> <div id="jive-html-wrapper-div">
> 
> Hi everyone.<br>
> <br>
> Sorry for the late response.&nbsp; I just saw
> this...<br>
> <br>
> Live CD isn't always a throw-away environment, as the
> booted
> environment is cpio-ed onto a target hard drive when
> installed.&nbsp;Driver
> Update is a project which takes advantage of this,
> using the DDU to
> install missing drivers in the booted environment,
> before the booted
> environment is cpio-ed to the hard drive.<br>
> <br>
> DDU (Device Driver Utility) will be enhanced to
> install SVR4 packages
> to get drivers.&nbsp; The contents file will need to
> be accurate on the hard
> drive w.r.t. these SVR4 packages.<br>
> <br>
Even with DDU, LiveCD by definition is a throw away environment.  
If for instance the system reboots the user will have to redo all the 
pkgadds. So there's no need for the contents file to consistent across
reboots.
> This thread has stated that this service runs<br>
> &nbsp;&nbsp;&nbsp; pkgadm sync -q<br>
> Seems that this command should be run in a live CD
> environment before
> doing a cpio install to the hard drive.<br>
> <br>
> This could be done when preparing the system for
> (cpio) install, or
> possibly after the DDU installs an SVR4 package on
> the live system.&nbsp;
> IMO the former is better, since that will capture all
> SVR4 packages,
> even those not installed via the DDU.<br>
> <br>
> &nbsp;&nbsp;&nbsp; Thanks,<br>
> &nbsp;&nbsp;&nbsp; Jack<br>
> <br>
> On 11/06/09 03:47, <a
> class="moz-txt-link-abbreviated"
> href="mailto:Casper.Dik at Sun.COM">Casper.Dik at Sun.COM</a
> > wrote:
> <blockquote
>  cite="mid:200911061147.nA6BlRKd036024 at dm-holland-02.u
> .sun.com"
>  type="cite">
>  <pre wrap="">
>  </pre>
>  <blockquote type="cite">
> <pre wrap="">They don't have by default, but we
> still might want to apply
> SVr4 packages to the booted install environment -
> please see below.
>     </pre>
> blockquote>
>   <pre wrap=""><!---->
>  </pre>
>  <blockquote type="cite">
> <pre wrap="">[3] Finished DC build and booted
>  resulting AI ISO image - it contains 
> ock files:
> 
> $ ls -la /var/sadm/install/
> total 18
> drwxr-xr-x   4 root     bin          512 Nov  6 01:08
> .
> drwxr-xr-x   8 root     sys          512 Nov  6 00:54
> ..
> -rw-r--r--   1 root     root           0 Nov  6 00:47
> .door
> lrwxrwxrwx   1 root     root          36 Nov  6 00:58
> .lockfile -&gt; 
> /mnt/misc/var/sadm/install/.lockfile
> lrwxrwxrwx   1 root     root          36 Nov  6 00:58
> .pkg.lock -&gt; 
> /mnt/misc/var/sadm/install/.pkg.lock
> lrwxrwxrwx   1 root     root          43 Nov  6 00:58
> .pkg.lock.client 
> -&gt; /mnt/misc/var/sadm/install/.pkg.lock.client
> drwxr-xr-x   2 root     bin          512 Nov  6 01:08
> admin
> lrwxrwxrwx   1 root     root          35 Nov  6 00:58
> contents -&gt; 
> /mnt/misc/var/sadm/install/contents
> lrwxrwxrwx   1 root     root          43 Nov  6 00:58
> gz-only-packages 
> -&gt; /mnt/misc/var/sadm/install/gz-only-packages
> dr-xr-xr-x   2 root     bin          512 Nov  6 00:47
> logs
> 
> However, those are just symbolic links to the
> read-only area:
>     </pre>
> blockquote>
>   <pre wrap=""><!---->
> /var/sadm/install is readable but /tmp/misc is not
>  readable?
> 
>   </pre>
> <blockquote type="cite">
>     <pre wrap=""># mount -p
> /tmp/solarismisc.zlib - /mnt/misc hsfs - no 
> ro,nosuid,noglobal,maplcase,rr,traildot
> ...
> 
> This leads to the situation when pkgserv SMF service
> goes
> sometimes to the maintenance mode during boot, but
> not always.
> 
> If I remove those 'dot' files from DC proto area and
> they don't end up
>     </pre>
> blockquote>
> <pre wrap=""><!---->in the final ISO imageimage,
>  pkgserv is allowed to manipulate them,
>  </pre>
>  <blockquote type="cite">
> <pre wrap="">since /var/sadm/install/ directory is
> writable:
>     </pre>
> blockquote>
>   <pre wrap=""><!---->
> o who creates the symbolic links?  I think we should
> not put the symbolic
> links there.
> 
> 
> 
>   </pre>
> <blockquote type="cite">
> <pre wrap="">Assuming that we still would like to
>  support installation of SVr4 packages
> uring the installation process into running
> installation environment as 
> well
> as on target (e.g. in order to deliver missing third
> party device driver
> which is in SVr4 format), can we still go with
> pkgserv SMF service disabled
> in the installation environment or is there any
> reason why it should be
> enabled ?
>     </pre>
> blockquote>
>   <pre wrap=""><!---->
> reason for the LiveCD &amp; AI-install; these are
>  really "throw away"
> nvironments.  You will need to have a readable
> /var/sadm/install 
> directory if you want to install packages, though.
> 
> Casper
> 
> _______________________________________________
> caiman-discuss mailing list
> <a class="moz-txt-link-abbreviated"
> href="mailto:caiman-discuss at opensolaris.org">caiman-di
> scuss at opensolaris.org</a>
> <a class="moz-txt-link-freetext"
> href="http://mail.opensolaris.org/mailman/listinfo/cai
> man-discuss">http://mail.opensolaris.org/mailman/listi
> nfo/caiman-discuss</a>
>   </pre>
> blockquote>
> <br>
> 
> </div>_______________________________________________
> caiman-discuss mailing list
> caiman-discuss at opensolaris.org
> http://mail.opensolaris.org/mailman/listinfo/caiman-di
> scuss
-- 
This message posted from opensolaris.org

Reply via email to