[pca] Patch weirdness -- any ideas?

[Rajiv Gunja]

Thanks Martin, I thought so too and have advised the same to the SysAdmin.

-GGR

--
Rajiv G Gunja
Blog: http://ossrocks.blogspot.com


2008/10/7 Martin Paul <martin at par.univie.ac.at>

> Hi,
>
>  Why is it that I am seeing a Solaris 8 patch on a Solaris 10 server and
>> that
>> too a bad patch?
>>
>> 04. Is it because of this?
>> Patch: 117350-29 Obsoletes: 112031-01, 112170-02, 113301-02, 113683-02,
>> 116897-01, 116901-02, 117004-01 Requires: 108528-29, 117000-05
>> Incompatibles:  Packages: FJSVhea
>>
>
> Yes. It seem's as if the admin took the FJSVhea package from a Solaris 8
> distribution and installed it on the Solaris 10 system. The package most
> probably had patch 117350-29 pre-installed - that's why the patch shows up
> as being installed. pca lists it as BAD, because 117350-29 is marked as such
> in the patchdiag.xref file (grep for "117350|29").
>
>  Obviously the SysAdmin has installed Fujitsu headers on a Sun Enterprise
>> E3500. But should this software effect the inclusion of kernel patch from
>> Solaris 8 which will not be installed, even if we tried it.
>>
>
> I guess the patch doesn't install because it requires other patches which
> are for packages that are missing on a Solaris 10 system, of course.
>
> The real question is why the admin didn't install the FJSVhea which comes
> with the Solaris 10 distribution that is installed on the machine? In that
> case, FJSVhea would be patched by the correct Solaris 10 kernel patch. pca
> (and patchadd) don't care about the OS release when judging whether a patch
> applies. If any of the packages included in a patch is installed on a
> system, the patch will apply.
>
> Hope that helps,
>
> Martin.
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://lists.univie.ac.at/mailman/private/pca/attachments/20081007/18f3ee4a/attachment.html