solaris 10 build 63 and lofiadm

solaris 10 build 63 and lofiadm

Post by Coy Hil » Tue, 24 Aug 2004 03:00:19



Based on the discussion here:
http://sunsolve.sun.com/pub-cgi/retrieve.pl?doc=fsrdb%2F50566

does the following logic work:

(a) The size and starting location of slice 1 on the 1st disk of 10
build 63 is still at offset 452 decimal in the vtoc.

Thus, I get the following:


0000000 0000000691 0000824960
0000010

which implies that slice 1 starts at block 691 and is 824960 blocks long.

The reason that I ask is that after using those calculations (and
creating a jumpstart server),  I used Live Upgrade to install 10_63 onto
  an inactive boot environment and found two interesting occurrences.  I
found that one package (SUNWmpatchmgr) did not install correctly during
the operation of luupgrade.  I also found that no user or role would
authenticate correctly to any of the smc suite of tools even though I
know the passwords are correct.

Any ideas?

Thanks,

--
Coy Hile

 
 
 

solaris 10 build 63 and lofiadm

Post by Coy Hil » Tue, 24 Aug 2004 06:09:50



> Thus, I get the following:


> 0000000 0000000691 0000824960
> 0000010

> which implies that slice 1 starts at block 691 and is 824960 blocks long.

I misspoke in the previous; that is *cylinder* 691 and block 442240 (691
* 640).

 
 
 

solaris 10 build 63 and lofiadm

Post by Scott Howar » Tue, 24 Aug 2004 08:42:46



> Based on the discussion here:
> http://sunsolve.sun.com/pub-cgi/retrieve.pl?doc=fsrdb%2F50566

I wrote the document, so I suppose I should reply :)

>> Thus, I get the following:


>> 0000000 0000000691 0000824960
>> 0000010

>> which implies that slice 1 starts at block 691 and is 824960 blocks long.
> I misspoke in the previous; that is *cylinder* 691 and block 442240 (691
> * 640).

That all look good to me. Realistically, if you can mount the new ISO
via lofiadm, and it doesn't give any errors/panic whilst doing so then
the odds of you having it wrong is pretty low.

Quote:>> The reason that I ask is that after using those calculations (and
>> creating a jumpstart server),  I used Live Upgrade to install 10_63 onto
>>  an inactive boot environment and found two interesting occurrences.  I
>> found that one package (SUNWmpatchmgr) did not install correctly during
>> the operation of luupgrade.  I also found that no user or role would
>> authenticate correctly to any of the smc suite of tools even though I
>> know the passwords are correct.

Where there any errors in the live-upgrade log? What version were you
live-upgrading from?

  Scott.

 
 
 

solaris 10 build 63 and lofiadm

Post by Coy Hil » Tue, 24 Aug 2004 21:47:01



> Where there any errors in the live-upgrade log? What version were you
> live-upgrading from?

I am upgrading from Solarsi 9, 9/04, and I don't recall seeing any
errors.  Once the *y download site lets me re-download the ISOs (rm
* in the wrong subdir is so much fun ;)), I will do the LiveUpgrade
again and let you know.

Unrelated, is something hosed on the download site?  I click on "I
already registered for a free D/L license" and I get to the instructions
window, but have no link to the actual log in to SLDC and download page
for Solaris Express 8/04.

--
Coy Hile

 
 
 

solaris 10 build 63 and lofiadm

Post by Gerry » Thu, 26 Aug 2004 01:05:59




> > Where there any errors in the live-upgrade log? What version were you
> > live-upgrading from?
> I am upgrading from Solaris 9, 9/04, and I don't recall seeing any
> errors.  Once the *y download site lets me re-download the ISOs (rm
> * in the wrong subdir is so much fun ;)), I will do the LiveUpgrade
> again and let you know.

Yes that is annoying. One mistake/ype and..

Quote:> Unrelated, is something hosed on the download site?  I click on "I
> already registered for a free D/L license" and I get to the instructions
> window, but have no link to the actual log in to SLDC and download page
> for Solaris Express 8/04.

Definitely. At least on Aug 23rd. Use the other link : <
 
 
 

solaris 10 build 63 and lofiadm

Post by Coy Hil » Thu, 26 Aug 2004 04:07:46



> Where there any errors in the live-upgrade log? What version were you
> live-upgrading from?

>   Scott.

Scott,

Here are the parts of the live-upgrade log (going from Sol9 9/04 ->
10_b63) that seem a bit... weird

------ START LOG EXCERPTS -------
Installing new packages

Copyright 2004 Sun Microsystems, Inc.  All rights reserved.
Use is subject to license terms.

Doing pkgadd of SUNWcsr to /.
1640 blocks
/a/etc/security/policy.conf updating entries for privileges(5), see
policy.conf(
4) for details.
/a/etc/security/policy.conf updating entry for LOCK_AFTER_RETRIES, see
pam_unix_
auth(5) for details.

Installation of <SUNWcsr> was successful.

Doing pkgadd of SUNWxrpcrt to /.

This appears to be an attempt to install the same architecture and
version of a package which is already installed.  This installation
will attempt to overwrite this package.

2125 blocks
/a/usr/bin/wscompile preserved
/a/usr/bin/wsdeploy preserved

Installation of <SUNWxrpcrt> was successful.

Doing pkgadd of SUNWmccom to /.
1687 blocks
Replacing current install root with //a.

Installation of <SUNWmccom> was successful.
Doing pkgadd of SUNWmcc to /.
2381 blocks
Integration Complete
See /tmp/dtappint.log file for more information

Installation of <SUNWmcc> was successful.
Doing pkgadd of SUNWmc to /.
1102 blocks
/a/usr/sadm/lib/smc/prereg/SUNWmc/SUNWmc.mcreg already exists

Installation of <SUNWmc> was successful.

Doing pkgadd of SUNWwbmc to /.
254 blocks
/a/usr/sadm/lib/smc/prereg/SUNWwbmc/SUNWwbmc.mcreg already exists

Installation of <SUNWwbmc> was successful.

Doing pkgadd of SUNWdclnt to /.
1066 blocks
/a/usr/sadm/lib/smc/prereg/SUNWdclnt/SUNWdclnt.mcreg already exists

Installation of <SUNWdclnt> was successful.

Doing pkgadd of SUNWlvmg to /.
1786 blocks
/a/usr/sadm/lib/smc/prereg/SUNWlvmg/SUNWlvmg.mcreg already exists

Installation of <SUNWlvmg> was successful.
Doing pkgadd of SUNWrmui to /.
352 blocks
/a/usr/sadm/lib/smc/prereg/SUNWrmui/SUNWrmui.mcreg already exists

Installation of <SUNWrmui> was successful.

Doing pkgadd of SUNWmozchat to /.
1105 blocks
Segmentation Fault - core dumped
Segmentation Fault - core dumped
Segmentation Fault - core dumped
Segmentation Fault - core dumped
Segmentation Fault - core dumped

Installation of <SUNWmozchat> was successful.
Doing pkgadd of SUNWmozgm to /.
21 blocks

Installation of <SUNWmozgm> was successful.
Doing pkgadd of SUNWmozmail to /.
22144 blocks
Segmentation Fault - core dumped
Segmentation Fault - core dumped
Segmentation Fault - core dumped
Segmentation Fault - core dumped
Segmentation Fault - core dumped

Installation of <SUNWmozmail> was successful.

Doing pkgadd of SUNWmpatchmgr to /.
497 blocks
No <path> specified

  **** [usage statement for smreg snipped] ****

  For more information, see smreg(1M).

pkgadd: ERROR: postinstall script did not complete successfully

Installation of <SUNWmpatchmgr> failed.
pkgadd return code = 1

--- END LOG EXCERPTS ----

I am going to bring the machine into work here tomorrow and re-create
the jumpstart tree using actual burns of the ISOs (rather than the
lofiadm procedure) and see if that causes any different results.  Will
advise.

Thanks,

--
Coy Hile

 
 
 

solaris 10 build 63 and lofiadm

Post by Coy Hil » Fri, 27 Aug 2004 22:24:47


I've just thought of another possible solution to the problem that I am
seeing.

When one runs 'smc' for the first time, it creates a bunch of  files
that evidently 'register' components or something and create toolboxes.
  Is there something stored in the filesystem that I could remove before
booting the new (upgraded) boot environment that would make the solaris
management console tools think that they are being run for the first
time and initialise themselves from scratch?

Thanks,

--
Coy Hile

 
 
 

1. Solaris 10 (build 63) upgrade fails on Solaris 8 as well as Solaris 9 sparcs

Hello
         Using the Solaris 10 build 63 for sparc.

Tried to do upgrades on the following two machines.

1. E4500  Solaris 8 (117350-11 Kernel level)
    Entire Distribution OEM support cluster, OBP 3.2.30.

2. Ultra 2  Solaris 9 Entire Distribution and OEM support cluster.

On both machines the upgrade starts when I put the CD 1 and and reaches to
the point where I have to select whether I want to do "Initial Install" or
"Upgrade".

When I check upgrade it starts initializing the system and halfway through I
get the follwoing error

"Could not reinitialize the system state".

At which point I have no option but to click OK on the error and exit out of
the install process.

Has anyone seen this can anyone provide any tips or pointers.?????

Thanks
Sandeep

2. rsh, rlogin and sr(!)

3. Solaris 10 -- Where to get gcc (or other compiler) built on Solaris 10?

4. Where is the net-pf-5 module??

5. gcc 4.1.1 build under Intel Solaris 10 does not build gcc

6. ANNOUNCEMENT: A Multi-Terabyte Capable Web Server

7. Solaris 10 1/06 OK, but Solaris Express Build 33 and 36 fail

8. How to partition for Linux?

9. Solaris 10, built new disk, now cannot boot off solaris 8 disk???

10. Jumpstart solaris 10 b69 and solaris 10 b72 on an Ultra 30

11. Live upgrade failure: Solaris 10 -> Solaris 10 update 1.

12. lsof 4.70D for Solaris 10 [was Re: Lsof on Solaris 10 x86?]

13. Solaris 10 build number?