Hello,
Thanks for the link - that's what I needed. I don't think I got down to
the appendix before I emailed the list.
I wrote up a wiki page for my own documentation on how the upgrade when,
I'm going to just copy and paste it here, sorry for the formatting!
Overall I'd say it went well, with one or two quirks that look harmless.
cheers,
Brian
-------------------------------------------------------------------------------------------
copy/pasted from my wiki page -
I'm running a dell studio XPS workstation for my desktop, using the
Nvidia drivers and the Gnome desktop on Solaris 11 Express, two monitors
each running it's own X display.
Not upgraded from OpenSolaris. (reinstalled from scratch from previous
Vista-64 install)
12 gig of memory, 3 internal hard disks (2 rpool mirrors, 1 it's own
pool, to be mirrored someday in the future), 1 cpu - x86 (GenuineIntel
106A4 family 6 model 26 step 4 clock 2660 MHz) Intel(r) Core(tm) i7 CPU
920 @ 2.67GHz. No non-global zones, but two virtual machines running on
VirtualBox 4.1.4 that had previously been migrated from VMware
Workstation on Vista.
3 things had been installed via pkgadd -
firefox-6.0.en-US.opensolaris-i386-pkg
VirtualBox-4.1.4-SunOS-r74291.pkg
pidgin-2.2.1-sol10-x86-local
We use sudo, not pfexec a root shell prompt, other than that I pretty
much followed this URL -
http://download.oracle.com/docs/cd/E23824_01/html/E24456/upgrade-1.html#scrolltoc
pkg publisher
PUBLISHER TYPE STATUS URI
solaris origin online http://pkg.oracle.com/solaris/release
My publisher matched that.
sudo pkg install pkg:/package/pkg
Packages to update: 2
Create boot environment: Yes
DOWNLOAD PKGS FILES XFER (MB)
Completed 2/2 366/366 2.0/2.0
PHASE ACTIONS
Removal Phase 36/36
Install Phase 128/128
Update Phase 291/291
PHASE ITEMS
Package State Update Phase 4/4
Package Cache Update Phase 2/2
Image State Update Phase 2/2
PHASE ITEMS
Reading Existing Index 8/8
Indexing Packages 2/2
A clone of maininstall exists and has been updated and activated.
On the next boot the Boot Environment solaris-1 will be mounted on '/'.
Reboot when ready to switch to this updated BE.
# beadm list
BE Active Mountpoint Space Policy Created
-- ------ ---------- ----- ------ -------
maininstall A / 88.5K static 2011-06-20 18:30
maininstall-1 R - 5.14G static 2011-10-21 14:31
solaris - - 13.33M static 2011-08-22 10:54
This step worked fine - I happened to create a new boot environment
after installation and before I modified anything, and called it
'maininstall', so it cloned that instead of 'solaris'.
Next I shutdown my Virtualbox vms, and rebooted with
sudo init 6
Came back up fine, added 3 service SMF definitions. I noticed that the
auto-snapshots feature was offline, for unrelated reasons I think, so I
disabled the services -
sudo svcadm disable svc:/system/filesystem/zfs/auto-snapshot:frequent svc:/system/filesystem/zfs/auto-snapshot:daily svc:/system/filesystem/zfs/auto-snapshot:hourly svc:/system/filesystem/zfs/auto-snapshot:monthly svc:/system/filesystem/zfs/auto-snapshot:weekly
I started my Virtualbox vms and kicked them off - as much as possible
I'm trying to do this while 'online' to simulate a production server I
want to avoid downtime on. They are a Windows 7 vm and a Solaris 10 vm
running two zones (nagios and rsync server).
Next was the upgrade piece.
sudo pkg update
Output looked kind of like -
Creating plan / (spinning thing)
Packages to remove: 291
Packages to install: 210
Packages to update: 599
Create boot environment: Yes
Create backup boot environment: No
DOWNLOAD PKGS FILES XFER (MB)
(a whole bunch of packages being downloaded)
During the downloading, I didn't notice much impact on the desktop, a
little bit more stuttering than normal on the Virtualbox keyboard input,
but nothing bad. The VMs look okay, chat and calendar fine on Windows 7,
nagios and RT running fine inside the zones on the Solaris 10 VM. prstat
-Z shows the pkg process doing things, but I'd expect that.
Next was the removal/install/update steps. During the removal phase the
nagios zone failed to ping localhost for at least 10 seconds (a
timeout). That could be complete coincidence, though that check normally
never fails. The network interfaces are Virtualbox mappings to exclusive
IP zones, but the ping check there is to 127.0.0.1. Interesting,
possibly there's a tiny bit of network flakiness as work is being done?
No idea, so many layers there with Virtualbox in the middle. In any
case, not something I'm terribly concerned about unless I see it in more
testing.
That looks like it finished more or less okay, here's the full output
(including above) from the pkg update command.
sudo pkg update
Packages to remove: 291
Packages to install: 210
Packages to update: 599
Create boot environment: Yes
Create backup boot environment: No
DOWNLOAD PKGS FILES XFER (MB)
Completed 1100/1100 70506/70506 1056.2/1056.2
PHASE ACTIONS
Removal Phase 29870/29870
Install Phase 71742/71742
Update Phase 77070/77070
PHASE ITEMS
Package State Update Phase 1699/1699
Package Cache Update Phase 890/890
Image State Update Phase 2/2
PHASE ITEMS
Reading Existing Index 8/8
Indexing Packages 1100/1100
Optimizing Index...
PHASE ITEMS
Indexing Packages 809/809
A clone of maininstall-1 exists and has been updated and activated.
On the next boot the Boot Environment maininstall-2 will be
mounted on '/'. Reboot when ready to switch to this updated BE.
The following unexpected or editable files and directories were
salvaged while executing the requested package operation; they
have been moved to the displayed location in the image:
var/saf -> /tmp/tmp6zILAD/var/pkg/lost+found/var/saf-20111110T092907Z
var/sadm/system/logs -> /tmp/tmp6zILAD/var/pkg/lost+found/var/sadm/system/logs-20111110T092907Z
var/run -> /tmp/tmp6zILAD/var/pkg/lost+found/var/run-20111110T092907Z
usr/lib/python2.4/site-packages -> /tmp/tmp6zILAD/var/pkg/lost+found/usr/lib/python2.4/site-packages-20111110T092908Z
etc/saf/zsmon -> /tmp/tmp6zILAD/var/pkg/lost+found/etc/saf/zsmon-20111110T092908Z
etc/saf -> /tmp/tmp6zILAD/var/pkg/lost+found/etc/saf-20111110T092908Z
---------------------------------------------------------------------------
NOTE: Please review release notes posted at:
http://www.oracle.com/pls/topic/lookup?ctx=E23824&id=SERNS
---------------------------------------------------------------------------
As you can see, there were some things it 'dealt with' by putting into a
lost+found directory under /var/pkg. I'm wondering if it has anything to
do with when I installed a few SVR4 packages -
firefox-6.0.en-US.opensolaris-i386-pkg
VirtualBox-4.1.4-SunOS-r74291.pkg
pidgin-2.2.1-sol10-x86-local
Other than those it's all Solaris 11 Express, this system wasn't
upgraded from OpenSolaris (it was reinstalled from scratch from Windows
Vista, thank goodness).
So, for whatever reason, before rebooting I'm not concerned about these
'fixed' files and directories on the boot image. I like that they'll
still be there on the old BE after reboot, in case there is an issue.
Guess I'll find out after the reboot, which is next. Sadly, I can't save
my desktop console log, will see if it shows up in a log file somewhere.
After shutting down my VMs, here goes.
sudo init 6
The reboot went fine - it did a reconfigure reboot, Configuring devices,
after converting the SMF from version 5 > 6, 6>7, and 7->8. All told,
about 10 minutes until it was booted back into the desktop and I had the
VMs starting again. Beadm output looks like this now.
beadm list
BE Active Mountpoint Space Policy Created
-- ------ ---------- ----- ------ -------
maininstall - - 11.39M static 2011-08-22 11:15
maininstall-1 - - 28.14M static 2011-11-10 08:45
maininstall-2 NR / 16.17G static 2011-11-10 09:28
solaris - - 13.33M static 2011-08-22 10:54
The VMs look good, nagios and RT are up, things are running well from
what I can tell at a first look. My repeat-key settings seem to be
speedier than before, but that's purely anecdotal. All in all pretty
smooth sailing for my first try. I'll have to try it on a test SPARC
server when I get a chance.
Post by Cindy SwearingenHi Brian,
http://download.oracle.com/docs/cd/E23824_01/html/E24456/upgrade-1.html
http://download.oracle.com/docs/cd/E23824_01/index.html
Let us know how the upgrade goes.
Thanks,
Cindy
Post by Brian WilsonCan't find the documentation for upgrading from Solaris 11 Express
yet...hmmm... maybe my google-foo is bad today.
cheers,
Brian
Post by Glynn FosterToday marks the release of Oracle Solaris 11, the first cloud OS.
Oracle Solaris 11 delivers ground-breaking features for secure and
agile deployment of services in large scale cloud environments and
enterprise data centers. With over 4,000 different new features,
Oracle Solaris 11 raises the bar on enterprise operating systems.
Oracle Solaris 11 is 7 years in the making and a whole new set of
capabilities, from advanced network virtualization to high
performance cryptography and virtualization, dependency aware
software packing and installation technologies.
http://www.oracle.com/technetwork/server-storage/solaris11/overview/whats-new/index.html
http://oracle.com/technetwork/server-storage/solaris11/downloads/index.htm
http://download.oracle.com/docs/cd/E19963-01/index.html#release-info
http://oracle.com/technetwork/server-storage/solaris11/documentation/index.html
http://oracle.com/technetwork/server-storage/solaris11/training/index.html
Blog: http://blogs.oracle.com/solaris
Facebook: http://facebook.com/oraclesolaris
Twitter: http://twitter.com/ORCL_Solaris
I'd like to express my thanks and congratulations to each and
everyone who has contributed in some way to this release. Help
spread the word!
Glynn Foster
On behalf of the Oracle Solaris Team
_______________________________________________
opensolaris-discuss mailing list
--
-----------------------------------------------------------------------------------
Brian Wilson, Solaris SE, UW-Madison DoIT
Room 3114 CS&S 608-263-8047
brian.wilson(a)doit.wisc.edu
'I try to save a life a day. Usually it's my own.' - John Crichton
-----------------------------------------------------------------------------------