Public SourceMage Discussion List

[Eric Sandall <eric AT sandall.us>]

Jeremy Blosser wrote:
<snip>
> This will BREAK systems which are in currently running in *production*
> environments for some of our users.  And it will break them in one of the
> worst ways possible -- they won't have any obvious indication things are
> broken until the next time their box reboots and fails to get past init.
> Whether we claim to support LVM before 1.0 or not, this is a non-option.
> 
> If no one has time to resolve the LVM issues in test, we should just
> backport the udevstart patch by itself.

Are the following udev commits okay?

commit fe9c8a2a68938efae95aed26c1ebb6c61d7c72ff
Author: Flavien Bridault <vlaaad AT sourcemage.org>
Date:   Sun Jun 18 00:41:34 2006 +0200

    udev updated to 094 for kernels > 2.6.14. We will now manage two
versions in
    order to avoid breaking boxes with old kernels. Lots of changes, please
    read HISTORY ;-)

commit a70f57e36897dcfe6661e88977635a6bd2cd410d
Author: Flavien Bridault <vlaaad AT sourcemage.org>
Date:   Sun Jun 18 00:37:45 2006 +0200

    Removed udevstart for kernel > 2.6.14, coldplug now works well
thanks to ude

If so, I'll just revert all of the other commit ids back to
5b866bfe90463d6d30e8f9de0193ae3d33519da1 (sync with stable-rc-0.4 from
Perforce), which *should* be enough, but please let me know.

We'll need the following commit to have init.d working with the current
kbd/console-tools in stable-rc-0.6:
commit ba0befcef1651a90112c04f55c2a08d640857508
Author: Juuso Alasuutari <iuso AT sourcemage.org>
Date:   Thu Jul 6 07:24:30 2006 +0300

    Removed keymap.sh & keymap.conf from init.d spell and moved to kbd &
    console-tools spells.

And will probably want these:
commit 58cd6000480b905de3950386cb45cdf1f02db388
Author: Juuso Alasuutari <iuso AT sourcemage.org>
Date:   Tue Jul 11 21:58:36 2006 +0300

    Changed exit 1's to return 1's, bug #12838.

commit 0d22246a175bc62dd371c6d7e222f8c22624188e
Author: Andra<C5><BE> Levstik <ruskie AT mages.ath.cx>
Date:   Tue Sep 5 12:28:21 2006 +0200

    init.d: unlink any symlinks in /var/run before removing stuff

commit 6b9eb857bd64e745c1dfc4837d6bce7746b990de
Author: Andra<C5><BE> Levstik <ruskie AT mages.ath.cx>
Date:   Tue Sep 5 19:40:48 2006 +0200

    init.d: changed wrong var

commit a00f744269041025eb0f84f6bb4c51320680b6ec
Author: Juuso Alasuutari <iuso AT sourcemage.org>
Date:   Thu Sep 21 02:20:27 2006 +0300

    init.d: [automated] Removed BUILD_API=2.

commit 95f13eb1ac5a9dc00b73497abf8ed6c792b0d0c7
Author: Jaka Kranjc <lynxlynxlynx AT sourcemage.org>
Date:   Sat Oct 21 10:53:15 2006 +0200

    init.d: fix #7311, proc not showing in /etc/mtab

commit 1b1ec0fd66472cd4e6caa69238c4a7fe0407bdb7
Author: Jaka Kranjc <lynxlynxlynx AT sourcemage.org>
Date:   Thu Oct 26 21:38:41 2006 +0200

    use TRACK_ROOT when linking #9163

commit 285e2024c765964cabf3ce49d41e78bcba36af99
Author: Thomas Orgis <sobukus AT sourcemage.org>
Date:   Mon Sep 4 00:43:18 2006 +0200

    bumping patchlevel for init.d

commit 557bc9028c2705ae3d65b7973234a8ed03a5dfd7
Author: Thomas Orgis <sobukus AT sourcemage.org>
Date:   Sun Sep 3 13:01:31 2006 +0200

    fixes to init.d according to bug #12964

commit c252edf200734efe6e77745401b81de10f8517d4
Author: Thomas Orgis <sobukus AT sourcemage.org>
Date:   Tue Sep 5 21:53:31 2006 +0200

    better question for static udev in init.d

OR, a simpler approach. ;)

Sync up with test, and revert these:
commit adbab1098ff1a2dd9f1eff2882cb6b47c1066ca7
Author: Thomas Orgis <sobukus AT sourcemage.org>
Date:   Thu Nov 2 20:12:10 2006 +0100

    dmsetup in init.d/mountall.sh failing is not fatal (simply no dm in
kernel),

commit 772f276d53fecceaf68441af953afc041887da62
Author: Thomas Orgis <sobukus AT sourcemage.org>
Date:   Sun Oct 29 14:39:03 2006 +0100

    enable device mapper nodes discovery in init.d/mountall.sh

Assuming those are the two changes that break LVM.

-sandalle

-- 
Eric Sandall                     |  Source Mage GNU/Linux Developer
eric AT sandall.us                  |  http://www.sourcemage.org/
http://eric.sandall.us/          |  SysAdmin @ Shock Physics @ WSU
http://counter.li.org/  #196285  |  http://www.shock.wsu.edu/