dmraid and promise fasttrak controller

[Heinz Mauelshagen]

On Fri, Jan 14, 2005 at 06:53:14PM +0100, Dick Marinus wrote:
> > This looks like bit 8 set in raid.status of the Promise metadata
> > (see dmraid -n output), which is used to indicate a broken mirror in
> > dmraid.
> Yes, I came to the same conclusion, I've hacked the source to 
> ((pdc)->raid.status & 0x00) 
> so it's always 0 :)

:)

Anyone out there to give a hand about status code definitions
to drill this one down ?

> After this dmraid correctly detected it.
> 
> > What does your Promise BIOS tell about the array status ?
> It tells me it's functional, and pdcraid from 2.4 is also working fine
> 
> Though...
> 
> After a initrd nightmare I've got a /dev/mapper/pdc_jjajccdd and partitions, 
> but only the first three partitions mounted fine. I couldn't mount the last 
> partition :-(

Flaw in lib/format/partition/dos.c.
Maybe you can drill this down and come up with a fix ?

> 
> During the initialisation from dmraid the kernel complained about an incorrect 
> size, it looked like the last block device was zero sized.
> 
> While hacking dmraid's code I was really amazed about the clean code! Keep up 
> the good work :-)

Thanks,
Heinz

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

Heinz Mauelshagen                                 Red Hat GmbH
Consulting Development Engineer                   Am Sonnenhang 11
                                                  56242 Marienrachdorf
                                                  Germany
Mauelshagen at RedHat.com                            +49 2626 141200
                                                       FAX 924446
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-