Skip to contentSkip to conversation

2.6.19-rc5: Can't get built-in raid support, modular works correctly

DISCUSSION J. Hart 2 participants 5 messages Jul 19, 2007 – Jul 20, 2007 MD RAID

lore
Share & tools
lore

J. Hart

author
#0 lore raw

When I configure a 2.6.19-rc5 linux kernel for "built-in" raid support, I do not get the expected /proc/mdstat entry. I set the following kernel parameters for this :

CONFIG_MD=Y
BLK_DEV_MD=y
MD_RAID0=y

When I configure the kernel for modular raid support in otherwise identical fashion, I do get the expected /proc/mdstat "file". I have no difficulty configuring "built-in" support under kernel 2.6.13 on another 32 bit X86 based system with IDE drives. The system I am trying to set up raid support for is X86_64 based, and has only SATA drives. I have not configured IDE support for it. Any ideas would be greatly appreciated.

#1 lore raw
Quoted 6 lines · J. Hart

When I configure a 2.6.19-rc5 linux kernel for "built-in" raid support, I do not get the expected /proc/mdstat entry. I set the following kernel parameters for this :

CONFIG_MD=Y
BLK_DEV_MD=y
MD_RAID0=y

Quoted 7 lines

When I configure the kernel for modular raid support in otherwise identical fashion, I do get the expected /proc/mdstat "file". I have no difficulty configuring "built-in" support under kernel 2.6.13 on another 32 bit X86 based system with IDE drives. The system I am trying to set up raid support for is X86_64 based, and has only SATA drives. I have not configured IDE support for it. Any ideas would be greatly appreciated.

-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Any reason you are using 2.6.19-rc5? Why not use 2.6.22.(1)?

Justin.

author
#2 lore raw
Quoted 2 lines · Justin Piszcz

Any reason you are using 2.6.19-rc5? Why not use 2.6.22.(1)?

I just wanted to try to understand the reason for the problem before changing to a new kernel. I had not heard that any such problem had been encountered, though I could have missed the news. Have you heard anything about it ?

J. Hart

#3 lore raw
Quoted 8 lines · J. Hart
Quoted 2 lines · Justin Piszcz

Any reason you are using 2.6.19-rc5? Why not use 2.6.22.(1)?

I just wanted to try to understand the reason for the problem before changing to a new kernel. I had not heard that any such problem had been encountered, though I could have missed the news. Have you heard anything about it ?

Quoted 7 lines

J. Hart

-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

I (normally) do not run -rcX release and I always compile in RAID support and have not seen that issue; then again, I did not try 2.6.19-rc5 that I can remember.

Justin.

author
#4 lore raw
Quoted 1 line · Justin Piszcz

I (normally) do not run -rcX release and I always compile in RAID support and have not seen that issue; then again, I did not try 2.6.19-rc5 that I can remember.

I've just discovered the problem (at 1 am of course).

I'm using Linux with my own custom kernel running on a quad Mac Pro (no Parallels, no Bootcamp, no Windows, no OS X etc, and hence no partition restrictions.....:-). Since this is an Intel Mac, running grub used to be a bit tricky owing to the A20 line problem, so I've been using lilo instead. After building the kernel, I dutifully typed the following, as any good lilo believing kernel builder should:

lilo

This ended up placing the boot block on the root partition. I had previously told the Mac EFI boot loader to look for it on the MBR of my first SATA drive, which in this case is a very different location. What I should have used was:

lilo -b /dev/sda

It was booting the old kernel of course, which was the same version release as the rebuilt one.

(....slapping forehead.....)

All is working just as expected now.....:-)

Many Thanks for your comments........