Subject: Problems using RAIDframe
To: None <current-users@netbsd.org>
From: Tom Ivar Helbekkmo <tih@eunetnorge.no>
List: current-users
Date: 09/17/2003 20:20:11
I'm running a very current current on my home Intel system, and
decided to rebuild it to use RAIDframe for everything -- I had enough
extra disks lying around, and I see that the boot blocks now check for
the extra 64 sectors offset in case there's a RAID 1 set.  After a
couple of false starts (like not realizing that a 'RAID' partition
can't start at the start of a disk), I think I'm figuring it out, but
I'm not certain I'm doing it right.  If someone could explain how
fdisk, disklabel and raidctl should be used to build a properly
bootable RAID 1 set these days, that would be great!

Meanwhile, I also need help understanding what happens when I try to
build a RAID 0 set, as follows:

*** Boot time messages:

sd4 at scsibus2 target 0 lun 0: <SEAGATE, ST39102LW, 0005> disk fixed
sd4: 8683 MB, 6962 cyl, 12 head, 212 sec, 512 bytes/sect x 17783240 sectors
sd4: sync (50.00ns offset 8), 16-bit (40.000MB/s) transfers, tagged queueing
sd5 at scsibus2 target 1 lun 0: <SEAGATE, ST39173W, 5958> disk fixed
sd5: 8683 MB, 7501 cyl, 10 head, 237 sec, 512 bytes/sect x 17783240 sectors
sd5: sync (50.00ns offset 8), 16-bit (40.000MB/s) transfers, tagged queueing
sd6 at scsibus2 target 2 lun 0: <SEAGATE, ST39102LW, 0005> disk fixed
sd6: 8683 MB, 6962 cyl, 12 head, 212 sec, 512 bytes/sect x 17783240 sectors
sd6: sync (50.00ns offset 8), 16-bit (40.000MB/s) transfers, tagged queueing
sd7 at scsibus2 target 3 lun 0: <SEAGATE, ST39173W, 4290> disk fixed
sd7: 8683 MB, 7501 cyl, 10 head, 237 sec, 512 bytes/sect x 17783240 sectors
sd7: sync (50.00ns offset 8), 16-bit (40.000MB/s) transfers, tagged queueing

*** fdisk setup for sd4/5/6/7:

Disk: /dev/rsd4d
NetBSD disklabel disk geometry:
cylinders: 6962, heads: 12, sectors/track: 212 (2544 sectors/cylinder)
total sectors: 17783240

BIOS disk geometry:
cylinders: 1023, heads: 255, sectors/track: 63 (16065 sectors/cylinder)
total sectors: 17783240

Partition table:
0: NetBSD (sysid 169)
    start 63, size 17783177 (8683 MB, Cyls 0-1106/243/42)
1: <UNUSED>
2: <UNUSED>
3: <UNUSED>
Bootselector disabled.

*** Disklabel for sd4/5/6/7:

5 partitions:
#        size    offset     fstype [fsize bsize cpg/sgs]
 a:  17783177        63       RAID                     # (Cyl.    0*-   6990*)
 c:  17783177        63     unused      0     0        # (Cyl.    0*-   6990*)
 d:  17783240         0     unused      0     0        # (Cyl.    0 -   6990*)


*** raidctl input file for building the set:

START array
1 4 0
START disks
/dev/sd4a
/dev/sd5a
/dev/sd6a
/dev/sd7a
START layout
64 1 1 0
START queue
fifo 128

*** disklabel for the RAID set:

4 partitions:
#        size    offset     fstype [fsize bsize cpg/sgs]
 a:  71132416         0     4.2BSD      0     0     0  # (Cyl.    0 -  17366*)
 d:  71132416         0     unused      0     0        # (Cyl.    0 -  17366*)

*** And then, I try to make a file system on it:

# newfs raid4a
/dev/rraid4a: 34732.6MB (71132416 sectors) block size 16384, fragment size 2048
        using 189 cylinder groups of 183.78MB, 11762 blks, 23040 inodes.
wtfs: write error for sector 160: Invalid argument

...and I'm stuck.  :-(  Anyone see something here that I don't?

-tih
-- 
Tom Ivar Helbekkmo, Senior System Administrator, EUnet Norway
www.eunet.no  T: +47-22092958 M: +47-93013940 F: +47-22092901