Problem with WD Advanced Format drive in LINUX (WD15EARS) [ Edited ]

[dafrizz]

Hi !

I've recently bought a 1.5TB EARS drive, unaware of the 4kB sector issue. Glad I found this thread.

I tried the fdisk approach, followed by creating partitions in gparted but that left me with the partition starting at 63, which I assume is no good.

I then tried the parted approach, which meant I had one big partition, starting at 40. So far so good. But that's not the layout I wanted, so I fired up gparted to modify the partition(s) to my requirements (2x 15GB, 1x for the rest).

I then ran into problems:

a) gparted didn't let me create the first partition as 15GB, 22GB was the minimum. Why ?

b) Went ahead and created 1x22GB, 1x15GB, I for the rest.

When finished I noticed that the first partition now started at 63. I'm assuming that using gparted is the issue. Does this mean I have to use parted to create the partitions, making sure I specify start sectors divisible by 8 ? Why 8 ?

Thanks for any help.

[dafrizz]

BTW Here's what I'm planning to do, can someone tell me if this will work:

mkpart primary ext2 40 3750000

I'm assuming this will create a partition from secot 40 to sector 3,750,000, which according to my calculations should be approx. 15GB

I would then create another partition like so

mkpart primary ext2 3750008 7,500,008

which hopefully creates another 15GB partition

Followed by

mkpart primary ext2 7500016 -1

which should create one more partition, taking up the rest of the space on the drive

Or am I going about this the wrong way ?

[klubad]

Hi,

I bought an external WD HD yesterday, a My Book Essential 1.5T. The HD in the box is a WDC WDC WD15EARS-00Z5B1

I was trying to follow and understand all the conversation here and there

http://thread.gmane.org/gmane.linux.utilities.util-linux-ng/2926

But it is difficult to find a final answer. Below the result of hdparm.What matters seems to be:

    Logical  Sector size:                   512 bytes
    Physical Sector size:                  4096 bytes
    Logical Sector-0 offset:                  0 bytes

With these results, do I need to format my disk with a special method?

For now, I formated the disk using this command:

mkfs.ext3 -T largefile4 /dev/sdb1

But the following gives me:

cat /sys/block/sdb/alignment_offset
0

I would like to be sure that I did good before the disk is filled at 80%...

Here my debian config:

linux kernel 2.6.32-8

fdisk -V
fdisk (util-linux-ng 2.16.2)
parted --version
parted (GNU parted) 1.8.8.git-dirty

hdparm -I /dev/sdb

/dev/sdb:

ATA device, with non-removable media
    Model Number:       WDC WD15EARS-00Z5B1                     
    Serial Number:      WD-WMAVUxxxxxxx
    Firmware Revision:  80.00A80
    Transport:          Serial, SATA 1.0a, SATA II Extensions, SATA Rev 2.5, SATA Rev 2.6
Standards:
    Supported: 8 7 6 5
    Likely used: 8
Configuration:
    Logical        max    current
    cylinders    16383    16383
    heads        16    16
    sectors/track    63    63
    --
    CHS current addressable sectors:   16514064
    LBA    user addressable sectors:  268435455
    LBA48  user addressable sectors: 2930277168
    Logical  Sector size:                   512 bytes
    Physical Sector size:                  4096 bytes
    Logical Sector-0 offset:                  0 bytes
    device size with M = 1024*1024:     1430799 MBytes
    device size with M = 1000*1000:     1500301 MBytes (1500 GB)
    cache/buffer size  = unknown
Capabilities:
    LBA, IORDY(can be disabled)
    Queue depth: 32
    Standby timer values: spec'd by Standard, with device specific minimum
    R/W multiple sector transfer: Max = 16    Current = 0
    Recommended acoustic management value: 128, current value: 254
    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=120ns  IORDY flow control=120ns

[monographix]

i wonder if its worth the time to try  find out, learn, figure out, understand what are you talking about in these 4k sector drives threads LOL

all and all i think those Green 4K hdd are not worth buying, unless you are some HDD guru and intend to use them only with an OS that can use them properly.  Same story with the SSDs as well, so complicated .... Moving ahead technologically tries to automate and simplify and make more user friendly some aspects of the the previous generation while adds other complexities and absolute productivity killers about aspects of the forthcoming generation: summary = no change in productivity and user friendliness (ms is the best example of this phenomenon)

[hans1967]

Hi community,

Have found a tutorial (source: http://www.brain4free.org/wiki/doku.php/blog:wd_advanced_format_hd_mit_linux), translated it from german:

Good description of the problem:

http://www.anandtech.com/storage/showdoc.aspx?i=3691&p=2

Using linux with WDxxEARS HDDs is possible but to install them correctly is a bit tricky:

For using WDxxEARS with linux no special software is required. But it is absolutely necessary to partition your HDD correctly. Under Windows XP or under linux using fdisk or parted with default settings WDxxEARS HDDs get partitioned incorrectly. To test, how and where your partitions got created, you  can use fdisk:

fdisk -l -u

A typical incorrect example will look like that:

 Device Boot      Start         End      Blocks   Id  System
/dev/sdc1              63  2930272064  1465136001   83  Linux 

It is absolutely necessary, that start- and end-parameter are divisible by 4. Default settings in fdisk, parted and Windows XP do not manage that problem. For Windows XP you need this separate Utility: http://www.wdc.com/en/products/advancedformat/index.asp

For linux, this tutorial will help:

http://thunk.org/tytso/blog/2009/02/20/aligning-filesystems-to-an-ssds-erase-block-size/

fdisk using this parameters creates all partitions aligned to 4 KByte blockboundaries:

fdisk -H 224 -S 56 /dev/sdb

Create Partitions and for security reasons test them in the end.

gparted-gui does not work. You have to use parted in console. You have to ensure that your partitions begin and end at 4 KByte blockboundaries. You manage that by swiching parted showing all units in KByte. Dead shot exactly in Kibibyte (http://en.wikipedia.org/wiki/Kibibyte). All Numbers you give in then have to be divisible by 4.

Example for an HDD with one partition (Notice the command "unit kib"):

parted /dev/sdb
(parted) unit kib
(parted) print
Model: ATA WDC WD15EARS-00Z (scsi)
Disk /dev/sdc: 1465138584kiB
Sector size (logical/physical): 512B/512B
Partition Table: msdos

Number  Start    End            Size           Type     File system  Flags
 1      31.5kiB  1465136032kiB  1465136001kiB  primary  xfs


(parted) mkpart
Partition type?  primary/extended? p
File system type?  [ext2]? xfs
Start? 32
End? 1465138584
(parted) print
Model: ATA WDC WD15EARS-00Z (scsi)
Disk /dev/sdc: 1465138584kiB
Sector size (logical/physical): 512B/512B
Partition Table: msdos

Number  Start    End            Size           Type     File system  Flags
 1      32.0kiB  1465138584kiB  1465138552kiB  primary

Hope this will help many people.

Best regards,

hans1967

[nuess0r]

hi dafrizz

i don't know where your problem with the minimum size partition comes from (normally you get this behaviour for partitions with data on it. am I right that your partition was empty?)

we have to wait for some patches for parted and gparted, so the default behavior is fine for the new drives.

go ahead with parted and you should be fine. i didn't checked your calculations but what you plan to do looks fine too me. you should get a properly performing drive by that.

[nuess0r]

hi klubad

please show us your partition details. do this by following what is described in the other posts:

To test, how and where your partitions got created, you can use fdisk:
fdisk -l -u

you only write that you formated the drive, who has created the partition on it? you or during manufacturing?

my guess is, that if you created the partition, you have to go over and follow the instructions here. If the partitioning was made during manufacturing you should be fine.

but thats a quess, please check AND POST your partition table, so we can provide correct instructions for newer My Book Essential drives.

[zackboll]

Hi, I recently upgraded my DirecTV HD-DVR HR21 to use the 1.5 Terabyte WD15EARS hard drive.  The OS of the DirecTV receiver is linux.

Could someone help be with the commands to resize the XFS partitions that the DVR creates at bootup?  The output from "fdisk -l -u" is

Disk /dev/sda: 1500.3 GB, 1500301910016 bytes

255 heads, 63 sectors/track, 182401 cylinders, total 2930277168 sectors

Units = sectors of 1 * 512 = 512 bytes

Disk identifier: 0x00000000

Device           Boot            Start                    End                      Blocks                ID         System

/dev/sda1                           63                       1060289             530113+           82         Linux swap / Solaris

/dev/sda2                           1060290            32531624          16735667+      83         Linux

/dev/sda3                           32531625          2930272064     1448870220    83         Linux

When I open the device in G-Parted, sda1 and sda3 show up as unknown, sda2 shows up as xfs.  I must be missing some libraries for g-parted not to identify sda1 and sda3.

Thanks,
Zack      

[dafrizz]

Well, I repartitioned the whole drive and this time didn't have any issues with any minimum sizes. This is the output from fdisk -l -u afterwards:

Disk /dev/sdb: 1500.3 GB, 1500301910016 bytes
255 heads, 63 sectors/track, 182401 cylinders, total 2930277168 sectors
Units = sectors of 1 * 512 = 512 bytes
Disk identifier: 0x000350f8

   Device Boot      Start         End      Blocks   Id  System
/dev/sdb1              64    30716279    15358108   83  Linux
/dev/sdb2        30716280    61432559    15358140   83  Linux
/dev/sdb3        63472816  2930272064  1433399624+  83  Linux
/dev/sdb4        61432560    63472814     1020127+  82  Linux swap / Solaris

Partition table entries are not in disk order

Looks to me like the partitions are properly aligned.

[STAMS]

Hi,

You can also use LVM to solve the problem:

x64etch:~# hdparm -i /dev/sdc
/dev/sdc:
 Model=WDC WD20EARS-00S8B1, FwRev=80.00A80, SerialNo=WD-WCAVY2798944
 Config={ HardSect NotMFM HdSw>15uSec SpinMotCtl Fixed DTR>5Mbs FmtGapReq }
 RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=50
 BuffType=unknown, BuffSize=unknown, MaxMultSect=16, MultSect=16
 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=3907027055
 IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
 PIO modes:  pio0 pio3 pio4
 DMA modes:  mdma0 mdma1 mdma2
 UDMA modes: udma0 udma1 udma2 udma3 udma4 udma5 *udma6
 AdvancedPM=no WriteCache=enabled
 Drive conforms to: Unspecified:  ATA/ATAPI-1,2,3,4,5,6,7

1/ delete all partition

x64etch:~# fdisk /dev/sdc
Delete all partition
w

2/ Install lvm (i use debian)

x64etch:~# apt-get install lvm2 dmsetup mdadm

3/ spend a little time to read the man page

x64etch:~# man pvcreate
    ...
    If the 2nd SCSI disk is a 4KB sector drive that compensates for windows partitioning (sector 7 is the lowest aligned logical block, the 4KB sectors start at LBA -1, and consequently sector 63 is
    aligned on a 4KB boundary) manually account for this when initializing for use by LVM:
        pvcreate --dataalignmentoffset 7s /dev/sdb
    ...

4/ Prepare the disk

x64etch:~# pvcreate --dataalignmentoffset 7s /dev/sdc
  Physical volume "/dev/sdc" successfully created

5/ Create a volume group

x64etch:~# vgcreate wd2tsata /dev/sdc
  Volume group "wd2tsata" successfully created

6/ and the end create the logical volume

x64etch:~# lvcreate --name vm_storage --size 300G wd2tsata
  Logical volume "vm_storage" created

7/ Format your logical volume

x64etch:~# mkfs.ext4 /dev/wd2tsata/vm_storage

You can change the size has you want. I use the disk on samba file server, and i transfert 4.4 G on it by network share in 45 seconds. I update later the post if i found a problem during the time of used.

have fun.

Source:

http://www.howtoforge.com/linux_lvm

http://www.wdc.com/wdproducts/library/WhitePapers/FRA/2579-771430.pdf

http://www.mentby.com/mike-snitzer-2/data-alignment.html

Stams