[ioates]

I recently set up a Linux server with all new hardware and shipped it across the country to a friend's data center. The testing was fine (other than the fact that this motherboard isn't compatible with Linux's soft reboot--yech) but now I'm having serious performance problems with the hard disk. Any help is greatly appreciated.

What happens is this: I'll try to load a large file (say 100M... nothing too crazy) and the CPU basically blocks and all other processes on the machine pretty much freeze. It's a web server, so this is a major problem. Here are some things I've dug up, I am happy to post more details if it'll help:

The motherboard is an Asus P4S800-MX with SiS661 chipset. The hard drive is a Seagate ST3120026A (120G.) Nothing fancy.

When I start accessing lots of data on the hard drive, the CPU looks like this in top:

Cpu(s): 1.7% us, 0.0% sy, 0.0% ni, 0.0% id, 98.3% wa, 0.0% hi, 0.0% si

The load average starts to go up very fast as the website CGIs block for hard disk access.

# uname -a
Linux wonderboy 2.6.12.3 #4 Wed Aug 3 21:59:39 EDT 2005 i686 unknown unknown GNU/Linux

# hdparm -I /dev/hda

/dev/hda:

ATA device, with non-removable media
Model Number: ST3120026A
Serial Number: 4JT0F8FQ
Firmware Revision: 8.01
Standards:
Used: ATA/ATAPI-6 T13 1410D revision 2
Supported: 6 5 4 3
Configuration:
Logical max current
cylinders 16383 65535
heads 16 1
sectors/track 63 63
--
CHS current addressable sectors: 4128705
LBA user addressable sectors: 234441648
LBA48 user addressable sectors: 234441648
device size with M = 1024*1024: 114473 MBytes
device size with M = 1000*1000: 120034 MBytes (120 GB)
Capabilities:
LBA, IORDY(can be disabled)
bytes avail on r/w long: 4 Queue depth: 1
Standby timer values: spec'd by Standard, no device specific minimum
R/W multiple sector transfer: Max = 16 Current = ?
Recommended acoustic management value: 128, current value: 0
DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 *udma5
Cycle time: min=120ns recommended=120ns
PIO: pio0 pio1 pio2 pio3 pio4
Cycle time: no flow control=240ns IORDY flow control=120ns
Commands/features:
Enabled Supported:
* READ BUFFER cmd
* WRITE BUFFER cmd
* Host Protected Area feature set
* Look-ahead
* Write cache
* Power Management feature set
Security Mode feature set
SMART feature set
* FLUSH CACHE EXT command
* Mandatory FLUSH CACHE command
* Device Configuration Overlay feature set
* 48-bit Address feature set
SET MAX security extension
* DOWNLOAD MICROCODE cmd
* General Purpose Logging feature set
* SMART self-test
* SMART error logging
Security:
Master password revision code = 65534
supported
not enabled
not locked
not frozen
not expired: security count
not supported: enhanced erase
HW reset results:
CBLID- above Vih
Device num = 0 determined by CSEL
Checksum: correct


# hdparm -i /dev/hda

/dev/hda:

Model=ST3120026A, FwRev=8.01, SerialNo=4JT0F8FQ
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
BuffType=unknown, BuffSize=8192kB, MaxMultSect=16, MultSect=off
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=234441648
IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5
AdvancedPM=no WriteCache=enabled
Drive conforms to: ATA/ATAPI-6 T13 1410D revision 2:

* signifies the current active mode

[Nerderello]

try disabling the smart feature (in BIOS), this may help a little.

Nerderello

[Marshalle]

On mine I disabled bus mastering (it says this boards bus mastering is only for windows), man there are too many hybrid, windows only software/hardware devices these days. Bus mastering should be hardware only! And yeah, the performance is crapola. This setting gave quite a boost,I have found a few times that ata based bus mastering or RAID devices perform very badly in linux. The only way to get true disk performance is to try to avoid those, the easiest way is to get SCSI, or turn those features off if possible.