Brian Elliott Finley wrote:
flags="boot"/>
flags="raid" raid_dev="/dev/md0" />
flags="raid" raid_dev="/dev/md1" />
flags="-" />
flags="raid" raid_dev="/dev/md0" />
flags="raid" raid_dev="/dev/md1" />
chunk_size="32" />
Perhaps consider
Andrea,
Comments below.
Thus spake Andrea Righi ([EMAIL PROTECTED]):
oops.. the attachment... :)
Cheers,
-Andrea
Perhaps consider including the raid_dev and spare_dev info with the
other raid stuff, instead of with the disk s
oops.. the attachment... :)
Cheers,
-Andrea
Hi all,
in order to support the feature in the subject (I need it to install a
server with this particular configuration) I've done a new patch, just
committed in the trunk. With this patch I've introduced a new way to
specify the software RAID volumes inside the autoinstallscript.conf,
inste
Anton,
That sounds like a good solution, well researched, and well thought
out. Please proceed.
Thanks!
-Brian
Anton Smith wrote:
> Hi - yes I'm interested in submitting a patch.
>
> But it depends on what people think is the desired behaviour. Maybe
> taking the non columnised output is the
Thanks, Anton.
Any chance you're interested in submitting such a patch?
Cheers, -Brian
Anton Smith wrote:
> Hi guys,
>
> I have been using systemimager from CVS (a couple of nights old) for a
> bit of a play.
>
> I've noticed something odd with lvdisplay (used in Common.pm to gather
> lvm disk
Hi guys,
I have been using systemimager from CVS (a couple of nights old) for a
bit of a play.
I've noticed something odd with lvdisplay (used in Common.pm to gather
lvm disk info).
Basically lvdisplay is reporting the size as twice the value it should
be. The output even differs between c
Brian Elliott Finley wrote:
Andrea,
I see what your at here...
If you try to use the $flags variable in the sections just above, does
it not do the right thing?
sorry... I think I've not understood well what you mean...
If it does, then I think that would be a cleaner way to go.
If it doesn't, th
Andrea,
I see what your at here...
If you try to use the $flags variable in the sections just above, does
it not do the right thing?
If it does, then I think that would be a cleaner way to go.
If it doesn't, then go ahead and commit this.
And can you send an example autoinstallscript.conf && mast
Andrea,
Looks good to me. let's go with that, and modify later if necessary.
Cheers, -Brian
Andrea Righi wrote:
This is another configuration that doesn't correspond to the
autoinstallscript.conf definition... but the differences are only in the
sizes.
# sfdisk -l -uM /dev/sda
Disk /dev/sda: 4
This patch adds the commands to retrieve the lvm group that every
partition belongs (if the partition is assigned to an lvm group).
Unfortunately I've had to add the $disk argument to the function
_print_to_auto_install_conf_file()... I've not found any other way...
IMHO it could a good solutio
Dann, Dann, the kernel mann,
Comments?
Andrea Righi wrote:
I think we must enable the device mapper module for all the
architectures that are using a 2.6 kernel...
Regards,
-Andrea
--- patches/linux.ppc64.config~ 2004-10-27
This is another configuration that doesn't correspond to the
autoinstallscript.conf definition... but the differences are only in the
sizes.
# sfdisk -l -uM /dev/sda
Disk /dev/sda: 4425 cylinders, 255 heads, 63 sectors/track
Units = megabytes of 1048576 bytes, blocks of 1024 bytes, counting from
On Wed, 2005-01-26 at 10:28 -0600, Brian Elliott Finley wrote:
> Dann, Dann, the kernel mann,
I've never used LVM, so I can't say. As long as its just enabling
modules, it shouldn't put any extra strain on size.
---
This SF.Net email is spon
Andrea Righi wrote:
Hi all,
here is a new patch for the LVM stuff (made from the CVS source). I've
added also the commands to create LVM groups.
Before to continue the development I'd like to know what do you think
about an opportune syntax to express and define LVM volumes, groups,
etc... For
I've applied all of the previous patches...
Just a little note... The autoinstallscript.conf example is not
correct... I've mistaken the cut&paste... sorry... In the xml example
schema the fs definition is not necessary in the logical volume
declaration.
This is the correct one:
I think we must enable the device mapper module for all the
architectures that are using a 2.6 kernel...
Regards,
-Andrea
--
Andrea Righi
System Management Group - CINECA - http://www.cineca.it
Via Magnanelli 6/3
40033 Casalecchio di Reno (BO) - Italy
e-mail: [EMAIL PROTECTED]
--- patches/linux.p
Hi all,
here is a new patch for the LVM stuff (made from the CVS source). I've
added also the commands to create LVM groups.
Before to continue the development I'd like to know what do you think
about an opportune syntax to express and define LVM volumes, groups,
etc... For now I've tried to us
18 matches
Mail list logo
