SATA & NCQ

Dear All,
I have a system as follows;
Processor  : Pentium IV 530 3.0Ghz 775 LGA
Mainboard : MSI 7028 915P Neo2-FR (7028-030)
Ram         : 512 x 2 Kingston Dual Ram 533
GCard       : Sparkle Geforce 6600GT 128 Mb Ram
Hdd         : Seagate 160 GB HDD 7200 rpm 8mb cache SATA, NCQ
The hard disk was selected espacially because of the NCQ (Native Command Queing) ability.
I have two questions (problems). I am not so familiar with the Bios or the hardware (as you will soon notice )
First;
When I enter Bios, there is the following key,
                          Integrated Peripheral -> SATA Devices Configuration -> ATA/IDE configuration
There are three choices for this key,
                          Disabled / Compatible / Enhanced.
If you select as Compatible or Enhanced, then another key appears
                           Configure SATA as
and there is only one value IDE. No selection for NCQ or anything else.
Default values are ATA/IDE Config = Enhanced and Configure SATA as = IDE
I did not quite understand what configure sata as IDE means. Am I using the SATA Drive as an IDE drive? and what can I do to use the hdd's ncq mode? Doesnt the mainboard support ncq? Are the above settings correct or should I change them to what?
I can not see AHCI or RAID in the choices as the mainboards handbook says. I tried to flash the bios with the latest version but it did not seem to work either. Then I understood that the mainboard has ICH6 chipset (not ICH6R). Still it has Via Raid chip on the mainboard. What should I do to use AHCI and NCQ?
Second;
What the hell does   Integrated Peripheral -> IDE Devices Configuration -> PCI IDE Busmaster key? Should it be enabled or disabled?
PLEASE HELP !!!
FROM THE HANDBOOK
ATA/IDE Configuration, Configure SATA as
These 2 items allow you to select the ATA/IDE and SATA configuration. Select [Disabled] in ATA/IDE Configuration if you want to disable both ATA/IDE configuration. Select [Compatible] or [Enhanced] to use the IDE, S-ATA and P-ATA devices. Refer to the following tables for details.
ATA/IDE Configuration
(Compatible)
SATA Only [SATA 1/3/2/4]
PATA Pri, SATA Sec [IDE1, SATA2/4]
SATA Pri, PATA Sec [SATA1/3, IDE1]
PATA Only [IDE1]
ATA/IDE Configuration
(Enhanced)
IDE [IDE1, SATA 1/2/3/4]
AHCI [IDE1, SATA 1/2/3/4]
RAID [IDE1, SATA 1/2/3/4],
[SATA support RAID 0 or 1]
For the setting options of Configure SATA as, select [IDE] if you want to have SATA as IDE function. Select [AHCI] to allow the SATA to have Advanced Host Controller Interface (AHCI) feature, which supports improved serial ATA disk performance with native command queuing & native hot plug. Select [RAID] to use SATA as RAID function. Setting options: [IDE], [AHCI], [RAID].

You definately want the first setting to be Enhanced. This may be enough to give you AHCI mode. I had a look at the manual for the 945 board and it has the same Phoenix bios, but the settings are rearranged.
I don't know how you test to see if it is enabled for real or not.
PCI IDE Busmaster key determines if a PCI controller card can be used as a boot controller. I have never seen a difference having it turned on or off, and when I used a Promise PCI card if the setting was off it got turned on anyways.

Similar Messages

  • Need help install SATA-NCQ HD on 865PENEO2

    Ive got a MSI 865PENEO2 MOBO and as far as i can see, ive got 2 IDEports and 2 SATA ports.
    My IDE HD busted a week ago so i got myself a new SATA DRIVE. Now i dont know what to do. Ive got a WINDOWS XP install cd with intergrated SP2. And ive pluged SATA cable+power cable in. I want to install XP on my new SATA drive. I have never delt with SATA so i dont know what i am doing.
    Do i need to install drivers? WIll i get that error of windows limiting how much HD space i can see? (my HD is a Seagate 300GIG SATA NCQ).
    Do i need to BIOS FLASH so it can reconsie my 300GIG SATA HD?
    Also after installing the SATA drive, my friend comes ouver sometimes with his IDE drive, can i just plug the IDE drive into ide port and will it show up as slave drive after the sata?
    thankyou people.

    Without RAID you probably won't need the F6 driver install, just make sure in your HDD priority that the SATA drive is the first HDD.

  • How to enable SATA NCQ/AHCI in MacBook (Late 2007)

    I am planning to upgrade the HDD to a bigger SATA-2 model.
    I would like to ask does *MacBook (late 2007)* with *Mac OS X 10.5.2* natively support NCQ / AHCI ?
    Do I have to enter the BIOS to *enable the AHCI* function of the HDD?

    deltatux wrote:
    Jonathan Kurtzman wrote:
    They all have a bios but you don't access it like on an old windows installation. If the drive is compatible, it will work.
    Wrong, EFI is NOT BIOS. They are two separate things. They do the same job, but EFI does it more efficiently due to the fact that it is a lot newer. You do not require BIOS to run EFI, or the other way around.
    This sounds more like a semantic discussion rather than the original poster's query whether or not something needed to be done in order to access a larger drive. Perhaps he used the term "BIOS", but we all know there have been means of updating the firmware on MacBooks, even if it isn't a traditional BIOS reflash.
    http://www.apple.com/support/downloads/macbookefifirmwareupdate10.html
    http://www.apple.com/support/downloads/macbookefifirmwareupdate11.html

  • SATA II NCQ X K8N Neo 4 Platinum ... whats happening??

    Hey everybody!
    So, im having big problems here since i've bought a Samsung SP2004C 200Gb Sata II NCQ and tries to install Windows XP on it... the deal is: I can normally install Windows XP (and at F6 screen, install the SATA driver).. so after about 2 weeks, my pc starts to get very slow and randomly crash corrupting files on Windows (like the last time, it was rundll32.dll... but this doens't mean it will be always this file, it can be another one) and so the Windows don't boot or softwares don't load (when Windows boot)...
    When this happens, i'm forced to reinstall Windows... and so it works really fine for 2 week and sometime every problem is back again... I'm really tired of this problem, this is the 3rd time i format my system... and I use my station for fun and work, so you can imagine the lot of work is formatting this pc lol... it takes about 5 or 6 hours to configure all softwares and everything else... heheehe ... but the deal is, in MSI web site, it says that nforce 4 (the native SATA controller) have problems with SATA NCQ tecnologies .... and so they say to disabled it... but... how can i disabled ?? can someone help me?? I need an answer before it starts to crash again lol!!! ...
    you can see what MSI says here: Click Here!
    and also, trying to not have the same problem, I tried to install this HD in the Silicon Controller (the 3rd part controller) and when I'm in the installation screen (Windows), it says that can't find my HD!! (ok! I configured the jumpers to 1.5gb)... I'm trying to put it in the SATA5/SATA6.. what's happening??
    I'm really desperated!
    Can someone help me??? 
    regard to everybody!
    Eduardo Wagner

    Quote from: Bas on 19-March-07, 05:27:20
    Sorry again....but it won't find your HD on that controller unless you loaded the floppy WITH the drivers for that controller during the CDrom-install-boot....
    Windows does NOT know about that controller or HD unless you load the drivers during install.
    Either you don't explain it all correctly, or I miss something here...
    In the very early stage of the booting of the CDrom it asks you to press F6 to load drivers, mostly talking about SCSI drivers but those drivers simular an need to be loaded at that stage.
    If you didn't load them, they HD isn't found.
    If you did all this and it doesn't show the HD, maybe the cable is wrong or try a differend brand HD on it.
    Sorry, I get puzzled by the explenation you give.
    Yes, you are right Bas.... sorry, my english is a lil bit rubish, never did an english course before hehehe... back to our convo, I said it really wrong.. its like you said..
    "In the very early stage of the booting of the CDrom it asks you to press F6 to load drivers, mostly talking about SCSI drivers but those drivers simular an need to be loaded at that stage."
    thats what I'm doing... loading the SATA driver with floppy (then appears "SiS Sata Driver XP/2000" can't remember exactly) in the F6 screen and in the partition screen the HD can't be found... i was using SATA5 and later SATA6... is there any problem with those ports? 
    best regards bas! 
    edit.: btw... I was reading again your post and realised what you said:
    If you did all this and it doesn't show the HD, maybe the cable is wrong or try a differend brand HD on it.
    The controller identifies the HD in the "after post screen"... so how can this be possible when the cable is wrong? 

  • NCQ Drive support

    Does anyone know if the SATA controller will support Seagate's 'NCQ' on their new drives. These things are fast! I'm pretty sure if you plug one in, it will run as normal SATA drive, but it be a waste to get NCQ version drives and Apple can't support it. Running dual 2.5 with 10.3.9
    Cheers

    You might want to head over to www.storagereview.com and click on Drive Comparison Benchmark and then check out the review and also what they have to say and found about NCQ.
    Apple jumped on SATA 1.0 and every vendor has their own specs and features (pick and choose, ala carte?).
    Some drives have NCQ built into a DSP chip in the drive.
    If you want a nice fast drive, try a 74GB Raptor (your PCI-x would require a SeriTek controller to boot and use the new and faster 150GB).
    The Raptor was actually not truly a native SATA drive, used TCQ (taken from SCSI world) and disabling TCQ improved desktop performance. Hitachi/IBM began incorporating TCQ in the 180GXP (now T7K250 but same Vancouver series) and has lately switched to the SATA NCQ, which is best suited to RAID usually.
    Then you have to worry about which drives ship with SSC enabled.
    SATA is now SATA II 2.5 and moving toward 3.0 specification.

  • Smart Array P400 and 516814-B21

    Smart Array P400 in Slot 6 Bus Interface: PCI Slot: 6 Serial Number:[Personal Information Removed] Cache Serial Number:[Personal Information Removed] RAID 6 (ADG) Status: Enabled Controller Status: OK Hardware Revision: D Firmware Version: 7.24 Rebuild Priority: Medium Expand Priority: Medium Surface Scan Delay: 3 secs Surface Scan Mode: Idle Wait for Cache Room: Disabled Surface Analysis Inconsistency Notification: Disabled Post Prompt Timeout: 15 secs Cache Board Present: True Cache Status: OK Cache Ratio: 25% Read / 75% Write Drive Write Cache: Disabled Total Cache Size: 512 MB Total Cache Memory Available: 464 MB No-Battery Write Cache: Disabled Cache Backup Power Source: Batteries Battery/Capacitor Count: 1 Battery/Capacitor Status: OK SATA NCQ Supported: True Array: A Interface Type: SAS Unused Space: 0 MB Status: OK Array Type: Data Logical Drive: 1 Size: 558.7 GB Fault Tolerance: 1+0 Heads: 255 Sectors Per Track: 32 Cylinders: 65535 Strip Size: 128 KB Full Stripe Size: 256 KB Status: OK Caching: Enabled Unique Identifier: 600508B1001047395356543259490033 Disk Name: /dev/cciss/c0d0 Mount Points: None Logical Drive Label: A37771AEP61630G9SVT2YI734D Mirror Group 0: physicaldrive 1I:1:2 (port 1I:box 1:bay 2, SAS, 300 GB, OK) physicaldrive 1I:1:3 (port 1I:box 1:bay 3, SAS, 300 GB, OK) Mirror Group 1: physicaldrive 2I:1:5 (port 2I:box 1:bay 5, SAS, 300 GB, OK) physicaldrive 2I:1:6 (port 2I:box 1:bay 6, SAS, 300 GB, OK) Drive Type: Data physicaldrive 1I:1:2 Port: 1I Box: 1 Bay: 2 Status: OK Drive Type: Data Drive Interface Type: SAS Size: 300 GB Rotational Speed: 15000 Firmware Revision: HPD6 Serial Number:[Personal Information Removed] Model: HP EF0300FATFD Current Temperature (C): 22 Maximum Temperature (C): 78 PHY Count: 2 PHY Transfer Rate: Unknown, 3.0Gbps physicaldrive 1I:1:3 Port: 1I Box: 1 Bay: 3 Status: OK Drive Type: Data Drive Interface Type: SAS Size: 300 GB Rotational Speed: 15000 Firmware Revision: HPD5 Serial Number:[Personal Information Removed] Model: HP EF0300FARMU Current Temperature (C): 25 Maximum Temperature (C): 55 PHY Count: 2 PHY Transfer Rate: Unknown, Unknown physicaldrive 2I:1:5 Port: 2I Box: 1 Bay: 5 Status: OK Drive Type: Data Drive Interface Type: SAS Size: 300 GB Rotational Speed: 15000 Firmware Revision: HPD6 Serial Number:[Personal Information Removed] Model: HP EF0300FATFD Current Temperature (C): 23 Maximum Temperature (C): 76 PHY Count: 2 PHY Transfer Rate: Unknown, Unknown physicaldrive 2I:1:6 Port: 2I Box: 1 Bay: 6 Status: OK Drive Type: Data Drive Interface Type: SAS Size: 300 GB Rotational Speed: 15000 Firmware Revision: HPD6 Serial Number:[Personal Information Removed] Model: HP EF0300FARMU Current Temperature (C): 27 Maximum Temperature (C): 87 PHY Count: 2 PHY Transfer Rate: Unknown, Unknown As you can see, disk models. Write speed of not more than 5MB / s If you make an array of disks only one model - more than 300MB / s What can be done?

    Hi:
    You may also want to copy and paste your issue on the HP Business Support Forum -- Disk Array section.
    http://h30499.www3.hp.com/t5/Disk-Array/bd-p/itrc-​195#.Urmg1-l3u9I

  • Extracting .rar files is slow, too slow

    hi
    i have problem with extracting files. it takes too much time.
    just for ex'  extracting 1G on linux took 3.5 minutes and 40 sec' on windows (on the same hardware)
    * i'm using xarchiver 0.4.6
    ty

    dids22 wrote:
    hi
    i have problem with extracting files. it takes too much time.
    just for ex'  extracting 1G on linux took 3.5 minutes and 40 sec' on windows (on the same hardware)
    ty
    I've noticed the same problem. Pacman grinds system (firefox) to a halt too.
    iirc, I dont think rar'ing was always this slow. Guess it might have something to do with sata, ncq and schedulers.
    unrar x 701MB.rar temp2/  4.69s user 4.51s system 4% cpu 3:12.77 total

  • Hard disk drive replacements for MacBookPro

    Good morning.
    I own a MBP, 2.16 Intel Duo Core.
    110G (original) hard disk at about 80%; running OS 10.4.11; need to upgrade to 10.5.6 to run iLife 9 for a class at school.
    1) Have received disk error -- "Volume Header Needs Minor Repair" Error: underlying task reported failure on exit.
    2) Repair from install disk does not correct the problem.
    3) Machine is still operational although it has not been shut down either.
    4) Since I need additional disk space, why not just upgrade the disk and the OS?
    5) Will either of these disk work? and well? why or why not?
    -Model ST904304N3; SATA NCQ interface 16MB cache, 320 G
    -Model ST3500641AS-RK; 500 G bare bones (?)
    6) Any one on the deep south side of chicago (Cal City/Hammond area) work such things?
    Thanks

    Welcome to the Apple discussions.
    Any 2.5" serial ATA (SATA) drive will work. Buying a new hard drive is an option, since the other repair you could try would be to buy Disk Warrior or TechTool Pro, both which cost a little more than a new hard drive.
    I usually get my hard drives from Newegg.com, with good prices and good user comments. Here are SATA drives they carry: http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&N=2010150380%201035915 133&name=SATA%203.0Gb%2fs . OWC is another source, and here are their drives: http://eshop.macsales.com/shop/hard-drives/2.5-Notebook/SATA .
    If you want to buy your own drive and have someone else install it, you might consider an Apple Authorized Service Provider. The closest one can be located at http://www.apple.com/buy/locator/service/ . Note Apple stores won't install a drive you bring in, however, generally the authorized service providers will.

  • RAW Video & File Corruption

    I am a photographer and filmmaker having serious corruptions issue with RAW photo and video files on Areca RAID arrays. RAW files (.CR2, .NEF, .DNG, etc...) are particularly affected but DSLR .MOV and Cinema DNG files are being corrupted as well.
    This is happening on two systems that are identical except for the Areca controllers and drives (ARC-1880, & ARC-1882) and different drives. Originally these arrays were both RAID-3 but I converted on to RAID5 to see if the problem was corrected. The issue still persists.
    I've contacted Areca for support but have not received a response.
    Any advice or help would be deeply appreciated. See system specs below:
    See these images for examples:
    http://i.imgur.com/zxeFnQu.jpg
    http://i.imgur.com/dnkW16h.jpg
    Model name=ARC-1882I
    Firmware Version=V1.50 2012-06-01
    Disk Vendor=Seagate
    Disk module=ST3000DM001-1CH166
    Disk Firmware Version=CC24
    Operating System=Windows 7
    Driver version=6.20.0.26
    Motherboard Vendor=Gigabyte
    Motherboard Model name=GA-X79-UD5
    Motherboard BIOS Version=F12

    Tony,
    I do have another theory, but before I begin, I'll say Eric Bowen from ADK is an absolute uber-tech and he knows a LOT more than I ever will .
    Here goes...
    What you are seeing is absolutely bizarre!
    Areca cards are awesome, but their support is not that great. You might want to pose this same question on the Areca's Owners Forum on hardforum.com. There are some really experienced Areca users that frequent that forum that may recognize this particular failure mode.
    Your firmware is pretty old; I updated all of my Areca's (1880 and 1882 series cards) to v1.52 a long time ago. I would suggest you do the same. Areca does not share all the "fixes" that they may do in firmware and in general newer is better.
    I read a long, long time ago about some subtle incompatibilities with some drives and Areca controller cards specifically dealing with SATA NCQ support. My number one guess, and hope for you, is that you have this turned on and that by turning it off you may resolve this crazy issue.
    Finally, I don't agree with Eric about only using non-Enterprise drives with Areca parity RAIDS. I've been violating that rule for too many years using WD and Hitachi drives to feel that is your only problem here. Before I ran Areca's with non-enterprise drives I did a lot of www research regarding users building 48 drive arrays using non-Enterprise drives and Areca controllers (for them Enterprise was way more expensive with that many drives in play). And personally, I've had probably a dozen Areca cards, used probably around 60 various drives in them over the years and never, never, ever lost any data. My use has been 100% RAID 5 (and some RAID 0) and while I have had varied hardware failures (5 bad drives - 3 were Enterprise and 2 were non-Enterprise), a bad RAM stick (Areca replaced on wtty. - never had any file errors at all, but the Areca error log showed about 6 events where memory ECC had to kick in to prevent a loss), a bad Areca CPU (totally cratered, but when then sent me a RMA card the drives picked right up and again, no data loss whatsoever).
    Yes, Enterprise drives are recommended, but Areca's newer firmware bios settings allow you to even disable TLER completely which can result is REALLY slow array performance but NO data loss under any circumstance.
    Best of luck! What you are dealing with sounds VERY frustrating!
    Jim

  • Would you be willing to install WinXP using a floppy and driver w/ BootCamp

    I think Apple may be able to provide us full SATA compatibility quickly if they were willing to program the AHCI controller device in enhanced AHCI mode.
    This would require us to install (reinstall) windows and use a USB floppy to install the intel AHCI drivers, since they are not a part of WinXP. This is generally what users of the server-class Intel motherboards have to do if they want full SATA performance.
    Users who are not accustomed to WinXP installation may find this too difficult.
    Perhaps instead of a BIOS set up, Apple can provide an OS X utility that allows configuration of certain system resources. Many WinTel manufacturers provide similar utilities, to simplify BIOS setup. Such a utility would allow more advanced users to enable enhanced-AHCI mode, while the mess experienced users can continue to work in PIO mode.
    If you are interested in learning more about the Intel SATA controller in the WinTel environment, you can find more information in the following links:
    http://www.microsoft.com/whdc/device/storage/serialATA_FAQ.mspx
    http://www.intel.com/support/chipsets/imsm/sb/cs-021736.htm
    ftp://download.intel.com/design/chipsets/manuals/30144401.pdf (Not specific to the 5000X chipset, but AHCI is a standard across multiple intel chipsets)
    Mac Pro 2.66   Mac OS X (10.4.7)  

    The real problem with the Mac Pro and Windows right now is how it sets up the SATA controller device in the CSM, I believe.
    Apple can fix this one of 2 ways:
    1. Give us PATA emulation with full DMA support, but this does not give us SATA NCQ.
    2. Give us Enhanced AHCI support, with NCQ, but this requires us to install an intel driver at windows install time using a USB floppy.
    I believe #2 shouldn't be hard for apple to implement, but it makes Windows harder to install. That is why I called for an OS X application that lets the user choose which mode they want the SATA controller to run in.
    I included the links above as background material into the full problem, from how to program the device in a BIOS (or CSM) to installing the drivers.
    I am asking people, if Apple could provide a firmware update, would you be willing to install Windows again using a USB floppy and driver, or would you rather wait for PATA emulation with DMA support.
    I previously wrote a posting that describes this problem:
    I have spent a few days researching SATA, and how it is handled under Windows, especially Intel's implementation. Intel is pushing a standardized controller interface called AHCI. AHCI does support legacy PATA emulation. In WinTel platform, it is a requirement of the BIOS to configure the AHCI device properly. The AHCI PCI device will present different PCI subcodes depending if it is running in PATA or native mode. These PCI subcodes are used to choose the appropriate driver at boot time.
    For best performance, SATA wants to run in native mode which, on the Mac Pro platform, means running the controller device in AHCI mode. AHCI fully supports DMA as well as Native Command Queuing (NCQ). Combining these two technologies give the best performance.
    The PATA emulation can support DMA, depending on the actual BIOS implementation. PATA can not support NCQ.
    With Boot Camp, Apple is providing a BIOS layer on top of EFI, the new standard for low-level PC firmware. This layer is called a CSM (Compatibility Support Module). I believe the CSM is not programming the device to enter AHCI mode. Also, the CSM is not providing anything other than Programmed I/O (PIO or non-DMA) in their PATA emulation support.
    How do we work around this? I don't know. I am hoping to bring others into this discussion to see if we can figure something out. Considering Apple released Boot Camp 1.1 with the same flaw, I don't believe we can expect them to resolve this issue soon.
    Putting the device in AHCI mode may present more difficult problems for Apple. They will have to change their CSM to support loading the boot sector using AHCI instead of PATA protocols.
    There is one other thing to note, when you boot with device in AHCI mode, and then install Windows, you must install a 3rd party driver provided by Intel at installation time. Pressing the F6 key early on in the install process, inserting a floppy, which contains the driver, and then choosing the driver for installation. The CSM will have to support USB floppies (I have not tested to see if it does, yet). It also means that any user that currently has Windows installed will have to reinstall Windows once we determine how to put the device in AHCI mode.
    If we were willing to give up NCQ, it might be easier for Apple to extend their CSM to include better PATA emulation. We may be limited to 100MB/s, or 133MB/s at best. This is more than enough bandwidth to support any single drive.
    Various Intel Macs   Mac OS X (10.4.6)  

  • The driver detected a controller error on \Device\Harddisk1\D

    Hope someone can help, yesterday in the event viewer I encountered the following error
    "The driver detected a controller error on \Device\Harddisk1\D
    ....I have added no new hardware for months, I checked this out on the microsoft website but they recommend changing the cable!
    I ran diagnostics on the hard drives and they found no errors at all ....I got a feeling it the motherboard, frustrating as this is the second board Ive had!
    Can anyone advise  on this error?
    My setup is as follows
    P4 3.2GHZ
    1GB Transcend Memory
    1 NEC 2510 DVD Wrtier
    1 IDE 120GB Maxtor Diamond Max
    2 X SATA HDD
    450 watt Thermaltake Butterfly PSU

    This seems to be a common problem between ICH5 chipsets and SATA NCQ drives. Apparantly a BIOS update will be able to fixed the problem but I'm not sure which BIOS you should use. I suggest that you check MSI's website to look into your mobo's BIOS release notes to find out more first.
    Please also note that there's an amount of risk to flashing your mobo's BIOS as and improper flash will caused the mobo to die immediately or having multiple errors due to bad flash. I would recommend that you flash through a floppy bootdisk which seems to be the best method till date.

  • Too slow ZFS

    Hello,
    I'm using Solaris 10 10/08 with ZFS.
    There are processes which flush to write data about 10~100MB every 10 minutes, but sometimes writing doesn't finish in 10 minutes!
    Exactly 30 processes are running and each process fires from 10 to 100MB every 10 minutes.
    It means that 1.5GB can be fired to write at a moment in the worst case.
    But each process read/write only files which exclusively belong to the process and flushes data at different time.
    The probability that the worst case I mentioned occur is very low.
    But too slow writing occurs quite often. That's the problem I wanna solve.
    The system makes a snapshot every hour, so there are very many snapshots of evolving data.
    Every data file has less than 1.1GB size.
    I wanna know what makes the writing too slow.
    The program is using the BerkeleyDB and the BerkeleyDB writes the data to file system.
    Of course there is some probability that the BerkeleyDB is the reason.
    If the problem is in the BerkeleyDB, CPU use should be high. But it's not.
    No CPU is busy. No process uses CPU more than 0.20%, No disk is busy. And the system has 16GB physical memory!
    Is 16GB memory too small for ZFS?
    The system specification is following.
    CPU : Intel Xeon 2.0GHz x2 (total 8 core)
    Memory : 16G real, 1116M free, 1174M swap in use, 7686M swap free
    [root@host:/tmp] # /opt/HPQacucli/sbin/hpacucli controller slot=2 physicaldrive all show
    Smart Array P800 in Slot 2
       array A
          physicaldrive 2E:1:1 (port 2E:box 1:bay 1, SATA, 500 GB, OK)
          physicaldrive 2E:1:2 (port 2E:box 1:bay 2, SATA, 500 GB, OK)
          physicaldrive 2E:1:3 (port 2E:box 1:bay 3, SATA, 500 GB, OK)
          physicaldrive 2E:1:4 (port 2E:box 1:bay 4, SATA, 500 GB, OK)
          physicaldrive 2E:1:5 (port 2E:box 1:bay 5, SATA, 500 GB, OK, spare)
       array B
          physicaldrive 2E:1:7 (port 2E:box 1:bay 7, SATA, 500 GB, OK)
          physicaldrive 2E:1:8 (port 2E:box 1:bay 8, SATA, 500 GB, OK)
          physicaldrive 2E:1:9 (port 2E:box 1:bay 9, SATA, 500 GB, OK)
          physicaldrive 2E:1:10 (port 2E:box 1:bay 10, SATA, 500 GB, OK)
          physicaldrive 2E:1:11 (port 2E:box 1:bay 11, SATA, 500 GB, OK, spare)
    [root@host:/] # /opt/HPQacucli/sbin/hpacucli controller slot=2 physicaldrive 2e:1:1 show
    Smart Array P800 in Slot 2
       array A
          physicaldrive 2E:1:1
             Port: 2E
             Box: 1
             Bay: 1
             Active Path:
                Port: 2E Box: 1 Bay: 1
             Status: OK
             Drive Type: Data Drive
             Interface Type: SATA
             Size: 500 GB
             Firmware Revision:     HPG3
             Serial Number: 5QG26WQV           
             Model: ATA     GB0500C4413    
             SATA NCQ Capable: True
             SATA NCQ Enabled: True
             PHY Count: 1
             PHY Transfer Rate: 1.5GBPS
    (*** Every physical drive has same specification. ***)
    [doomvas@host user]$ top
    load averages:  0.09,  0.12,  0.23
    138 processes: 137 sleeping, 1 on cpu
    CPU states: 98.9% idle,  0.3% user,  0.8% kernel,  0.0% iowait,  0.0% swap
    Memory: 16G real, 995M free, 1177M swap in use, 7570M swap free
       PID USERNAME LWP PRI NICE  SIZE   RES STATE    TIME    CPU COMMAND
      2692 doomvas   13  59    0  118M  115M sleep   12:25  0.05% db
      2537 doomvas   13  59    0   57M   55M sleep    6:37  0.03% db
      2329 doomvas   13  59    0   52M   50M sleep    6:27  0.03% db
      2472 doomvas   13  59    0   51M   48M sleep    7:34  0.03% db
      2368 doomvas   13  59    0   54M   52M sleep    6:31  0.03% db
      2303 doomvas   13  59    0   66M   63M sleep    5:18  0.03% db
      2407 doomvas   13  59    0   42M   39M sleep    5:29  0.02% db
      2381 doomvas   13  59    0   43M   41M sleep    5:15  0.02% db
      2394 doomvas   13  59    0   45M   43M sleep    4:34  0.02% db
      2420 doomvas   13  59    0   42M   39M sleep    5:29  0.02% db
      2446 doomvas   13  59    0   38M   35M sleep    5:05  0.02% db
      2316 doomvas   13  59    0   36M   33M sleep    4:29  0.02% db
      2342 doomvas   13  59    0   38M   35M sleep    4:28  0.02% db
      2459 doomvas   13  59    0   29M   27M sleep    4:08  0.02% db
      2550 doomvas   13  59    0   37M   35M sleep    3:49  0.02% db
    (*** CPU use for a process doesn't be over 0.20% ***)I made a dtrace script to monitor every fbt:zfs:: probe and profiled for about 30 minutes.
    the most frequent probe : zio_wait_for_children
    total count = 5855583
    total time (nanoseconds) = 7941992752
    average time = 1356
    the most long-take probe : txg_thread_wait
    total count = 270
    total time = 3235383561647 (about 3235 seconds)
    average time = 12072326722 (about 12 seconds)
    If anyone has solution, knows the reason or has an idea to find out the reason, please reply.
    Thank you.
    Edited by: shintaek on Jul 23, 2009 11:48 PM

    exactly, I think we were too impatient.
    It seemed working well after deleting half of snapshots, but just next day it occurred again.
    of course the system is maintaining snapshots just half of the original, but it became slow because unknown reason.
    we've tested something like following because we doubt that zpool was related to the too low performance.
    case 1 : 1 zpool, 1*500MB mkfile, 50*100MB mkfile at any time
    case 2 : 20 zpool, 1*500MB mkfile to ONE zpool, 2~3*100MB mkfile PER zpool at any time
    case 3 : 1 zpool, just 1*500MB mkfile
    in case 1 and 2, 'at any time' means that 100MB mkfile processes were maintained to restart when finished again and again.
    we estimated the time to write 500MB for each case.
    case 1 result : about 6 minutes
    case 2 result : about 16 minutes
    case 3 result : about 4 seconds
    in the case 1, the test was harder than our real-world situation, but it took only 6 minutes.
    our system log told us that writing about 1GB took more than 50 minutes!!!
    do anyone know why the tests above result like this?
    and our zfs performance is still problem.
    please reply anything about our situation.
    we'll of course test the ARC size. thank about this.
    thank you.

  • Dl180 g6 storage errors

    Server running proxmox linux (kernel 2.6.32-29-pve). It's works fine about two month. Yesterday it was terribly slow down.
    Syslog messages:
    Jun 17 08:27:20 proxmox kernel: hpsa 0000:06:00.0: Abort request on C6:B0:T0:L2
    Jun 17 08:27:20 proxmox kernel: hpsa 0000:06:00.0: Abort request on C6:B0:T0:L2
    Jun 17 08:27:20 proxmox kernel: hpsa 0000:06:00.0: Abort request on C6:B0:T0:L2
    Jun 17 08:27:20 proxmox kernel: hpsa 0000:06:00.0: Abort request on C6:B0:T0:L2
    Jun 17 08:27:20 proxmox kernel: hpsa 0000:06:00.0: Abort request on C6:B0:T0:L2
    Jun 17 08:27:20 proxmox kernel: hpsa 0000:06:00.0: Abort request on C6:B0:T0:L2
    Jun 17 08:27:20 proxmox kernel: hpsa 0000:06:00.0: Abort request on C6:B0:T0:L2
    Jun 17 08:27:20 proxmox kernel: hpsa 0000:06:00.0: Abort request on C6:B0:T0:L2
    Jun 17 08:27:20 proxmox kernel: hpsa 0000:06:00.0: Abort request on C6:B0:T0:L2
    Jun 17 08:27:20 proxmox kernel: hpsa 0000:06:00.0: Abort request on C6:B0:T0:L2
    Jun 17 08:27:20 proxmox kernel: hpsa 0000:06:00.0: Abort request on C6:B0:T0:L0
    Jun 17 08:27:20 proxmox kernel: hpsa 0000:06:00.0: Abort request on C6:B0:T0:L0
    Jun 17 08:27:20 proxmox kernel: hpsa 0000:06:00.0: Abort request on C6:B0:T0:L1
    Jun 17 08:27:20 proxmox kernel: hpsa 0000:06:00.0: Abort request on C6:B0:T0:L1
    Jun 17 08:27:20 proxmox kernel: hpsa 0000:06:00.0: Abort request on C6:B0:T0:L1
    Jun 17 08:27:20 proxmox kernel: hpsa 0000:06:00.0: Abort request on C6:B0:T0:L1
    Jun 17 08:27:20 proxmox kernel: hpsa 0000:06:00.0: Abort request on C6:B0:T0:L1
    Jun 17 08:27:20 proxmox kernel: hpsa 0000:06:00.0: Abort request on C6:B0:T0:L1
    Jun 17 08:27:20 proxmox kernel: hpsa 0000:06:00.0: Abort request on C6:B0:T0:L1
    Jun 17 08:27:20 proxmox kernel: hpsa 0000:06:00.0: Abort request on C6:B0:T0:L1
    After rebooting it works fine about 10 hours and symptons returns.
    Smart Array P410 in Slot 1
       Bus Interface: PCI
       Slot: 1
       Serial Number: PACCR0L9VYT1Q3A
       Cache Serial Number: PAAVP0E9VYW10HE
       RAID 6 (ADG) Status: Enabled
       Controller Status: OK
       Hardware Revision: C
       Firmware Version: 6.40
       Rebuild Priority: Medium
       Expand Priority: Medium
       Surface Scan Delay: 15 secs
       Surface Scan Mode: Idle
       Queue Depth: Automatic
       Monitor and Performance Delay: 60  min
       Elevator Sort: Enabled
       Degraded Performance Optimization: Disabled
       Inconsistency Repair Policy: Disabled
       Wait for Cache Room: Disabled
       Surface Analysis Inconsistency Notification: Disabled
       Post Prompt Timeout: 0 secs
       Cache Board Present: True
       Cache Status: OK
       Cache Ratio: 100% Read / 0% Write
       Drive Write Cache: Disabled
       Total Cache Size: 512 MB
       Total Cache Memory Available: 400 MB
       No-Battery Write Cache: Disabled
       Cache Backup Power Source: Batteries
       Battery/Capacitor Count: 1
       Battery/Capacitor Status: OK
       SATA NCQ Supported: True

    Hi:
    I recommend you copy and paste your post to the HP Business Support Forum -- DL Servers section.
    http://h30499.www3.hp.com/t5/ProLiant-Servers-ML-D​L-SL/bd-p/itrc-264#.U5-EOv1OXGg

  • ****MARKOUL RELOADED****

    Yes!! it is true!!..
    ... after 4 1/2years (record time) i had the nForce k7n420 Pro (MS- 6373) it is time also for Markoul to change ship! 
    ... yes you heard correct !!
    ***MARKOUL IS JOINING TO THE NFORCE 4 CLUB!!...***           
    below you can see my new signature and some snapshots from this cosmogonic event 
    Quote
    ...due to signature size restrictions applied also add to the signature below the following devices:
    Barracuda 7200.8 SATA NCQ 250GB
    Barracuda IV 60GB ATA100
    Nec 2500A DVD Recorder /  Pioneer 106S DVD Player
    Creative On Board SoundBlaster Live! 24 bit/96KHz 7.1 ch HD / MSI Audio drv v.5.12.1.353[/font]
    So Forum,
    I salute You!!   
    Manolis aka Markoul
    snap2
    snap3
    snap4
    snap5

    Manolis,
    Welcome to the Athlon64 wing of the nForce Club!!
    Good to see you have finally crossed over my Padawan Learner.
    That Zalman CNPS7700-Cu sure is something to see inside that case. I was thinking it might be time for a new case though. Something like a Full Tower with a lot of cooling fans to keep that baby nice and cool during the Greek Summers. 
    Richard "Jedi Master" Deathstalker

  • NewtorkCard Problem

    I have problem with my Integrated NewtworkCard. When she trying get ip from dhcp server windows display bsod afd.sys 0x000000CB. I connect crossover cable to it and another computer. Configure ip, mask etc. it's working. For now
    i must have another networkcard to connect internet. What's problem is with this card ? It's broken ? Thanks for replies. Sorry for my english.
    My Specs:
    AMD64 2800+
    NEO Platnium
    512 MB Ram Kingston
    Radeon 9600XT
    Seagate SATA NCQ 80GB

    Hi, Beverley.  
    Thank you for visiting Apple Support Communities.  
    I understand that you are unable to trim clips using iMovie.  I would need a little more information on how you were attempting to trim a clip to provide a better answer.  However, make sure you have the movie project open and are selecting the clip that you wish to trim as this will activate the clip trimmer.  
    Trim clips
    -Jason H.  

Maybe you are looking for

  • I want to find out opening balance as on date in fi(for customer)

    hi experts, my req is user enter date through selection screen like (07.04.09) i want to find opening balance. thanks ajay.

  • Is there a way to open CD Tray in G4 with a G3 keyboard?

    I am using a G4 with a G3 keyboard. I can't open the CD Tray. Is there a key stroke or other method to open the CD tray? Thanks

  • Problem with Database Connections in the designer - Catalog not updated

    This is Crystal Reports 2008 (12.2.0290). We have a number of reports that are viewed using the .Net runtime with the connection details set at runtime using ODBC to talk to SQL Server on different systems. This all works fine and is not the problem.

  • Subject Area for GL

    Hi, As far as I understand for BI Apps Fin implementation on top of EBS there are three parts: AP, AR and GL. Witch are the corresponding subject areas that should be executed in order to obtain GL information? For AP I suppose the corresponding subj

  • Service Call Close

    Hi Experts,      Can we close Service call in SAP 8.81 PL 7. I have closed the Service call through DI API(SDK). But its not reflecting the SAP , even though I have given resolution . Kindly help me on this. My code Dim Service As SAPbobsCOM.ServiceC