kernelID and patch112438-03
I am having problem in installing patch 112438-03 in 3 sunfire v240s running Solaris 8.
On one server the patch installed with no problem but on the other twos I am getting the following error
# patchadd /var/spool/patch/112438-03
Checking installed patches...
Verifying sufficient filesystem capacity (dry run method)...
Patch 112438-03 failed to install due to a failure produced by pkgadd.
See /var/sadm/patch/112438-03/log for details
Patchadd is terminating.
# more /var/sadm/patch/112438-03/log
Package not patched:
PKG=SUNWcarx.us
Original package not installed.
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.
/var/spool/patch/112438-03/SUNWcarx.u/install/checkinstall: /var/spool/patch/112438-03/SUNWcarx.u/install/checkinstall: cannot open
pkgadd: ERROR: checkinstall script did not complete successfully
Dryrun complete.
No changes were made to the system.
#
I was puzzled because all 3 servers had Solaris 8 installed from the same jumpstart server. Didn't understand why the patch installed flawless in one server but not in order two. A investigation revelead that on the server were the patch was installed the kernelID is Generic_117350-27 whereas the other two servers where the patch didn't installed the kernelID is Generic_117350-26 (?!) Have no idea how the servers endup with different kernelIDs considering that they all have the same patch cluster installed and the hardware configuration is the same in all 3 servers.
Question: how one update the kernelid?
# 1
For the record.
Found way to fix my problem.
To update the kernelID, I unzipped the latest patch_cluster in /tmp and installed from there, and the kernel was patched
$ unzip 8_Recommended -d /tmp
....
$ uname -v
Generic_..117350.-26
$ cd /tmp/8_R*
$ su root
# ./install_cluster
...
# shutdown -i6 -g0 -y
...
$ uname -v
Generic_117350-29
$
Earlier to install patch 112438-03, see email exchange below
>
> Darren, thanks a lot!!!!
>
> able to install the patch using the second method, adding an install
> account install:x:0:1:installpatch braindamage:/:/bin/true
>
> of course, for security reasons, the account was deleted after the patch
> was installed.
> not very comfortable with the idea of leaving an account with uid=0 even
> though it is a locked account.
> I think Sun has to do something to fix this. Creating an account with
> uid=0 and leave it on the system is kind of risky. The auditors of my
> company will not buy it.
I've never had to use that method. Normally it's more of a problem with
the directory that the patch is in (especially if NFS is involved).
So extracting the patch into /tmp and applying from there has always
been sufficient for me.
--
Darren Dunham<a href="mailto:ddunham@taos.com" target="_blank">ddunham@taos.com</a>
Senior Technical Consultant TAOS<a href="http://www.taos.com/" target="_blank">http://www.taos.com/</a>
Got some Dr Pepper?San Francisco, CA bay area
< This line left intentionally blank to confuse you. >
__
Solaris-Users mailing list
<a href="mailto:Solaris-Users@filibeto.org" target="_blank">Solaris-Users@filibeto.org</a>
<a href="http://www.filibeto.org/mailman/listinfo/solaris-users" target="_blank">http://www.filibeto.org/mailman/listinfo/solaris-users</a >
> /var/spool/patch/112438-03/SUNWcarx.u/install/checkinstall:
> /var/spool/patch/112438-03/SUNWcarx.u/install/checkinstall: cannot open
> pkgadd: ERROR: checkinstall script did not complete successfully
> Dryrun complete.
> No changes were made to the system.
> #
Does the FAQ entry apply to your system?
<a href="http://www.science.uva.nl/pub/solaris/solaris2.html#q5.59" target="_blank">http://www.science.uva.nl/pub/solaris/solaris2.html#q5.59< ;/a>
5.59) Patch installation often fails with "checkinstall" errors.
When installing a patch, the Solaris 2.5+ patch installation procedure
will execute the script "checkinstall" with uid nobody.
If any of the patch files or if any part of the path leading up to the
patch directory cannot be read by nobody, an error similar to the
following will appear:
patchadd .# or ./installpatch .
Generating list of files to be patched...
Verifying sufficient filesystem capacity (exhaustive method) ...
Installing patch packages...
pkgadd: ERROR: checkinstall script did not complete successfully
....
You can workaround this in two ways, one is to make sure that the user
"nobody" can read all patch files and execute a "pwd" in the patch
directory or add an account "install" to /etc/passwd:
install:x:0:1:installpatch braindamage:/:/bin/true
Installpatch and patchadd use "nobody" as a fallback if it cannot find
the "install" user.
--
Darren Dunham<a href="mailto:ddunham@taos.com" target="_blank">ddunham@taos.com</a>
Senior Technical Consultant TAOS<a href="http://www.taos.com/" target="_blank">http://www.taos.com/</a>
Got some Dr Pepper?San Francisco, CA bay area
< This line left intentionally blank to confuse you. >
__
Solaris-Users mailing list
<a href="mailto:Solaris-Users@filibeto.org" target="_blank">Solaris-Users@filibeto.org</a>
<a href="http://www.filibeto.org/mailman/listinfo/solaris-users" target="_blank">http://www.filibeto.org/mailman/listinfo/solaris-users</a >
