Hi,
We have some problems with one IDE-RAID machine, however
the problem seem not to be with the RAID setup, actually the ide sub
system.
Basic system information:
=========================
AMD k6-2 with onboard ALI15X3 IDE (disabled)
2*PCI HPT366 IDE cards
PCI SCSI
PCI NIC
4*Maxtor 93652U8 (as master, one on each channel) on the HPT
controller.
The 4 maxtor disk are used for the RAID 5 (softraid)
Problem description:
====================
With DMA disabled everything is _stable_ but the performance is _bad_:
fileserver:~# hdparm -tT /dev/md0
/dev/md0:
Timing buffer-cache reads: 128 MB in 2.64 seconds = 48.48 MB/sec
Timing buffered disk reads: 64 MB in 19.12 seconds = 3.35 MB/sec
With DMA enabled, performance is a better ~10MB/sec but the system is
very unstable. The system locks within seconds or minutes after the
boot.
The message printed before the system locks is something like this:
hdx: timeout wainting for DMA ide_dmaproc: chipset supported ide_1
dma_timeout func only = 14 (where x is one of the four harddisks)
Anyone who has a clue about this? Below is some additional
information.
Best regards,
Peter
dmesg:
******
Uniform Multi-Platform E-IDE driver Revision: 6.31
ide: Assuming 33MHz system bus speed for PIO modes; override with
idebus=xx
HPT366: IDE controller on PCI bus 00 dev 90
PCI: Enabling device 00:12.0 (0005 -> 0007)
PCI: Found IRQ 5 for device 00:12.0
PCI: The same IRQ used for device 00:12.1
HPT366: chipset revision 1
HPT366: not 100% native mode: will probe irqs later
ide0: BM-DMA at 0xd000-0xd007, BIOS settings: hda:pio, hdb:pio
HPT366: IDE controller on PCI bus 00 dev 91
PCI: Found IRQ 5 for device 00:12.1
PCI: The same IRQ used for device 00:12.0
HPT366: chipset revision 1
HPT366: not 100% native mode: will probe irqs later
ide1: BM-DMA at 0xd400-0xd407, BIOS settings: hdc:pio, hdd:pio
HPT366: IDE controller on PCI bus 00 dev a0
PCI: Enabling device 00:14.0 (0005 -> 0007)
PCI: Found IRQ 11 for device 00:14.0
PCI: The same IRQ used for device 00:14.1
HPT366: chipset revision 1
HPT366: not 100% native mode: will probe irqs later
ide2: BM-DMA at 0xd900-0xd907, BIOS settings: hde:pio, hdf:pio
HPT366: IDE controller on PCI bus 00 dev a1
PCI: Found IRQ 11 for device 00:14.1
PCI: The same IRQ used for device 00:14.0
HPT366: chipset revision 1
HPT366: not 100% native mode: will probe irqs later
ide3: BM-DMA at 0xdd00-0xdd07, BIOS settings: hdg:pio, hdh:pio
hda: Maxtor 33073U4, ATA DISK drive
hdc: Maxtor 33073U4, ATA DISK drive
hde: Maxtor 33073U4, ATA DISK drive
hdg: Maxtor 33073U4, ATA DISK drive
ide0 at 0xdfa8-0xdfaf,0xdfe2 on irq 5
ide1 at 0xdff0-0xdff7,0xdfe6 on irq 5
ide2 at 0xdf90-0xdf97,0xdf9a on irq 11
ide3 at 0xdfa0-0xdfa7,0xdf9e on irq 11
hda: 60030432 sectors (30736 MB) w/512KiB Cache, CHS=59554/16/63
hdc: 60030432 sectors (30736 MB) w/512KiB Cache, CHS=59554/16/63
hde: 60030432 sectors (30736 MB) w/512KiB Cache, CHS=59554/16/63
hdg: 60030432 sectors (30736 MB) w/512KiB Cache, CHS=59554/16/63
Partition check:
hda: [PTBL] [3736/255/63] hda1
hdc: [PTBL] [3736/255/63] hdc1
hde: hde1
hdg: hdg1
[...]
raid5 personality registered
raid5: measuring checksumming speed
8regs : 570.800 MB/sec
32regs : 344.400 MB/sec
pII_mmx : 684.800 MB/sec
p5_mmx : 739.200 MB/sec
raid5: using function: p5_mmx (739.200 MB/sec)
md driver 0.90.0 MAX_MD_DEVS=256, MD_SB_DISKS=27
md.c: sizeof(mdp_super_t) = 4096
autodetecting RAID arrays
(read) hda1's sb offset: 30009280 [events: 00000001]
(read) hdc1's sb offset: 30009280 [events: 00000001]
(read) hde1's sb offset: 30015104 [events: 00000001]
(read) hdg1's sb offset: 30015104 [events: 00000001]
autorun ...
considering hdg1 ...
adding hdg1 ...
adding hde1 ...
adding hdc1 ...
adding hda1 ...
created md0
bind<hda1,1>
bind<hdc1,2>
bind<hde1,3>
bind<hdg1,4>
running: <hdg1><hde1><hdc1><hda1>
now!
hdg1's event counter: 00000001
hde1's event counter: 00000001
hdc1's event counter: 00000001
hda1's event counter: 00000001
md: md0: raid array is not clean -- starting background reconstruction
md0: max total readahead window set to 744k
md0: 3 data-disks, max readahead per data-disk: 248k
raid5: device hdg1 operational as raid disk 3
raid5: device hde1 operational as raid disk 2
raid5: device hdc1 operational as raid disk 1
raid5: device hda1 operational as raid disk 0
raid5: allocated 4312kB for md0
raid5: raid level 5 set md0 active with 4 out of 4 devices, algorithm 2
raid5: raid set md0 not clean; reconstructing parity
RAID5 conf printout:
--- rd:4 wd:4 fd:0
disk 0, s:0, o:1, n:0 rd:0 us:1 dev:hda1
disk 1, s:0, o:1, n:1 rd:1 us:1 dev:hdc1
disk 2, s:0, o:1, n:2 rd:2 us:1 dev:hde1
disk 3, s:0, o:1, n:3 rd:3 us:1 dev:hdg1
RAID5 conf printout:
--- rd:4 wd:4 fd:0
disk 0, s:0, o:1, n:0 rd:0 us:1 dev:hda1
disk 1, s:0, o:1, n:1 rd:1 us:1 dev:hdc1
disk 2, s:0, o:1, n:2 rd:2 us:1 dev:hde1
disk 3, s:0, o:1, n:3 rd:3 us:1 dev:hdg1
md: updating md0 RAID superblock on device
hdg1 [events: 00000002](write) hdg1's sb offset: 30015104
md: syncing RAID array md0
md: minimum _guaranteed_ reconstruction speed: 100 KB/sec/disc.
md: using maximum available idle IO bandwith (but not more than 100000
KB/sec) for reconstruction.
md: using 124k window, over a total of 30009280 blocks.
hde1 [events: 00000002](write) hde1's sb offset: 30015104
hdc1 [events: 00000002](write) hdc1's sb offset: 30009280
hda1 [events: 00000002](write) hda1's sb offset: 30009280
.
.. autorun DONE.
fileserver:~# cat /proc/interrupts
CPU0
0: 353176 XT-PIC timer
1: 794 XT-PIC keyboard
2: 0 XT-PIC cascade
5: 1585145 XT-PIC ide0, ide1
9: 5261 XT-PIC DC21143 (eth0)
10: 6553 XT-PIC NE2000
11: 1586982 XT-PIC ide2, ide3
12: 1748 XT-PIC aic7xxx
NMI: 0
ERR: 0
fileserver:~# cat /proc/ide/drivers
ide-disk version 1.10
--
Peter Lohmann [EMAIL PROTECTED] Tel.: 05207/920446
www.openstuff.de - Linux-Stofftiere, Linux-Mousepads, Linux-Tastaturen,
Koffein Getränke - alles was den Linuxfan begeistert :-)
--> We are but packets in the internet of life -- Mike (UF)
--
=- To unsubscribe, email [EMAIL PROTECTED] with the -=
=- body of "unsubscribe linux-abit". -=
