Hi all !
I would like to ask about problems with ATA RAID. Thank's to all very much for
compliance.
HW: Intel BX mainboard, HotRod PCi card (with HPT370 chip with 2.0.0919 BIOS) +
2x Seagate ATA100 60 GB hdd (ST360020A) as masters on first + second channels,
PIII550@733 CPU, 512 MB RAM/133 MHz
SW : RedHat 7.2 + packages from RH RawHide, kernel 2.4.16-0.9
Problems :
a} When I set up HPT chip only as IDE controler, both disks works fine. I have in
/etc/sysconfig/hardisks* parameters for hdpam (-m16 -c3 -u1 -d1 -a128 -X69). Both
disks get this params and works without error reports
b) after creating RAID0 array on this two 60Gb disks system slows down while root
filesystems is remounting RO -> RW (on refered RIAD array is only data volume - root
filesystem is on another single disk). In this time is both bog disks only inicialised, but
filesystem on it is NOT mounted and hd-paramsis NOT set yet.
In this time syslog reports same errors into /var/log/messages log too :
hde: read_intr: error=0x10 { SectorIdNotFound }, LBAsect=204877970, sector=6
hde: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }
(repeated many times).
After system loading is all OK, raid drivers can be loaded (without any message) and
volume on RAID is available, but _very_ slow - kernel compilation is faster on root
filesystem (older Seagate ST38410A disk connected on iBX chipset in UDMA/2
mode).
I detect that first disk in array (in this case primary on first channel = hde) haven't any
settings and works into PIO0 mode ?!?
c) after deleting RAID0 array and re-creating in reverse order (first disk in array is
master on second channel == hdg) I got similar error messages, but system reports
errors on hdg disk.
Questions :
1) How can I supress error messages about errors on first disk in RAID0 array ?
2} How can I set up disk in array into fastest availaible mode ? Can I set up params on
both disks via /dev/hde + /dev/hdg ?
3) How can I set up modules in /etc/modules.conf config file ? I use this commands :
alias block-major-114 ataraid
above ataraid hptraid
Is this correct and accurate ?
4) Both disks in array reported different geometry (after RAID0 array created) :
first disk in array :
/dev/hdg:
multcount = 0 (off)
I/O support = 0 (default 16-bit)
unmaskirq = 0 (off)
using_dma = 0 (off)
keepsettings = 0 (off)
nowerr = 0 (off)
readonly = 0 (off)
readahead = 128 (on)
geometry = 7297/255/63, sectors = 117231408, start = 0
second disk in array :
/dev/hde:
multcount = 16 (on)
I/O support = 3 (32-bit w/sync)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 0 (off)
nowerr = 0 (off)
readonly = 0 (off)
readahead = 128 (on)
geometry = 116301/16/63, sectors = 117231408, start = 0
Gemetry of RAID0 array is :
Disk /dev/ataraid/d0: 255 heads, 63 sectors, 14594 cylinders
Units = cylinders of 16065 * 512 bytes
Is this correct ?
5) After repartitioning (fdisk from util-linux-2.11f-16) + writing PTBL I got this
message :
Command (m for help): w
Tha parition table has been altered !
Calling ioctl() to re-read partiton table.
Re-read failed with error 22: Invalid argument.
Reboot your system to ensure tha partition table is updated.
When I probe format partiton I got this message :
# mkfs -j -m 1 -L big /dev/ataraid/d0p2
mkefs 1.25 (20-Sep-2001)
mkfs.ext2: Device size reported to be zero. Invalid parition specified, or
partiton table wasn't reread after running fdisk, due to
a modified partition being busy and in use. You may need to reboot
to re-read your partition table.
Only way is re-loading ataraid + hptraid modules before formatting. Is this feature or
bug ?
Thank's very much for any answer.
Miroslav BENES
system administrator
TENEZ a.s.
Chotebor
Czech republic
www.tenez.cz
