SATA slow??

Dusty · 2005-06-12 18:51:38

Thanks, that's actually the drive I had my eye on before, what coincidence. I think I'll switch to the via controller again, the promise drive detection seems to add several sections to boot, which is plenty long enough, seeing as it takes a long time for even lilo to load!

Dusty

cactus · 2005-06-12 18:53:09

so john..you are using devfs? Dusty, are you using udev?

Dusty · 2005-06-12 19:46:34

yes, udev here.

I also tried the ubuntu 64 CD, same results on the hard drive, although it was a little faster (over 2MB/s on one trial), which I assume is to be expected without the 32 bit emulation layer. I've now tried several kernels with no cigar...

Is it at all likely that both controllers on the motherboard are hooped in such a way that they are working only a little... maybe like the communication between the controller and the memory or something? maybe they dropped a screwdriver on the board when they installed it or something like that?

Yeah, I'm grasping at straws now...

Dusty

Edit: BTW, when I run hdparm in ubuntu64 live CD, it doesn't have the HDIO_DRIVE_CMD errors. I don't think this means anything to my problem, but its an interesting anecdote.

johnisevil · 2005-06-12 21:47:36

Yes, I'm using udev also. The problem seemed to be udev trying to use devfs device naming (/dev/discs/disc0/etc...)

Dusty · 2005-06-14 03:15:02

Check this out!!!

I'm running ubuntu64 right now. When I run hdparm in it, I still get the abismal timings, but I can run it on the disk without getting the errors. I checked the output of hdparm -I, I don't see why my disk is so slow. Can anybody see anything wrong with this?:

root@ubuntu:/home/ubuntu # hdparm -tT /dev/sda

/dev/sda:
 Timing cached reads:   3048 MB in  2.00 seconds = 1522.71 MB/sec
 Timing buffered disk reads:    8 MB in 12.16 seconds = 673.84 kB/sec
root@ubuntu:/home/ubuntu # hdparm /dev/sda

/dev/sda:
 IO_support   =  0 (default 16-bit)
 readonly     =  0 (off)
 readahead    = 256 (on)
 geometry     = 14593/255/63, sectors = 120034123776, start = 0
root@ubuntu:/home/ubuntu # hdparm -I /dev/sda

/dev/sda:

ATA device, with non-removable media
        Model Number:       ST3120827AS
        Serial Number:      5MT0JE97
        Firmware Revision:  3.42
Standards:
        Used: ATA/ATAPI-6 T13 1410D revision 2
        Supported: 6 5 4 3
Configuration:
        Logical         max     current
        cylinders       16383   16383
        heads           16      16
        sectors/track   63      63
        --
        CHS current addressable sectors:   16514064
        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: 32
        Standby timer values: spec'd by Standard, no device specific minimum
        R/W multiple sector transfer: Max = 16  Current = 16
        Recommended acoustic management value: 254, current value: 0
        DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 udma5 *udma6
             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
           *    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
Checksum: correctroot@ubuntu:/home/ubuntu # hdparm -tT /dev/sda

/dev/sda:
 Timing cached reads:   3048 MB in  2.00 seconds = 1522.71 MB/sec
 Timing buffered disk reads:    8 MB in 12.16 seconds = 673.84 kB/sec
root@ubuntu:/home/ubuntu # hdparm /dev/sda

/dev/sda:
 IO_support   =  0 (default 16-bit)
 readonly     =  0 (off)
 readahead    = 256 (on)
 geometry     = 14593/255/63, sectors = 120034123776, start = 0
root@ubuntu:/home/ubuntu # hdparm -I /dev/sda

/dev/sda:

ATA device, with non-removable media
        Model Number:       ST3120827AS
        Serial Number:      5MT0JE97
        Firmware Revision:  3.42
Standards:
        Used: ATA/ATAPI-6 T13 1410D revision 2
        Supported: 6 5 4 3
Configuration:
        Logical         max     current
        cylinders       16383   16383
        heads           16      16
        sectors/track   63      63
        --
        CHS current addressable sectors:   16514064
        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: 32
        Standby timer values: spec'd by Standard, no device specific minimum
        R/W multiple sector transfer: Max = 16  Current = 16
        Recommended acoustic management value: 254, current value: 0
        DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 udma5 *udma6
             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
           *    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
Checksum: correct

I've also been running seagate diagnostics off of ultimate boot disk. Its not reporting any errors. I can't tell if its slow or not... can't find a program to actually measure the access time on there.

Dusty

Dusty · 2005-06-15 03:30:28

hdparm wrote:

Recommended acoustic management value: 254, current value: 0

Somebody told me this setting could be adversly affecting my speed, but if I run hdparm -M 254 /dev/sda, I get the ioctl errors. Anyone know of another way I could set it? I can't find anything in the seagate utilities on ultimate boot CD.

Dusty

cactus · 2005-06-15 04:02:42

Command

bonnie++ -fd /home/eliott/bonkerz/

Output

Writing intelligently...done
Rewriting...done
Reading intelligently...done
start 'em...done...done...done...
Create files in sequential order...done.
Stat files in sequential order...done.
Delete files in sequential order...done.
Create files in random order...done.
Stat files in random order...done.
Delete files in random order...done.
Version  1.03       ------Sequential Output------ --Sequential Input- --Random-
                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
hermes.solarnet  2G           39247  14 12466   3           30478   4 169.4   0
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 16 21908 100 +++++ +++ 20570 100 21489  99 +++++ +++ 19345  99
hermes.solarnet,2G,,,39247,14,12466,3,,,30478,4,169.4,0,16,21908,100,+++++,+++,20570,100,21489,99,+++++,+++,19345,99

Dusty · 2005-06-15 04:30:38

That's interesting, these results are comparable...:

Writing intelligently...done
Rewriting...done
Reading intelligently...done
start 'em...done...done...done...
Create files in sequential order...done.
Stat files in sequential order...done.
Delete files in sequential order...done.
Create files in random order...done.
Stat files in random order...done.
Delete files in random order...done.
Version  1.03       ------Sequential Output------ --Sequential Input- --Random-
                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
buchuki          2G           36008  11 13707   3           22708   3 214.5   0
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 16 30443  97 +++++ +++ 25746  99 29978  98 +++++ +++ 24655  99
buchuki,2G,,,36008,11,13707,3,,,22708,3,214.5,0,16,30443,97,+++++,+++,25746,99,29978,98,+++++,+++,24655,99

What's up with that? Now I'm really confused...

Dusty

Dusty · 2005-06-18 00:15:53

I got my new maxtor. Its on sdb:

root:dusty # hdparm -tT /dev/sda

/dev/sda:
 Timing cached reads:   3148 MB in  2.00 seconds = 1571.88 MB/sec
HDIO_DRIVE_CMD(null) (wait for flush complete) failed: Inappropriate ioctl for device
 Timing buffered disk reads:   10 MB in  4.97 seconds =   2.01 MB/sec
HDIO_DRIVE_CMD(null) (wait for flush complete) failed: Inappropriate ioctl for device
root:dusty # hdparm -tT /dev/sdb

/dev/sdb:
 Timing cached reads:   2784 MB in  2.00 seconds = 1390.82 MB/sec
HDIO_DRIVE_CMD(null) (wait for flush complete) failed: Inappropriate ioctl for device
 Timing buffered disk reads:  152 MB in  3.01 seconds =  50.57 MB/sec
HDIO_DRIVE_CMD(null) (wait for flush complete) failed: Inappropriate ioctl for device
root:dusty #

I'm off to harrass Seagate and partition the maxtor.

Dusty

IceRAM · 2005-06-18 00:47:49

1. What Seagate is that?

2. What Maxtor is that?

3. Please make the "checking the HDD temperature by touching"-test I've performed when I rescued my data (painfully) from my ex-less-than-6-months-old-close-to-death-120GB-SATA-Maxtor. I found Maxtor sensibly hotter than my new Seagate (I believe I've checked this while idle). I just want to verify this with others. 10x

P.S. Good luck with Maxtor!

Dusty · 2005-06-18 01:57:48

Seagate model is:
ST3120827AS

Maxtor model is same as johnisevil's:
6Y080M0

Both drives are currently active, as I am copying between the two (very very VERY slowly) at the moment. The Maxtor may be a little hotter than the Seagate, hard to tell for sure. It's a smaller drive (80GB vs 120GB seagate). Also, the Maxtor is working and the Seagate isn't. I just don't know whether to return the Seagate or assume the problem is in the kernel... none of the diagnostic tools on the ultimate boot disk found a problem, but I wasn't able to test timing sone way or another.

Dusty

IceRAM · 2005-06-18 13:15:47

Dusty, I'm sorry to break it to you, but my 2 Seagate HDDs are both ST3120827AS (same as yours) and they both work fine (see my above tests). The total number of sectors is a bit different, but that doesn't seem to bother them while working (I don't use any RAID, yet).

Now that I look over kernel.log, you might want to check their revision number.

# cat /var/log/kernel.log | grep ATA

Jun 18 15:26:52 aquarium hda: HL-DT-ST DVDRAM GSA-4163B, ATAPI CD/DVD-ROM drive
Jun 18 15:26:52 aquarium hda: ATAPI 40X DVD-ROM DVD-R-RAM CD-R/RW drive, 2048kB Cache, UDMA(33)
Jun 18 15:26:52 aquarium ata1: SATA max UDMA/133 cmd 0xC000 ctl 0xC402 bmdma 0xD000 irq 18
Jun 18 15:26:52 aquarium ata2: SATA max UDMA/133 cmd 0xC800 ctl 0xCC02 bmdma 0xD008 irq 18
Jun 18 15:26:52 aquarium ata1: dev 0 ATA, max UDMA/133, 234439535 sectors: lba48
Jun 18 15:26:52 aquarium ata2: dev 0 ATA, max UDMA/133, 234441648 sectors: lba48
Jun 18 15:26:52 aquarium Vendor: ATA       Model: ST3120827AS       Rev: 3.42
Jun 18 15:26:52 aquarium Vendor: ATA       Model: ST3120827AS       Rev: 3.42

Dusty · 2005-06-18 17:53:22

That's interesting. Here's my output:

Jun 18 10:51:02 buchuki hda: HL-DT-ST DVDRAM GSA-4163B, ATAPI CD/DVD-ROM drive
Jun 18 10:51:02 buchuki hda: ATAPI 40X DVD-ROM DVD-R-RAM CD-R/RW drive, 2048kB Cache, UDMA(33)
Jun 18 10:51:02 buchuki ata1: SATA max UDMA/133 cmd 0xD000 ctl 0xC802 bmdma 0xB800 irq 20
Jun 18 10:51:02 buchuki ata2: SATA max UDMA/133 cmd 0xC400 ctl 0xC002 bmdma 0xB808 irq 20
Jun 18 10:51:02 buchuki ata1: dev 0 ATA, max UDMA/133, 234441648 sectors: lba48
Jun 18 10:51:02 buchuki ata2: dev 0 ATA, max UDMA/133, 160086528 sectors:
Jun 18 10:51:02 buchuki Vendor: ATA       Model: ST3120827AS       Rev: 3.42
Jun 18 10:51:02 buchuki Vendor: ATA       Model: Maxtor 6Y080M0    Rev: YAR5

What is your acoustic value if you run hdparm -I on the seagates? (hdparm -I only works for me when I'm in a 64 bit distro). Mine is 0 for the Seagate, and recommended something like 128 or so... and the Maxtor is a higher number like that. That's the only thing that could be a problem, but I don't know how to set it because the hdparm option to set acoustic value doesn't seem to work. :-(

Maybe I should take the drive to a technician to be tested.

Dusty

IceRAM · 2005-06-18 18:35:35

# hdparm -I /dev/sda

/dev/sda:
 HDIO_DRIVE_CMD(identify) failed: Inappropriate ioctl for device

Doesn't work here. And I have no 64bit system to test. Sorry.

Have you tried moving the Seagate on the 2nd SATA port and run a test on it then?

Can't you test it in another computer, with a LiveCD?

(your SATA controller on the MB might have something against Seagate's NCQ - although I haven't heard of such incompatibilities / I have a non-NCQ capable chipset on my MB)

Dusty · 2005-06-18 19:29:36

IceRAM wrote:

[
Can't you test it in another computer, with a LiveCD?

I don't really have another system to test it on, nothing that supports sata. If I get to know the CS guys at work, maybe they can help me test it. Since this particular model works with Linux for you, the problem must either be in the hard drive, or the hard drive/sata controller combination. BUT I tried it with both the on-board via controller and the on-board promise controller, and it worked with neither. I think I better talk to NCIX and/or Seagate about returning it? I'd rather verify the problem first though... :-/

Dusty

IceRAM · 2005-06-19 05:41:28

I was thinking today that you might want to try compiling a really old kernel and try it then... by really old I'm thinking of something like... 2.6.7 (or.. you already did this by using LiveCDs with old kernels?).

Dusty · 2005-06-19 16:04:48

Yeah, my Knoppix CD uses 2.6.7, I believe, maybe 2.6.6.

One thing, its not just Linux that's really slow; it took LiLo a long time to boot off of this drive too. This again suggests its the hard drive and/or controller.

The Seagate people, while not very helpful, told me that its impossible for a SATA drive to slow down and it must be the controller. But I've tried both on-board controllers with this drive and it works with a Maxtor so... it must be the drive???

I'm starting to talk in circles here. I have to make friends with somebody who has a SATA motherboard and try to settle the issue.

Dusty

Dusty · 2005-06-20 22:09:20

FTR, I'm returning the drive to NCIX for testing and replacement if it is defective.

Dusty

Dusty · 2005-07-09 00:29:40

Ok, if anybody comes across this thread, I had a new Seagate drive shipped to replace the old one. It works fine. So I don't want anybody to think that Seagate or the Seagate/Asus combination is a problem; I just had a faulty drive.

Dusty

IceRAM · 2005-07-09 08:21:13

Glad to head everything is solved now.

Arch Linux

#26 2005-06-12 18:51:38

Re: SATA slow??

#27 2005-06-12 18:53:09

Re: SATA slow??

#28 2005-06-12 19:46:34

Re: SATA slow??

#29 2005-06-12 21:47:36

Re: SATA slow??

#30 2005-06-14 03:15:02

Re: SATA slow??

#31 2005-06-15 03:30:28

Re: SATA slow??

#32 2005-06-15 04:02:42

Re: SATA slow??

#33 2005-06-15 04:30:38

Re: SATA slow??

#34 2005-06-18 00:15:53

Re: SATA slow??

#35 2005-06-18 00:47:49

Re: SATA slow??

#36 2005-06-18 01:57:48

Re: SATA slow??

#37 2005-06-18 13:15:47

Re: SATA slow??

#38 2005-06-18 17:53:22

Re: SATA slow??

#39 2005-06-18 18:35:35

Re: SATA slow??

#40 2005-06-18 19:29:36

Re: SATA slow??

#41 2005-06-19 05:41:28

Re: SATA slow??

#42 2005-06-19 16:04:48

Re: SATA slow??

#43 2005-06-20 22:09:20

Re: SATA slow??

#44 2005-07-09 00:29:40

Re: SATA slow??

#45 2005-07-09 08:21:13

Re: SATA slow??

Board footer