Mary, yes, I did create that shell script as well and replaced /sbin/ biosdev. But putting it back in place after every upgrade got annoying. So creating different fdisk tables was the more suitable permanent "fix" as there is no BIOS for my motherboard (GA-K8NF-9) that does EDD - as far as I can tell anyway.
You missed listing the fdisk partition table change that as 3. option :) - mo (must chose a better motherboard NeXT time round ...) On 22 Mar 2007, at 00:56, mary ding wrote: > Moritz and Matty: > > This issue had came up before and Vikram Hegde had the following > response: > > > 1. First run biosdev -d and get the output. > > i.e: > > solaris-devx# biosdev -d > adding /pci at 0,0/pci-ide at 1f,2/ide at 0/cmdk at 0,0 > adding /pci at 0,0/pci-ide at 1f,2/ide at 1/cmdk at 0,0 > matching biosdev-0x80 > match_edd bdf 0 31 2 > interface type ATA pci channel 0 target 0 > matched thru edd > 0x80 /pci at 0,0/pci-ide at 1f,2/ide at 0/cmdk at 0,0 > matching biosdev-0x81 > match_edd bdf 0 31 2 > interface type ATA pci channel 0 target 0 > matched thru edd > 0x81 /pci at 0,0/pci-ide at 1f,2/ide at 0/cmdk at 0,0 > 0x80 /pci at 0,0/pci-ide at 1f,2/ide at 0/cmdk at 0,0 matchcount 2 > 0x81 /pci at 0,0/pci-ide at 1f,2/ide at 0/cmdk at 0,0 matchcount 2 > biosdev: Could not match any!! > > 2. > > In the above case, this appears to be a BIOS bug which prevents > biosdev from figuring out the BIOS disk ID. There is an RFE filed > > 6513832 biosdev should try first block match if BIOS EDD info is buggy > > that will help reduce the impact of such BIOS bugs. > > In the meantime, you have a couple of options > > 1. Upgrade your BIOS to see if that fixes the biosdev problem. > > or > > 2. Create an executable shell script like the following > > #!/bin/sh > echo 0x80 /pci at 0,0/pci-ide at 1f,2/ide at 0/cmdk at 0,0 > echo 0x81 /pci at 0,0/pci-ide at 1f,2/ide at 1/cmdk at 0,0 > > and replace /sbin/biosdev with this shell script. > > Moritz Willers wrote: >> I'm slightly confused on how many disks you have in the system, >> but ... when I run into this a year ago this helped me: http:// >> bugs.opensolaris.org/view_bug.do?bug_id=6362108 (biosdev fails >> with "biosdev: Could not match any!!"). >> My board didn't do "EDD" (I think, whatever that is) and biosdev >> had to resort to comparing the first block on the disk as read >> from the disk against what the bios says the first block is. >> Both my disks had the same first block, so it couldn't tell the >> difference between them from the first block. >> Once I changed the fdisk layout to be different on the two >> biosdev was able to tell the difference, but then I run into >> other problems ... But now I'm happily using LU with alternating >> BE on two partitions on the same disk. >> good luck! - mo >> On 18 Mar 2007, at 16:01, Matty wrote: >>> Howdy, >>> >>> I am trying to activate an ABE on an x86 server, and I keep >>> getting the following error: >>> >>> $ luactivate Solaris10_11_06_B >>> biosdev: Could not match any!! >>> ERROR: Device mapping command </sbin/biosdev> failed. Please >>> reboot and try again. >>> ERROR: GRUB slice lookup failed. >>> ERROR: Cannot determine GRUB slice. >>> >>> When I run biosdev with the debug option, I get the following: >>> >>> $ /sbin/biosdev -d >>> adding /pci at 0,0/pci1000,30 at 10/sd at 0,0 >>> adding /pci at 0,0/pci1000,30 at 10/sd at 1,0 >>> adding /pci at 0,0/pci1000,30 at 11/sd at 0,0 >>> adding /pci at 0,0/pci1000,30 at 11/sd at 1,0 >>> adding /pci at 0,0/pci1000,30 at 11/sd at 2,0 >>> adding /pci at 0,0/pci1000,30 at 11/sd at 3,0 >>> adding /pci at 0,0/pci1000,30 at 11/sd at 4,0 >>> adding /pci at 0,0/pci1000,30 at 11/sd at 5,0 >>> adding /pci at 0,0/pci1000,30 at 11/sd at 6,0 >>> adding /pci at 0,0/pci1000,30 at 11/sd at 8,0 >>> matching biosdev-0x80 >>> magic not valid 0 pathinfolen 0 >>> matched first block >>> 0x80 /pci at 0,0/pci1000,30 at 10/sd at 0,0 >>> matching biosdev-0x81 >>> magic not valid 0 pathinfolen 0 >>> matched first block >>> 0x81 /pci at 0,0/pci1000,30 at 11/sd at 0,0 >>> matching biosdev-0x82 >>> magic not valid 0 pathinfolen 0 >>> matched first block >>> 0x82 /pci at 0,0/pci1000,30 at 11/sd at 0,0 >>> matching biosdev-0x83 >>> magic not valid 0 pathinfolen 0 >>> matched first block >>> 0x83 /pci at 0,0/pci1000,30 at 11/sd at 0,0 >>> matching biosdev-0x84 >>> magic not valid 0 pathinfolen 0 >>> matched first block >>> 0x84 /pci at 0,0/pci1000,30 at 11/sd at 0,0 >>> matching biosdev-0x85 >>> magic not valid 0 pathinfolen 0 >>> matched first block >>> 0x85 /pci at 0,0/pci1000,30 at 11/sd at 0,0 >>> matching biosdev-0x86 >>> magic not valid 0 pathinfolen 0 >>> matched first block >>> 0x86 /pci at 0,0/pci1000,30 at 11/sd at 0,0 >>> matching biosdev-0x87 >>> magic not valid 0 pathinfolen 0 >>> matched first block >>> 0x87 /pci at 0,0/pci1000,30 at 10/sd at 0,0 >>> 0x80 /pci at 0,0/pci1000,30 at 10/sd at 0,0 matchcount 2 >>> 0x81 /pci at 0,0/pci1000,30 at 11/sd at 0,0 matchcount 6 >>> 0x82 /pci at 0,0/pci1000,30 at 11/sd at 0,0 matchcount 6 >>> 0x83 /pci at 0,0/pci1000,30 at 11/sd at 0,0 matchcount 6 >>> 0x84 /pci at 0,0/pci1000,30 at 11/sd at 0,0 matchcount 6 >>> 0x85 /pci at 0,0/pci1000,30 at 11/sd at 0,0 matchcount 6 >>> 0x86 /pci at 0,0/pci1000,30 at 11/sd at 0,0 matchcount 6 >>> 0x87 /pci at 0,0/pci1000,30 at 10/sd at 0,0 matchcount 2 >>> biosdev: Could not match any!! >>> >>> Does anyone happen to know what might be causing the luactivate >>> to fail? >>> >>> Thanks, >>> - Ryan >>> >>> >>> This message posted from opensolaris.org >>> _______________________________________________ >>> install-discuss mailing list >>> install-discuss at opensolaris.org >>> http://opensolaris.org/mailman/listinfo/install-discuss >> --------------------------------------------------------------------- >> --- >> _______________________________________________ >> install-discuss mailing list >> install-discuss at opensolaris.org >> http://opensolaris.org/mailman/listinfo/install-discuss > > > -- > ------------------------------------------------------------------- > _/_/_/ _/ _/ _/ _/ Mary Ding > _/ _/ _/ _/_/ _/ Member of Technical Staff > _/_/_/ _/ _/ _/ _/ _/ UMPK17-307 > _/ _/ _/ _/ _/_/ Menlo Park, CA. 94043 > _/_/_/ _/_/_/ _/ _/ mary.ding at Eng.Sun.COM > 650-786-8317 (phone) > M I C R O S Y S T E M S > -------------------------------------------------------------------- -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 3452 bytes Desc: not available URL: <http://mail.opensolaris.org/pipermail/install-discuss/attachments/20070322/fa8251d8/attachment.bin>
