Sorry about the crossposting.
I wrote on the Yellow Dog Linux list when somebody asked about software
RAID on YDL about my experiences with it:
quoted
The one really big gotcha is that the Macintosh partitioning scheme
can't tell the Linux kernel that certain partitions are to be
considered "Linux RAID autodetect" (as in x86 using the DOS partition
table type 0xfd). This means that you can't boot a Mac Linux system
directly from RAID because the kernel won't be able to autostart the
RAID devices. You have to work around this by creating an initial RAM
disk that uses the raidstart command to start your metadevices, then
swaps the initrd out of the way and proceeds to start the real system.
to which Tim Seufert replied on the same list:
Hmmm. That would seem to be a lack in the Linux RAID code, since the
Macintosh partition table has a vastly more flexible partition type
field than DOS: instead of a single byte it's a string. It would mean
breaking from the convention of using the "Apple_SVR2_UNIX" type for
Linux partitions, but that really is just a convention as far as I know.
Perhaps the PPC Linux developers and the Linux RAID developers should
get together on this and make some decisions so as to make it happen.
--
Atro Tossavainen (Mr.) / The Institute of Biotechnology at
Systems Analyst, Techno-Amish & / the University of Helsinki, Finland,
+358-9-19158939 UNIX Dinosaur / employs me, but my opinions are my own.
< URL : http : / / www . helsinki . fi / %7E atossava / > NO FILE ATTACHMENTS
** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/
From: Sven Luther <hidden> Date: 2004-01-27 09:32:13
On Tue, Jan 27, 2004 at 10:01:17AM +0200, Atro Tossavainen wrote:
Sorry about the crossposting.
I wrote on the Yellow Dog Linux list when somebody asked about software
RAID on YDL about my experiences with it:
quoted
quoted
The one really big gotcha is that the Macintosh partitioning scheme
can't tell the Linux kernel that certain partitions are to be
considered "Linux RAID autodetect" (as in x86 using the DOS partition
table type 0xfd). This means that you can't boot a Mac Linux system
directly from RAID because the kernel won't be able to autostart the
RAID devices. You have to work around this by creating an initial RAM
disk that uses the raidstart command to start your metadevices, then
swaps the initrd out of the way and proceeds to start the real system.
to which Tim Seufert replied on the same list:
quoted
Hmmm. That would seem to be a lack in the Linux RAID code, since the
Macintosh partition table has a vastly more flexible partition type
field than DOS: instead of a single byte it's a string. It would mean
breaking from the convention of using the "Apple_SVR2_UNIX" type for
Linux partitions, but that really is just a convention as far as I know.
Perhaps the PPC Linux developers and the Linux RAID developers should
get together on this and make some decisions so as to make it happen.
Seems ok for me. Also, i guess that there are other partition types,
like the amiga partitition table the pegasos boxes mostly use, which has
a 32bit identifier for partition types. I guess it is the task of the
RAID code to have some per partition type checking for this RAID autodetect
magic.
Friendly,
Sven Luther
** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/
On Tue, Jan 27, 2004 at 10:32:13AM +0100, Sven Luther wrote:
On Tue, Jan 27, 2004 at 10:01:17AM +0200, Atro Tossavainen wrote:
quoted
Sorry about the crossposting.
I wrote on the Yellow Dog Linux list when somebody asked about software
RAID on YDL about my experiences with it:
quoted
quoted
The one really big gotcha is that the Macintosh partitioning scheme
can't tell the Linux kernel that certain partitions are to be
considered "Linux RAID autodetect" (as in x86 using the DOS partition
table type 0xfd). This means that you can't boot a Mac Linux system
directly from RAID because the kernel won't be able to autostart the
RAID devices. You have to work around this by creating an initial RAM
disk that uses the raidstart command to start your metadevices, then
swaps the initrd out of the way and proceeds to start the real system.
to which Tim Seufert replied on the same list:
quoted
Hmmm. That would seem to be a lack in the Linux RAID code, since the
Macintosh partition table has a vastly more flexible partition type
field than DOS: instead of a single byte it's a string. It would mean
breaking from the convention of using the "Apple_SVR2_UNIX" type for
Linux partitions, but that really is just a convention as far as I know.
Perhaps the PPC Linux developers and the Linux RAID developers should
get together on this and make some decisions so as to make it happen.
Seems ok for me. Also, i guess that there are other partition types,
like the amiga partitition table the pegasos boxes mostly use, which has
a 32bit identifier for partition types. I guess it is the task of the
RAID code to have some per partition type checking for this RAID autodetect
magic.
in general we need to stop using Apple_* types for our partitions,
apple clearly documents that the entire Apple_* namespace is reserved
solely for Apple use, and 3rd party developers must use something,
anything else for thier partitions.
the trick is coming up with a generic enough and sensible enough
naming convention. (i think its quite silly to include specific
filesystem names in partition types, and really it seems silly to
specify "Linux" for filesystems which can and are used by OSes other
then "Linux" (ext2 is used in Hurd for example)).
--
Ethan Benson
http://www.alaska.net/~erbenson/
** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/
On Tue, Jan 27, 2004 at 10:32:13AM +0100, Sven Luther wrote:
Seems ok for me. Also, i guess that there are other partition types,
like the amiga partitition table the pegasos boxes mostly use, which has
a 32bit identifier for partition types. I guess it is the task of the
RAID code to have some per partition type checking for this RAID autodetect
magic.
it's done in fs/partitions/. it could be made anything, as long as it's put
in the raid docs so people know.
here's a patch of mine for alpha, to give you an idea.
--
Tom Vier [off-list ref]
DSA Key ID 0xE6CB97DA
From: Sven Luther <hidden> Date: 2004-01-29 07:22:22
On Wed, Jan 28, 2004 at 08:01:01PM -0500, Tom Vier wrote:
On Tue, Jan 27, 2004 at 10:32:13AM +0100, Sven Luther wrote:
quoted
Seems ok for me. Also, i guess that there are other partition types,
like the amiga partitition table the pegasos boxes mostly use, which has
a 32bit identifier for partition types. I guess it is the task of the
RAID code to have some per partition type checking for this RAID autodetect
magic.
it's done in fs/partitions/. it could be made anything, as long as it's put
in the raid docs so people know.
Ok, thanks. On amiga filesystem, partitition types are in 32bit ints,
but can also be represented as 4 chars, so maybe `RAID` would be a good
type for that ? I will try patching the kernel, and adding support in
libparted for recognizing such partitions, and try it out.
Does it make sense to do Software Raid on two partititons of the same
disk, just for testing purpose ?
here's a patch of mine for alpha, to give you an idea.
Ok, thanks.
Friendly,
Sven Luther
** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/