Re: Migrating Solaris
- From: AGT <usenetpersongerryt@xxxxxxxxx>
- Date: Sun, 22 Jun 2008 19:45:37 -0700
On Sun, 22 Jun 2008 21:09:15 -0400, Richard B. Gilbert wrote:
I am doing some testing and I want to restore our production box onto
a box on the testbench. The servers are both the same class and the
operating system is 2.6 (hopefully as a result of this testing they
will upgrade to something a little more modern!).
Dont count on it. Companies that far back cant seem to get out of
that rut : <
If you are going to go through the pain of upgrading, you should shoot
for something a LOT more modern; e.g. Solaris 10. The upgrade can be
done in 30 - 45 minutes. Testing the applications to make sure that
they run correctly under the new version may take months!
I restore all of the necessary slices / /var /opt and /usr. I put a
bootblk on the new disk, I regen the the device tree and reboot.
regen..? But did you change the relevant files in /etc??
On reboot I get the message during the startup /dev/ip device busy.
I try ifconfig -a and get
ifconfig:socket: device busy
I boot into single user mode and ifconfig -a
and successfully get back le0
le0?? - Nothing newer than Solaris 9 will see the ether card.
Just so you know.
I try ifconfig le0 plumb and get the message
Open /dev/ip: device busy.
Its already plumbed. So its kinda busy.
Any suggestions would really be appreciated.
What hardware is this...?
Are you sure that BOTH boxes have "le0"? If one is le0 and the other is
hme0, you are going to have a few problems.
Even if both boxes have le0, they both have the same IP address now. If
you are going run both boxes on the same network You will need to run
sys-unconfig on the test box.
Sounds like theres a FEW reasons to do THAT : >
- Prev by Date: Re: Sanity check of filesystem design
- Next by Date: Re: build 91 fonts UGLY
- Previous by thread: Re: Migrating Solaris
- Next by thread: build 91 fonts UGLY