Loading ...
Sorry, an error occurred while loading the content.

Welland Greenstar HDD spindown problem

Expand Messages
  • tjrundy
    Hello, Apologies for a 2nd email, but I thought that by splitting out topics into threads, it might be easier for people to refer to these in the future. I m
    Message 1 of 8 , Jul 13, 2008
    • 0 Attachment
      Hello,

      Apologies for a 2nd email, but I thought that by splitting out topics
      into threads, it might be easier for people to refer to these in the
      future.

      I'm running Unslung 6.10 beta. I have two hard drives, one natively
      formatted on port 2, and a 1TB FAT 32 formatted drive on port 1
      (Samsung HD103UJ). The Enclosure for port 1 is a Welland Greenstar:
      http://www.welland.com.tw/html/green/740pss.html.

      I got it since it was recommended on the wiki:
      http://www.nslu2-linux.org/wiki/FAQ/SpinDownUSBHarddisks#method2

      The Enclosure has 4 power states:

      1. On - Read / Write normal RPM
      2. Idle - reduced RPM
      3. Spindown
      4. 'Power Off'

      It would appear that this power off actually turns the power to the
      disc off in such a way that it disconnects the disc from the USB. When
      it's connected to an XP machine and the disc is accessed when in this
      'Power Off' mode it makes the sound of a USB device disconnected and
      then reconnected, before you can access the files.

      When connected to the NSLU2, at boot it mounts it as /dev/sdb and
      HDD_1_1_1. When the drive is subsequently accessed and it's in this
      'Power Off' state, it then mounts it as /dev/sdc and HDD_1_2_1. If the
      drive powers off again, next time it will appear as /dev/sdd and
      HDD_1_3_1, and so on.

      Unfortunately there's no way to set the enclosure to spindown but
      never 'power-off'.

      Please does anyone know how I can force subsequent mounts to use the
      same device and HDD_1_1_1?

      Many thanks,

      Tom.
    • Carl Lowenstein
      ... I find state 2 really hard to believe. Disk drives aren t built that way. The heads fly dynamically on a thin layer of air, and the drive platters must
      Message 2 of 8 , Jul 17, 2008
      • 0 Attachment
        On Sun, Jul 13, 2008 at 2:01 PM, tjrundy <tjrundy@...> wrote:
        > Hello,
        >
        > Apologies for a 2nd email, but I thought that by splitting out topics
        > into threads, it might be easier for people to refer to these in the
        > future.
        >
        > I'm running Unslung 6.10 beta. I have two hard drives, one natively
        > formatted on port 2, and a 1TB FAT 32 formatted drive on port 1
        > (Samsung HD103UJ). The Enclosure for port 1 is a Welland Greenstar:
        > http://www.welland.com.tw/html/green/740pss.html.
        >
        > I got it since it was recommended on the wiki:
        > http://www.nslu2-linux.org/wiki/FAQ/SpinDownUSBHarddisks#method2
        >
        > The Enclosure has 4 power states:
        >
        > 1. On - Read / Write normal RPM
        > 2. Idle - reduced RPM
        > 3. Spindown
        > 4. 'Power Off'

        I find state 2 really hard to believe. Disk drives aren't built that
        way. The heads fly dynamically on a thin layer of air, and the drive
        platters must be running at full speed for this to happen. Otherwise
        you get a head crash, which is not good for the head, the platter, or
        the data.

        carl
        --
        carl lowenstein
        marine physical lab, u.c. san diego
        clowenstein@...
      • Rod Whitby
        ... See the conclusion paragraph at http://www.tomshardware.com/reviews/wd-caviar-gp,1703-8.html The main feature in WD s GreenPower portfolio is the dynamic
        Message 3 of 8 , Jul 18, 2008
        • 0 Attachment
          Carl Lowenstein wrote:
          > On Sun, Jul 13, 2008 at 2:01 PM, tjrundy <tjrundy@...> wrote:
          >> The Enclosure has 4 power states:
          >>
          >> 1. On - Read / Write normal RPM
          >> 2. Idle - reduced RPM
          >> 3. Spindown
          >> 4. 'Power Off'
          >
          > I find state 2 really hard to believe. Disk drives aren't built that
          > way. The heads fly dynamically on a thin layer of air, and the drive
          > platters must be running at full speed for this to happen. Otherwise
          > you get a head crash, which is not good for the head, the platter, or
          > the data.

          See the conclusion paragraph at
          http://www.tomshardware.com/reviews/wd-caviar-gp,1703-8.html

          "The main feature in WD's GreenPower portfolio is the dynamic adjustment
          of the drive's spindle speed. It can run at speeds between 5,400 RPM and
          7,200 RPM to save energy whenever high performance isn't needed."

          Dunno about what the enclosure in question can do, but the latest
          GreenPower WD drives apparently can reduce the RPM of the spindles to
          save power.

          -- Rod
        • tjrundy
          ... adjustment ... and ... to ... Right, so on the face of it, it seems a pretty nifty piece of kit. Now I m back from my summer hols, I m back on the case of
          Message 4 of 8 , Oct 1, 2008
          • 0 Attachment
            > See the conclusion paragraph at
            > http://www.tomshardware.com/reviews/wd-caviar-gp,1703-8.html
            >
            > "The main feature in WD's GreenPower portfolio is the dynamic
            adjustment
            > of the drive's spindle speed. It can run at speeds between 5,400 RPM
            and
            > 7,200 RPM to save energy whenever high performance isn't needed."
            >
            > Dunno about what the enclosure in question can do, but the latest
            > GreenPower WD drives apparently can reduce the RPM of the spindles
            to
            > save power.
            >
            > -- Rod

            Right,

            so on the face of it, it seems a pretty nifty piece of kit. Now I'm
            back from my summer hols, I'm back on the case of trying to get this
            HDD working with Openslug.

            Remember my original problem was:

            When connected to the NSLU2, at boot it mounts it as /dev/sdb and
            HDD_1_1_1. When the drive is subsequently accessed and it's in this
            'Power Off' state, it then mounts it as /dev/sdc and HDD_1_2_1. If the
            drive powers off again, next time it will appear as /dev/sdd and
            HDD_1_3_1, and so on.

            Unfortunately there's no way to set the enclosure to spindown but
            never 'power-off'.

            Please does anyone know how I can force subsequent mounts to use the
            same device and HDD_1_1_1?

            Does anyone have any idea about this? Please, Pretty please? ;)
          • Mike (mwester)
            tjrundy wrote: ... OpenSlug (aka SlugOS) or Unslung? Your comments below seem to imply Unslung. ... If this is, in fact, Unslung, it s pretty unlikely that
            Message 5 of 8 , Oct 1, 2008
            • 0 Attachment
              tjrundy wrote:
              ...
              > so on the face of it, it seems a pretty nifty piece of kit. Now I'm
              > back from my summer hols, I'm back on the case of trying to get this
              > HDD working with Openslug.

              OpenSlug (aka SlugOS) or Unslung? Your comments below seem to imply
              Unslung.

              > Remember my original problem was:
              >
              > When connected to the NSLU2, at boot it mounts it as /dev/sdb and
              > HDD_1_1_1. When the drive is subsequently accessed and it's in this
              > 'Power Off' state, it then mounts it as /dev/sdc and HDD_1_2_1. If the
              > drive powers off again, next time it will appear as /dev/sdd and
              > HDD_1_3_1, and so on.
              >
              > Unfortunately there's no way to set the enclosure to spindown but
              > never 'power-off'.
              >
              > Please does anyone know how I can force subsequent mounts to use the
              > same device and HDD_1_1_1?
              >
              > Does anyone have any idea about this? Please, Pretty please? ;)

              If this is, in fact, Unslung, it's pretty unlikely that you'll get this
              to work the way you want it. The fundamental problem is that the
              mechanism that handles this is proprietary Linksys code.

              For SlugOS (and other distros like Debian) it should just work -- udev
              should take care of the situation, and if you are mounting by UUID or
              label, it doesn't matter if it gets a new /dev node.

              Mike (mwester)
            • tjrundy
              ... this ... udev ... or ... Thanks for your help mike. I re-checked my notes, I changed to SlugOS 4.8 beta after I posted that message, but the problem
              Message 6 of 8 , Oct 1, 2008
              • 0 Attachment
                > If this is, in fact, Unslung, it's pretty unlikely that you'll get
                this
                > to work the way you want it. The fundamental problem is that the
                > mechanism that handles this is proprietary Linksys code.
                >
                > For SlugOS (and other distros like Debian) it should just work --
                udev
                > should take care of the situation, and if you are mounting by UUID
                or
                > label, it doesn't matter if it gets a new /dev node.
                >
                > Mike (mwester)
                >

                Thanks for your help mike. I re-checked my notes, I changed to SlugOS
                4.8 beta after I posted that message, but the problem persists. Just
                to incite a little more help, here's a copy of the log files when I
                plug in the device:

                Oct 1 21:40:13 (none) syslog.info -- MARK --
                Oct 1 21:45:01 (none) user.info kernel: usb 1-2: new high speed USB
                device usin g ehci_hcd and
                address 3
                Oct 1 21:45:01 (none) user.err kernel: usb 1-2: device descriptor
                read/64, erro r -71
                Oct 1 21:45:01 (none) user.info kernel: usb 1-2: configuration #1
                chosen from 1 choice
                Oct 1 21:45:01 (none) user.info kernel: scsi1 : SCSI emulation for
                USB Mass Sto rage devices
                Oct 1 21:45:01 (none) user.debug kernel: usb-storage: device found at
                3
                Oct 1 21:45:01 (none) user.debug kernel: usb-storage: waiting for
                device to set tle before
                scanning
                Oct 1 21:45:06 (none) user.notice kernel: scsi 1:0:0:0: Direct-Access
                SAMSU NG HD103UJ
                PQ: 0 ANSI: 2 CCS
                Oct 1 21:45:06 (none) user.notice kernel: SCSI device sdb: 1953525168
                512-byte hdwr sectors
                (1000205 MB)
                Oct 1 21:45:06 (none) user.notice kernel: sdb: Write Protect is off
                Oct 1 21:45:06 (none) user.debug kernel: sdb: Mode Sense: 00 38 00 00
                Oct 1 21:45:06 (none) user.err kernel: sdb: assuming drive cache:
                write through
                Oct 1 21:45:06 (none) user.notice kernel: SCSI device sdb: 1953525168
                512-byte hdwr sectors
                (1000205 MB)
                Oct 1 21:45:06 (none) user.notice kernel: sdb: Write Protect is off
                Oct 1 21:45:06 (none) user.debug kernel: sdb: Mode Sense: 00 38 00 00
                Oct 1 21:45:06 (none) user.err kernel: sdb: assuming drive cache:
                write through
                Oct 1 21:45:08 (none) user.info kernel: usb 1-2: USB disconnect,
                address 3
                Oct 1 21:45:08 (none) user.info kernel: sd 1:0:0:0: SCSI error:
                return code = 0 x00010000
                Oct 1 21:45:08 (none) user.warn kernel: end_request: I/O error, dev
                sdb, sector 0
                Oct 1 21:45:08 (none) user.err kernel: Buffer I/O error on device
                sdb, logical block 0
                Oct 1 21:45:08 (none) user.info kernel: sd 1:0:0:0: SCSI error:
                return code = 0 x00010000
                Oct 1 21:45:08 (none) user.warn kernel: end_request: I/O error, dev
                sdb, sector 0
                Oct 1 21:45:08 (none) user.err kernel: Buffer I/O error on device
                sdb, logical block 0
                Oct 1 21:45:08 (none) user.warn kernel: unable to read partition
                table
                Oct 1 21:45:08 (none) user.notice kernel: sd 1:0:0:0: Attached scsi
                disk sdb
                Oct 1 21:45:08 (none) user.debug kernel: usb-storage: device scan
                complete
                Oct 1 21:45:08 (none) user.notice root: udev/mount.sh [/dev/sdb] is
                blacklisted , ignoring
                Oct 1 21:45:09 (none) user.notice root: udev/mount.sh [/dev/sdb] is
                blacklisted , ignoring
                Oct 1 21:45:18 (none) user.info kernel: usb 1-2: new high speed USB
                device usin g ehci_hcd and
                address 4
                Oct 1 21:45:19 (none) user.err kernel: usb 1-2: device descriptor
                read/64, erro r -71
                Oct 1 21:45:19 (none) user.info kernel: usb 1-2: configuration #1
                chosen from 1 choice
                Oct 1 21:45:19 (none) user.info kernel: scsi2 : SCSI emulation for
                USB Mass Sto rage devices
                Oct 1 21:45:19 (none) user.debug kernel: usb-storage: device found at
                4
                Oct 1 21:45:19 (none) user.debug kernel: usb-storage: waiting for
                device to set tle before
                scanning
                Oct 1 21:45:24 (none) user.notice kernel: scsi 2:0:0:0: Direct-Access
                SAMSUNG HD103UJ PQ: 0 ANSI: 2 CCS
                Oct 1 21:45:24 (none) user.notice kernel: SCSI device sdb: 1953525168
                512-byte hdwr sectors (1000205 MB)
                Oct 1 21:45:24 (none) user.notice kernel: sdb: Write Protect is off
                Oct 1 21:45:24 (none) user.debug kernel: sdb: Mode Sense: 00 38 00 00
                Oct 1 21:45:24 (none) user.err kernel: sdb: assuming drive cache:
                write through
                Oct 1 21:45:24 (none) user.notice kernel: SCSI device sdb: 1953525168
                512-byte hdwr sectors (1000205 MB)
                Oct 1 21:45:24 (none) user.notice kernel: sdb: Write Protect is off
                Oct 1 21:45:24 (none) user.debug kernel: sdb: Mode Sense: 00 38 00 00
                Oct 1 21:45:24 (none) user.err kernel: sdb: assuming drive cache:
                write through
                Oct 1 21:45:25 (none) user.info kernel: sdb1
                Oct 1 21:45:25 (none) user.notice kernel: sd 2:0:0:0: Attached scsi
                disk sdb
                Oct 1 21:45:25 (none) user.debug kernel: usb-storage: device scan
                complete
                Oct 1 21:45:25 (none) user.notice root: udev/mount.sh [/dev/sdb] is
                blacklisted, ignoring
                Oct 1 21:45:25 (none) user.err kernel: FAT: utf8 is not a recommended
                IO charset for FAT filesystems, filesystem will be case sensitive!
                Oct 1 21:45:26 (none) user.notice root: mount.sh/automount Auto-mount
                of [/media/sdb1] successful

                if I run fdisk:

                root@storage:~$ fdisk -l

                Disk /dev/sda: 80.0 GB, 80026361856 bytes
                255 heads, 63 sectors/track, 9729 cylinders
                Units = cylinders of 16065 * 512 = 8225280 bytes

                Device Boot Start End Blocks Id System
                /dev/sda1 1 366 2939863+ 83 Linux
                /dev/sda2 367 9670 74734380 83 Linux
                /dev/sda3 9671 9729 473917+ 83 Linux

                Disk /dev/sdb: 1000.2 GB, 1000204886016 bytes
                255 heads, 63 sectors/track, 121601 cylinders
                Units = cylinders of 16065 * 512 = 8225280 bytes

                Device Boot Start End Blocks Id System
                /dev/sdb1 1 121601 976760001 c Win95 FAT32
                (LBA)

                Now I wait for 10 minutes for the disk drive to go into its low power
                state...

                now I try and access it:

                root@storage:/etc$ mount
                rootfs on / type rootfs (rw)
                /dev/root on /initrd type jffs2 (ro)
                /dev/sda1 on / type ext3 (rw,data=ordered)
                proc on /proc type proc (rw)
                sysfs on /sys type sysfs (rw)
                /dev/sda1 on /dev/.static/dev type ext3 (rw,data=ordered)
                tmpfs on /dev type tmpfs (rw)
                /dev/sda2 on /home type ext3 (rw,data=ordered)
                tmpfs on /var/volatile type tmpfs (rw)
                tmpfs on /dev/shm type tmpfs (rw)
                usbfs on /proc/bus/usb type usbfs (rw)
                devpts on /dev/pts type devpts (rw)
                /dev/sdb1 on /media/sdb1 type vfat
                (rw,sync,fmask=0022,dmask=0022,codepage=cp437,iocharset=utf8)
                root@storage:/etc$ cd /media
                root@storage:/media$ ls
                card hdd net realroot sda2 sdc1 union
                cf mmc1 ram sda1 sdb1 sdd1
                root@storage:/media$ cd sdb1
                root@storage:/media/sdb1$ ls
                ls: ./Music: Input/output error
                ls: ./Software: Input/output error
                ls: ./Films: Input/output error
                System Volume Information
                root@storage:/media/sdb1$

                and the messages log contains lots of rejection I/O to dead device:

                Oct 1 22:18:49 (none) user.err kernel: scsi 3:0:0:0: rejecting I/O to
                dead device
                Oct 1 22:18:49 (none) user.err kernel: FAT: Directory bread(block
                477041) failed
                Oct 1 22:18:50 (none) user.err kernel: scsi 3:0:0:0: rejecting I/O to
                dead device
                Oct 1 22:18:50 (none) user.err kernel: FAT: FAT read failed (blocknr
                4906)
                Oct 1 22:18:50 (none) user.err kernel: scsi 3:0:0:0: rejecting I/O to
                dead device
                Oct 1 22:18:50 (none) user.err kernel: FAT: FAT read failed (blocknr
                4906)
                Oct 1 22:18:50 (none) user.err kernel: scsi 3:0:0:0: rejecting I/O to
                dead device
                Oct 1 22:18:50 (none) user.err kernel: FAT: FAT read failed (blocknr
                4906)
                Oct 1 22:18:50 (none) user.notice root: mount.sh/automount Not
                removing non-empty directory [/media/sdb1]
                Oct 1 22:18:50 (none) user.notice root: udev/mount.sh [/dev/sdb] is
                blacklisted, ignoring
                Oct 1 22:18:57 (none) user.info kernel: usb 1-2: new high speed USB
                device using ehci_hcd and address 6
                Oct 1 22:18:57 (none) user.info kernel: usb 1-2: configuration #1
                chosen from 1 choice
                Oct 1 22:18:57 (none) user.info kernel: scsi4 : SCSI emulation for
                USB Mass Storage devices
                Oct 1 22:18:57 (none) user.debug kernel: usb-storage: device found at
                6
                Oct 1 22:18:57 (none) user.debug kernel: usb-storage: waiting for
                device to settle before scanning
                Oct 1 22:19:02 (none) user.notice kernel: scsi 4:0:0:0: Direct-Access
                SAMSUNG HD103UJ PQ: 0 ANSI: 2 CCS
                Oct 1 22:19:02 (none) user.notice kernel: SCSI device sdc: 1953525168
                512-byte hdwr sectors (1000205 MB)
                Oct 1 22:19:02 (none) user.notice kernel: sdc: Write Protect is off
                Oct 1 22:19:02 (none) user.debug kernel: sdc: Mode Sense: 00 38 00 00
                Oct 1 22:19:02 (none) user.err kernel: sdc: assuming drive cache:
                write through
                Oct 1 22:19:02 (none) user.notice kernel: SCSI device sdc: 1953525168
                512-byte hdwr sectors (1000205 MB)
                Oct 1 22:19:02 (none) user.notice kernel: sdc: Write Protect is off
                Oct 1 22:19:02 (none) user.debug kernel: sdc: Mode Sense: 00 38 00 00
                Oct 1 22:19:02 (none) user.err kernel: sdc: assuming drive cache:
                write through
                Oct 1 22:19:03 (none) user.info kernel: sdc1
                Oct 1 22:19:03 (none) user.notice kernel: sd 4:0:0:0: Attached scsi
                disk sdc
                Oct 1 22:19:03 (none) user.debug kernel: usb-storage: device scan
                complete
                Oct 1 22:19:03 (none) user.notice root: udev/mount.sh [/dev/sdc] is
                blacklisted, ignoring
                Oct 1 22:19:04 (none) user.err kernel: FAT: utf8 is not a recommended
                IO charset for FAT filesystems, filesystem will be case sensitive!
                Oct 1 22:19:04 (none) user.notice root: mount.sh/automount Auto-mount
                of [/media/sdc1] successful

                and re-mounts it on sdc1!

                So it must take about 15 seconds to spin up out of its low power
                state. So this causes havoc with Samba, naturally because the shares
                don't reference the right device now!

                Sorry to be a continued pain, but anyone have any thoughts on how to
                fix this?

                Cheers,

                Tom.

                ps. the Mrs isn't too keen on having hdd's spinning all the time, esp
                when I'm away!
              • tjrundy
                ... Further to Mike s suggestion, I spent a couple of hours last night trying to get the UUID of the disk (which was only 8 chars long as it s VFAT formatted),
                Message 7 of 8 , Oct 2, 2008
                • 0 Attachment
                  --- In nslu2-general@yahoogroups.com, "tjrundy" <tjrundy@...> wrote:
                  >
                  >
                  > > If this is, in fact, Unslung, it's pretty unlikely that you'll get
                  > this
                  > > to work the way you want it. The fundamental problem is that the
                  > > mechanism that handles this is proprietary Linksys code.
                  > >
                  > > For SlugOS (and other distros like Debian) it should just work --
                  > udev
                  > > should take care of the situation, and if you are mounting by UUID
                  > or
                  > > label, it doesn't matter if it gets a new /dev node.
                  > >
                  > > Mike (mwester)
                  > >

                  Further to Mike's suggestion, I spent a couple of hours last night
                  trying to get the UUID of the disk (which was only 8 chars long as
                  it's VFAT formatted), and then trying to mount it by the UUID, or the
                  label. Neither worked, and I noticed that much of the web-material
                  about this was referenced to the Debian dist...

                  I'm thinking that perhaps Debian might be a better bet?

                  However before I start all over again, can anyone confirm whether UUID
                  support only works for ext2/3 and if it's supported under VFAT how I
                  would get a 'proper' UUID string?

                  Thanks again guys,

                  Tom.
                • Mike (mwester)
                  tjrundy wrote: ... Check out blkid -- it puts the current list of UUIDs in /etc/blkid.tab (or similar) Mike (mwester)
                  Message 8 of 8 , Oct 6, 2008
                  • 0 Attachment
                    tjrundy wrote:
                    ...
                    > However before I start all over again, can anyone confirm whether UUID
                    > support only works for ext2/3 and if it's supported under VFAT how I
                    > would get a 'proper' UUID string?

                    Check out "blkid" -- it puts the current list of UUIDs in /etc/blkid.tab
                    (or similar)

                    Mike (mwester)
                  Your message has been successfully submitted and would be delivered to recipients shortly.