From 1be510f9529cb082f802408b472a77d074b394c0 Mon Sep 17 00:00:00 2001 From: Nicolas Vigier Date: Sun, 14 Apr 2013 13:46:12 +0000 Subject: Add zarb MLs html archives --- zarb-ml/mageia-dev/2012-April/014459.html | 130 ++++++++++++++++++++++++++++++ 1 file changed, 130 insertions(+) create mode 100644 zarb-ml/mageia-dev/2012-April/014459.html (limited to 'zarb-ml/mageia-dev/2012-April/014459.html') diff --git a/zarb-ml/mageia-dev/2012-April/014459.html b/zarb-ml/mageia-dev/2012-April/014459.html new file mode 100644 index 000000000..ce01c8348 --- /dev/null +++ b/zarb-ml/mageia-dev/2012-April/014459.html @@ -0,0 +1,130 @@ + + + + [Mageia-dev] [soft-commits] [4017] Include persistent-storage udev rules ( needed for UUID info on root disks during install) + + + + + + + + + +

[Mageia-dev] [soft-commits] [4017] Include persistent-storage udev rules ( needed for UUID info on root disks during install)

+ Colin Guthrie + mageia at colin.guthr.ie +
+ Wed Apr 18 21:51:16 CEST 2012 +

+
+ +
'Twas brillig, and Thierry Vignaud at 18/04/12 18:17 did gyre and gimble:
+> On 11 April 2012 10:41, Colin Guthrie <gmane at colin.guthr.ie> wrote:
+>>>> Include persistent-storage udev rules (needed for UUID info on root disks
+>>>> during install)
+>>>
+>>> For the record, what does this fix?
+>>
+>> Sorry, I should have been more verbose.
+>>
+>> Basically because of the previous commit:
+>> http://svnweb.mageia.org/soft?view=revision&revision=4014
+>>
+>> We can now generate a "hostonly" initrd, i.e. a nice small, "normal"
+>> initrd that we typically produce, tailored for the system in question.
+>>
+>> As a fallout from this, when running the install with an encrypted root
+>> partition (and possibly just a normal LVM, not sure), various udev
+>> properties were not injected into the udev database. These properties
+>> (FS_UUID - can be seen via e.g. udevadm info --query=property
+>> --name=/dev/sda5) as needed by dracut to detect things properly (rather
+>> than resulting to lots of weird and wonderful probing methods dracut
+>> tries to use udev as it's primary source of info).
+>>
+>> So these rules ensured that the UUID of /dev/sd* partitions were
+>> available in udev.
+> 
+> Maybe should we include them into rescue too, shouldn't we?
+
+Yeah that makes sense IMO. Sorry I should have fixed it there too. We
+may also want to investigate whether or not  we should "fake" a dracut
+boot (i.e. mount /run and make sure the folder /run/initramfs exists).
+The only real requirement is that all the necessary info is available in
+udev to generate an initrd. I think with the rules and with udev
+starting before any lvm/crypt commands then that should be fine.
+
+
+Longer term, I think we can maybe look to use dracut itself to create a
+rescue image itself (and maybe even stage1)... but that's for another day :D
+
+Col
+
+-- 
+
+Colin Guthrie
+colin(at)mageia.org
+http://colin.guthr.ie/
+
+Day Job:
+  Tribalogic Limited http://www.tribalogic.net/
+Open Source:
+  Mageia Contributor http://www.mageia.org/
+  PulseAudio Hacker http://www.pulseaudio.org/
+  Trac Hacker http://trac.edgewall.org/
+
+ + + + + + + + + + + + + + + + + + + + + + + + +
+

+ +
+More information about the Mageia-dev +mailing list
+ -- cgit v1.2.1