Professional Web Applications Themes

SCO 5.0.7 - 73Gb Hard disk appears as 4Gb - SCO

Hi, Although I've seen a post from Bela, I'm posting this just in case anyone might have any idea Got an LSI Card (1020) with a 73 Gb Fujitsu disk. After loading the BTLD containing the lsil driver version 10.03.26, the disk was recognized with 69Gb. So far so good... Later on instalation, divvy reports only 4 Gb. After several tries, placed an Adaptec 39160 (ad160), low level initialized the disk from the controller bios, since dparam didn't work, and tried to install once again. Same thing happened with the Adaptec card, leaving me quite confused. I was convinced the ...

  1. #1

    Default SCO 5.0.7 - 73Gb Hard disk appears as 4Gb

    Hi,

    Although I've seen a post from Bela, I'm posting this just in case anyone
    might have any idea

    Got an LSI Card (1020) with a 73 Gb Fujitsu disk.
    After loading the BTLD containing the lsil driver version 10.03.26, the disk
    was recognized with 69Gb. So far so good... Later on instalation, divvy
    reports only 4 Gb.

    After several tries, placed an Adaptec 39160 (ad160), low level initialized
    the disk from the controller bios, since dparam didn't work, and tried to
    install once again.

    Same thing happened with the Adaptec card, leaving me quite confused. I was
    convinced the initial setup would be a LSILogic driver problem. Or not?

    Putting biosgeom in the bootstring didn't work also. Panic rigth after the
    post install boot.

    Sometimes I end up with a 2Gb harddisk.

    With a 36 Gb hd, everything works fine. The problem is that I really need
    73Gb.

    TIA

    Paulo Couceiro


    Paulo Guest

  2. #2

    Default Re: SCO 5.0.7 - 73Gb Hard disk appears as 4Gb

    Paulo Couceiro wrote:
     

    What you're seeing is a known problem with "lsil" versions 1.03.26 and
    earlier. If you ask LSI tech support for 1.03.27, they're supposed to
    have a way to get you a copy of it even though it hasn't gotten through
    all their testing yet.
     

    It was initially an "lsil" problem. I'm not sure how it propagated to
    the "ad160" setup. One possibility: low-level formatting in the BIOS
    doesn't necessarily do anything. If you had written a dparam stamp to
    the disk under "lsil", it might have persisted even through low-level
    format; then it would have affected "ad160". If that's the cause, you
    should be able to correct it by overwriting the masterboot sector:

    # dd if=/dev/zero of=/dev/rhd00 bs=512 count=1 ### DANGER!!!

    Only do this if you intend to lose all contents of the disk!!!
     

    You should be able to get it. Start with a zero'd masterboot sector,
    then use either "ad160" or an updated "lsil".
     
    Bela Guest

  3. #3

    Default Re: SCO 5.0.7 - 73Gb Hard disk appears as 4Gb

    "Paulo Couceiro" <telepac.pt> wrote in message
    news:3f5672cf$0$2269$telepac.pt... 
    disk 
    initialized 
    was 
    the 

    with which version of SCO?


    Stuart Guest

  4. #4

    Default Re: SCO 5.0.7 - 73Gb Hard disk appears as 4Gb

    > > With a 36 Gb hd, everything works fine. The problem is that I really
    need 
    >
    > with which version of SCO?[/ref]

    My apologies.. I'm just blind..


    Stuart Guest

  5. #5

    Default Re: SCO 5.0.7 - 73Gb Hard disk appears as 4Gb - SOLVED

    Sorry for the long post...


    FYI: This is a Fujitsu HD model MAP3735NC wich has the following geometry:
    8924 Cyl 255 Heads, 63 Sec

    Had to use a third party utilitiy to zero out the MBR. (Seagate's Disk
    Wizard), since it was still reporting a 4 Gb HD

    Boot with biosgeom=(8924,255,63) in the bootstring

    After the post install boot, where it would usually panic, tried biosgeom
    once again. All went fine.

    After booting normally, I've edited /etc/default/boot ant inserted biosgeom
    into the bootstring.

    It worked fine with both Adaptec 39160 SCSI controller (ad160) and LSILogic
    1020/30 SCSI (lsil)

    Note: the driver used for lsil was 1.03.26, the one supposed to be buggy...

    Event though I've overcome this problem, I'd like some advice on any
    possible disadvantage using the method described.


    Thank you all for the help


    Paulo Couceiro



     
    disk 
    initialized 
    was 


    Paulo Guest

  6. #6

    Default Re: SCO 5.0.7 - 73Gb Hard disk appears as 4Gb - SOLVED

    Paulo Couceiro wrote:
     

    That geometry is really just a figment of OSR5's imagination. But it's
    a very dearly held figment, you have to get it right...
     

    You should have been able to do that natively with OSR5, but good that
    it's done, anyway.
     

    The 1.03.26 bug is simply that it automatically calculates a bad
    geometry. By overriding it, you are achieving much the same results as
    you would have gotten with the newer driver.

    I'm still puzzled about "ad160". Did you try an install in which you
    had zeroed the masterboot record, then installed with "ad160" without
    "biosgeom"? It should have natively come up with a good geometry.
    Um... unless the 39160 BIOS settings tell it to do otherwise. I forget
    exactly what they are, but that BIOS has settings like "extended DOS
    translation for large disks". There may be more than one setting that
    seem to relate to disk size and/or geometry translation. There exists a
    combination of those settings which would have allowed you to install
    with "ad160", without "biosgeom".
     

    Using a "biosgeom" bootstring may in some cases cause the OS to insist
    on the same geometry for _all_ disks in the system. You don't want
    that.

    You should be able to fix it now by booting up, running:

    dparam -w /dev/rhd00
    dparam /dev/rhd00 `dparam /dev/rhd00`

    and then rebooting without "biosgeom". Except this will not work on
    OSR507 as shipped; it is unable to stamp parameters, due to a bug in
    masterboot. That bug is fixed in OSR507UP1 (not MP1), and will also be
    fixed in the upcoming "wd" update (which will work on OSR507 with or
    without UP1).
     
    Bela Guest

  7. #7

    Default Re: SCO 5.0.7 - 73Gb Hard disk appears as 4Gb - SOLVED


    "Bela Lubkin" <com> wrote in message
    news:com... [/ref]
    geometry: 
    >
    > That geometry is really just a figment of OSR5's imagination. But it's
    > a very dearly held figment, you have to get it right...

    >
    > You should have been able to do that natively with OSR5, but good that
    > it's done, anyway.
    > [/ref]
    biosgeom [/ref]
    biosgeom [/ref]
    LSILogic [/ref]
    buggy... 

    Yes, I tried with a zeroed MBR. The 39160 controller reported 4471 cyl, 64
    h,32 sec as the LSI Card did. This happened ONLY after the install has
    finished.
    The extended bios translation was off in the controller Bios.

    Could this be caused by this particular disk model and brand?
     
    >
    > Using a "biosgeom" bootstring may in some cases cause the OS to insist
    > on the same geometry for _all_ disks in the system. You don't want
    > that.
    >
    > You should be able to fix it now by booting up, running:
    >
    > dparam -w /dev/rhd00
    > dparam /dev/rhd00 `dparam /dev/rhd00`
    >
    > and then rebooting without "biosgeom". Except this will not work on
    > OSR507 as shipped; it is unable to stamp parameters, due to a bug in
    > masterboot. That bug is fixed in OSR507UP1 (not MP1), and will also be
    > fixed in the upcoming "wd" update (which will work on OSR507 with or
    > without UP1).[/ref]

    I'll try that during the next few days. I'll keep the follow up in this
    thread.

    Paulo Couceiro [/ref]


    Paulo Guest

  8. #8

    Default Re: SCO 5.0.7 - 73Gb Hard disk appears as 4Gb - SOLVED

    Paulo Couceiro wrote:
     [/ref][/ref]
     

    That's not clear to me. Do you mean that during ISL (booting from the
    CD) both HBAs showed a geometry of 8924/255/63, but both switched to
    4471/64/32 when booting from the hard disk? Or more likely:

    ISL Boot from hard disk
    lsil 4471/64/32 4471/64/32
    ad160 8924/255/63 4471/64/32

    Is that what you saw?
     

    Here you're talking about the 39160 BIOS, right? I seem to remember
    there are at least two different settings that can affect its idea of
    geometry.
     

    I don't think so. Neither the HBA driver nor the OS care at all what
    kind of disk it is; in determining "geometry", the only information used
    from the _drive_ is its response to the question: "how big are you in
    blocks?" Note: "ad160" also uses information it gets from the
    controller _BIOS_ -- information that's influenced by "extended
    translation" and other settings.
     
    Bela Guest

  9. #9

    Default Re: SCO 5.0.7 - 73Gb Hard disk appears as 4Gb - SOLVED

    This is what I saw.

    ISL Boot from hard disk
    lsil 4471/64/32 4471/64/32
    ad160 4471/64/32 4471/64/32

    I managed to repeat the whole process from the very beggining. Zeroed the
    MBR, attached the HD to the 39160. Seagate's utility returns CHS as
    8924,255,63. No partitions defined.

    ad160 returns the info above. Extended translation for DOS Drives>1Gbyte is
    disabled.

    Go to same steps once again, this time with HD on LSI controller. Exact
    same results.

    Paulo Couceiro



    "Bela Lubkin" <com> wrote in message
    news:com... [/ref]
    geometry: 
    >
    > That geometry is really just a figment of OSR5's imagination. But it's
    > a very dearly held figment, you have to get it right...

    >
    > You should have been able to do that natively with OSR5, but good that
    > it's done, anyway.
    > [/ref]
    biosgeom [/ref]
    biosgeom [/ref]
    LSILogic [/ref]
    buggy... 
    >
    > Using a "biosgeom" bootstring may in some cases cause the OS to insist
    > on the same geometry for _all_ disks in the system. You don't want
    > that.
    >
    > You should be able to fix it now by booting up, running:
    >
    > dparam -w /dev/rhd00
    > dparam /dev/rhd00 `dparam /dev/rhd00`
    >
    > and then rebooting without "biosgeom". Except this will not work on
    > OSR507 as shipped; it is unable to stamp parameters, due to a bug in
    > masterboot. That bug is fixed in OSR507UP1 (not MP1), and will also be
    > fixed in the upcoming "wd" update (which will work on OSR507 with or
    > without UP1).
    > [/ref]


    Paulo Guest

Similar Threads

  1. RH8 hard disk install
    By Alok in forum Linux Setup, Configuration & Administration
    Replies: 12
    Last Post: September 22nd, 03:59 AM
  2. What is LVD hard disk ?
    By Jose Perez in forum AIX
    Replies: 1
    Last Post: September 15th, 03:08 PM
  3. Copy DVD to hard disk?
    By mark@markscoleman.com in forum Mac Applications & Software
    Replies: 3
    Last Post: August 5th, 02:23 AM
  4. [PHP] Hard disk number
    By Adrian in forum PHP Development
    Replies: 0
    Last Post: July 27th, 11:35 AM

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139