don't know what shell
uses option tabcomplete.
AFAIK, real ksh will do tab completion in either emacs or vi editin
They are all editors, vi , vim or emacs.I would guess that emacs is
in an additional package not included in the default install and is an
editor that full time programmers like. You can use any of them for
reading but that is a bit of overkill. Unix also has "view" which
behaves like v
Understandable confusion since "uname -r" does return "5.11" which is
needed for consistency.
desk:/home/lw ->more /etc/release
OpenSolaris 2009.06 snv_111b X86
Copyright 2009 Sun Microsystems, Inc. All Rights Reserved.
Use is sub
I am puzzled why user jack has been changed to an automount location
for the newer builds. I had a finish script that removed the jack
dataset and then called the command "/usr/sbin/userdel -r jack". This
works fine for 2009.06 but fails for the latest builds. It seems an
odd change when y
with bugzilla searches, how did you structure your
search please.
Thanks...Les
Glenn Lagasse wrote, On 02/04/10 16:23:
* Leslie H Wood (leslie.w...@sun.com) wrote:
I am puzzled why user jack has been changed to an automount location for
the newer builds. I had a finish script
Change the menu.lst timeout line
timeout 30
to
timeout -1
It will pass straight through then
Hardeep Narwan wrote, On 02/12/10 13:24:
Hi All,
I've just installed the latest version of OpenSolaris 2009.06 to hopefully find
an OS that finally fits. Things look good so far. I just wante
NIC interfaces OK - full duplex?
pfexec dladm show-phys
Michelle Knight wrote, On 02/16/10 14:56:
CPU more than 90% idle, no disk activity and nothing going on on the network
... yet it takes 15 seconds from entering an SSH command for me to get a
password prompt. This kind of delay isn't
Does traceroute or ping -s from the machine you are doing an ssh from
look good?
What about ssh response out from the OpenSolaris machine ?
Michelle Knight wrote, On 02/16/10 15:37:
Forgot to mention, the ssh is via IP addy not hostname.
___
op
Michelle Knight wrote, On 02/16/10 23:42:
telnet is not enabled; this is an out of the box installation. All that is done is the system drive
is mirrored, SMB is added and the IP is fixed. Incidentally, switching to manual on the IP addy
caused DNS not to work until "dns" was added manually
Pretty sure we are only talking about DNS client here so setup is easy
and I do not see any overhead.At home, see if you can figure out
what your internet provider is supplying.
1. If you are using their DHCP then you may have DNS already setup when
the Opensolaris DHCP client connects.
B wrote, On 02/21/10 23:44:
Hi,
Can someone please tell me clearly how to access single user mode. I need to
update my 132 build to 133 and only have 1 GB on my x64 notebook. This was not
quite enough memory to complete the new update.
If the system is up and running than you can use th
man init :-)
The above shows you can get back to multi-user as well but a reboot will
test that any changes you made will survive a reboot. Yes a reboot
should start mult-user mode.
B wrote, On 02/22/10 10:01:
Thanks machine is up and running so will use init s command. Will machine
reve
root # svcs -a | grep milestone
disabled Feb_20 svc:/milestone/patching:default
online Feb_20 svc:/milestone/name-services:default
online Feb_20 svc:/milestone/devices:default
online Feb_20 svc:/milestone/network:default
online Feb_20 svc:/milestone
lti-user:default
online 11:14:00 svc:/milestone/multi-user-server:default
Leslie H Wood wrote, On 02/22/10 12:17:
root # svcs -a | grep milestone
disabled Feb_20 svc:/milestone/patching:default
online Feb_20 svc:/milestone/name-services:default
online Feb_20
No, once you login to the single user root session then a will
exit single user and return the system to the multi-user state.
You should use a non root account for most tasks and only use the actual
root login for major system fixes or updates. Single user mode you
pretty much just have roo
You can always create an ABE before you try something unknown.
I append "-ABE" to the BE I want to protect
beadm list
BE Active Mountpoint Space Policy Created
-- -- -- - -- ---
opensolaris-sru7 - - 20.30M static
beadm activate opensolaris-sru8-ABE
Oops, I missed the target name on the beadm activate command...
Leslie H Wood wrote, On 02/27/10 10:08:
You can always create an ABE before you try something unknown.
I append "-ABE" to the BE I want to protect
beadm list
BE
The rpool disk cannot use EFI labels on 2009.06 and I am not sure if
this is still true for 2010.03. You basically format the disk using
a standard Solaris label and (normally) assign the whole disk to the
rpool on slice 0.
This is why you have to use c9t0d0s0 (include the slice) when setti
When you attached your mirror did you you install a bootblock on the
mirror disk ?
You have to manually create a bootblock or an additional grub menu item
(x86) for the new mirrored disk which is a known issue.
Sparc: installboot -F zfs /usr/platform/`uname -i`/lib/fs/zfs/bootblk
/dev/rdsk/c0
I run that command after I attach the mirror... Not sure if that makes
a difference.
...Les
Geoff wrote, On 03/18/10 11:52:
Hi Leslie. Yup, this is the command I used before attaching the drive to the
mirror:
"pfexec installgrub -m /boot/grub/stage1 /boot/grub/stage2 /dev/rdsk/c8t1d0s0"
->rmformat -l
Looking for devices...
1. Logical Node: /dev/rdsk/c3t0d0p0
Physical Node: /p...@0,0/pci-...@4/i...@0/s...@0,0
Connected Device: ATAPIDVD A DH18A1P ZS51
Device Type: CD Reader
Bus: IDE
Size:
Label:
Access permissions:
philo neo
Not seen a reply to this as yet but the only method I know is for Sun
x86 hardware and that is ipmitool. The prtdiag -v works for Sparc but
is not that useful for x86.
On x86 the SP monitors this data and ipmitool can access it.
ipmitool sdr list
PWRBS| 0 unspecified | nc
AC
The only way for Solaris which is what I assume you are asking about is
to scan the /var/sadm/system/admin/clustertoc file. This is good for
the package only and you can see which cluster it is included in or not.
You can only associate the files to the package on the installed
packages via
Use the "format" command which will put you into a mini environment with
options. Look at "man format" since some tests will wipe your drive
and others do not. It does warn you if you pick an option that will
overwrite and allows you to escape.
Leank Soon wrote, On 04/22/10 05:06:
Hi,
I
Erich Weiler wrote, On 04/30/10 14:47:
Actually I just saw this in the documentation for the SunFire x4540:
--
Installing Solaris OS Onto an X4540 Using a Compact Flash Card
When installing Solaris 10 onto an X4540 Server using a Compact Flash (CF) card
as the boot media, do the foll
Erich Weiler wrote, On 04/30/10 15:18:
Probably more to do with the fact that 8 GB for root is pretty
minimal. It also is not a good idea to have swap on a CF so you
need enough memory for swap to be accommodated. We do ship some
servers with the OS on a CF and following the above works qu
pkgchk is a system 5 utility and is only legacy support in
OpenSolaris. Look at the IPS command "man pkg"
motie:/home/lw31248 ->pkg search /usr/bin/cp
INDEX ACTION VALUE PACKAGE
path file usr/bin/cp pkg:/sun...@0.5.11-0.75
path file usr/bin/cp pkg:/sun...@0.5.11-0.93
On 5/21/2010 7:47 PM, Dr Lou wrote:
some recent testing on a 'sandbox' machine has raised an interesting
question... One of our techs, in fiddling with this machine, seems to have
deleted some key files; at least part of the /etc directory as well as - at
least - some of the devices director
Dr Lou wrote, On 05/22/10 04:03:
LW,
Thanks for your reply - but my question wasn't on [i]prophylaxis...[/i]
The point was to generate a little 'creative destruction', then to test what
resources openSolaris has for [i]treatment.[/i]
So now, if this [i][/i]were, really,[b][/b] a damaged pro
export PKG_CLIENT_TIMEOUT=900
export PKG_TIMEOUT_MAX=8
The above used to be recommended before starting an install but this was
back in b118 days and I am not sure if it is still needed.
weekleyj wrote, On 06/05/10 17:00:
I've been having an intermittent problem while getting
packages.openso
There have been 10 Support Repository Updates (SRU) for 2009.06 so far
if that helps to justify a contract.
Alan Coopersmith wrote, On 06/17/10 07:18:
Stephan Ferraro wrote:
Normally they should be available patches to patch only the kernel to gurantee that a release is stable. I
Edward Ned Harvey wrote, On 06/17/10 19:54:
From: opensolaris-help-boun...@opensolaris.org [mailto:opensolaris-
help-boun...@opensolaris.org] On Behalf Of Simon Gao
I replaced one faulty disk with a new disk in the same disk slot. The
faulty disk has a device name as c2t6d0. I expect the syste
Richard L. Hamilton wrote, On 06/18/10 16:15:
Hi
Currently I have 400+ users with quota set to 500MB
limit. Currently the file system is using veritas
file system.
I am planning to migrate all these home directory to
a new server with ZFS. How can i migrate the quotas.
I can create 400+ file
The look at "devfsadm" instead of "cfgadm"
devfsadm -C -c disk
bonso wrote, On 06/21/10 04:21:
Hello,
Having only replaced one failing disk I'm not really the voice of experience
and I'll even make an assumption but I'd say this is normal. My reasoning is
that you replace one disk with a
Automount at work which I guess was updated when you created you userID
Look at /etc/auto_home
This default home directory is assumed to be /home/user_name and the
real location is else where but standard practice usually uses the
export directory /export/home/user_name so you can have your h
Thommy M. Malmström wrote, On 06/23/10 12:11:
/etc/dfs/dfstab
is empty and autohome is
schattenlocalhost:/export/home/&
+auto_home
automaster is
+auto_master
/net-hosts -nosuid,nobrowse
/home auto_home -nobrowse
So because of auto_master I have
Correction:
is totally controlled by the "schatten" line in /etc/auto_home
Should be
is totally controlled by the "schatten" line in /etc/auto_home and
etc/passwd
Leslie H Wood wrote, On 06/23/10 12:51:
Thommy M. Malmström wrote, On 06/23/10 12:11:
/etc/dfs
Since the rpool can not use EFI labels you need to format the mirror disk so
it is the same as your rpool disk.
Example and your disk target will/may differ.
prtvtoc /dev/rdsk/c7t0d0s2 | fmthard -s - /dev/rdsk/c7t1d0s2 (s2 is the
whole disk)
Then just attach the disk to be mirrored
That is RBAC and by default in OpenSolaris, root is a role for more
security.Your user ID was probably given "roles=root" so you can use
"pfexec" to run privileged commands from your normal unprivileged user login.
You can disable root as a role if you prefer not to have the RBAC setup.
S
Since the rpool is created by default as a single drive and also it is
different from additional pools in that it cannot be an EFI file system then the
"attach" option is how you add the mirror as discussed in earlier e-mail.The
additional pool consists of 5 mirrored pairs constructed how
It is also a good idea to add a cron to scrub each pool that you have at
regular intervals.
# zpool scrub commands
5 0 * * 0 /usr/sbin/zpool scrub pool
5 1 * * 0 /usr/sbin/zpool scrub rpool
Leslie H Wood wrote, On 07/12/10 20:04:
Since the rpool is created by default as a single drive and
Correction: "EFI file system" should have been "EFI partition"
Leslie H Wood wrote, On 07/12/10 20:04:
Since
the rpool is created by default as a single drive and also it is
different from additional pools in that it cannot be an EFI
To check it's status
svcs -l svc:/application/graphical-login/gdm:default
To manipulate it
svcadm svc:/application/graphical-login/gdm:default
dave wrote, On 07/23/10 08:05:
dmesg | tail doesnt show any serious errors other than nwam.
Whats the command to get the display.
___
dave wrote, On 07/23/10 08:05:
dmesg | tail doesnt show any serious errors other than nwam.
Whats the command to get the display.
svcs -xv
Will show if your start up precess have issues and a clean result would be
nothing.
mine:/home/xxx ->svcs -xv
mine:/home/xxx ->
I am sorry but I deleted the prior mail so I have no idea what version of
OpenSolaris your are running. There are stable releases of OpenSolaris and
Development releases.Looks like you are on 2009.06 (build 111b) per the
$HOME/.dmrc issue.I would suggest moving to 134 from the
Redirect to a file or start a "script" (man script).
svcs -xv > svcs.dump
dave wrote, On 07/23/10 09:10:
svcs -xv shows a long list that moves rapid, cant copy. Gedit lacks vi or
script capabilities for saving the output.
___
opensolaris-help m
"gedit svcs.dump" or "vi svcs.dump" or "less svcs.dump"or is your
install hosed to the point that they do not work ?
dave wrote, On 07/23/10 09:33:
How to access the svcs.dump
___
opensolaris-help mailing list
opensolaris-help@opensolaris.or
A quick search on http://opensolaris.org found the following which may help.
It does have a pointer for new network drivers. I would still suggest a better
starting point would be build 134 since it has a bunch of updated items.
http://hub.opensolaris.org/bin/view/Community+Group+documenta
1 - 100 of 129 matches
Mail list logo
