e280r: panic[cpu1] UE WDU Error(s) - SUN

This is a discussion on e280r: panic[cpu1] UE WDU Error(s) - SUN ; hello all, my E280R has panic-ed tonight, and now, unable to reboot it: Sun Fire 280R (2 X UltraSPARC-III+) , No Keyboard Copyright 1998-2002 Sun Microsystems, Inc. All rights reserved. OpenBoot 4.5, 3072 MB memory installed, Serial #50922130. Ethernet address ...

+ Reply to Thread
Results 1 to 6 of 6

Thread: e280r: panic[cpu1] UE WDU Error(s)

  1. e280r: panic[cpu1] UE WDU Error(s)

    hello all,
    my E280R has panic-ed tonight, and now, unable to reboot it:
    Sun Fire 280R (2 X UltraSPARC-III+) , No Keyboard
    Copyright 1998-2002 Sun Microsystems, Inc. All rights reserved.
    OpenBoot 4.5, 3072 MB memory installed, Serial #50922130.
    Ethernet address 0:3:ba:9:2:92, Host ID: 83090292.

    Boot device: disk File and args:
    Loading ufs-file-system package 1.4 04 Aug 1995 13:02:54.
    FCode UFS Reader 1.12 00/07/17 15:48:16.
    Loading: /platform/SUNW,Sun-Fire-280R/ufsboot
    Loading: /platform/sun4u/ufsboot
    Corrected ECC Error
    {0} ok

    thru the RSC, i saw that erros was due to:

    SUNW-MSG-ID: SUNOS-8000-0G, TYPE: Error, VER: 1, SEVERITY: Major
    EVENT-TIME: 0x476493df.0x2fefe0 (0x3f21d061a2b10)
    PLATFORM: SUNW,Sun-Fire-280R, CSN: -, HOSTNAME: scr
    SOURCE: SunOS, REV: 5.10 Generic_127111-03
    DESC: Errors have been detected that require a reboot to ensure system
    integrity. See http://www.sun.com/msg/SUNOS-8000-0G for more information.
    AUTO-RESPONSE: Solaris will attempt to save and diagnose the error telemetry
    IMPACT: The system will sync files, save a crash dump if needed, and reboot
    REC-ACTION: Save the error summary below in case telemetry cannot be saved


    ereport.cpu.ultraSPARC-IIIplus.ue ena=21d06180e7000401 detector=[ version=1

    scheme="cpu" cpuid=1 cpumask=22 serial="112D09014E5" ] afsr=1000040000019c

    afar-status=1 afar=bc9c6690 pc=12191c4 tl=0 tt=32 privileged=1 multiple=0

    syndrome-status=1 syndrome=19c error-type="U" error-disposition=0

    l2-cache-ways=1 l2-cache-data=[...] dcache-ways=0 icache-ways=0 resource=[

    version=0 scheme="mem" unum="J0100 J0202 J0304 J0406" ]

    Any chance to have this machine rebooted?

    Thanks in advance,

    gerard

  2. Re: e280r: panic[cpu1] UE WDU Error(s)

    "jdh13" wrote in message
    news:4764d9bd$0$27284$426a74cc@news.free.fr...
    > hello all,
    > my E280R has panic-ed tonight, and now, unable to reboot it:
    > Sun Fire 280R (2 X UltraSPARC-III+) , No Keyboard
    > Copyright 1998-2002 Sun Microsystems, Inc. All rights reserved.
    > OpenBoot 4.5, 3072 MB memory installed, Serial #50922130.
    > Ethernet address 0:3:ba:9:2:92, Host ID: 83090292.
    >
    > Boot device: disk File and args:
    > Loading ufs-file-system package 1.4 04 Aug 1995 13:02:54.
    > FCode UFS Reader 1.12 00/07/17 15:48:16.
    > Loading: /platform/SUNW,Sun-Fire-280R/ufsboot
    > Loading: /platform/sun4u/ufsboot
    > Corrected ECC Error
    > {0} ok


    If I had to guess I'd say you have 4x256 MB DIMMs and 4x512 MB DIMMs as well
    as 2 CPUs.
    You can try turning the key to the diag position and see if anything fails
    during POST.
    The OBP is at 4.5 and 4.16.4 is the latest, but you can't flash it if you
    can't boot.

    If nothing fails POST, then try removing whatever modules are in slots J0100
    J0202 J0304 J0406.
    This is all of Group 0 which is every other DIMM starting closest to the
    CPUs.
    If that fixes the problem, then at least you'll know it's one of those 4.
    Flashing to the latest OBP firmware and running max diags with the memory
    back in at that point may find which is the fault one.

    If it's not the memory, then try removing CPU1 since that seems to be the
    one called out as far as processors go.

    Trinean



  3. Re: e280r: panic[cpu1] UE WDU Error(s)

    Trinean wrote:

    >
    > If I had to guess I'd say you have 4x256 MB DIMMs and 4x512 MB DIMMs as well
    > as 2 CPUs.
    > You can try turning the key to the diag position and see if anything fails
    > during POST.
    > The OBP is at 4.5 and 4.16.4 is the latest, but you can't flash it if you
    > can't boot.
    >
    > If nothing fails POST, then try removing whatever modules are in slots J0100
    > J0202 J0304 J0406.
    > This is all of Group 0 which is every other DIMM starting closest to the
    > CPUs.
    > If that fixes the problem, then at least you'll know it's one of those 4.
    > Flashing to the latest OBP firmware and running max diags with the memory
    > back in at that point may find which is the fault one.
    >
    > If it's not the memory, then try removing CPU1 since that seems to be the
    > one called out as far as processors go.
    >


    thanks for your answer. I will do it tomorrow.

    gerard

  4. Re: e280r: panic[cpu1] UE WDU Error(s)

    Trinean wrote:


    >
    > If nothing fails POST, then try removing whatever modules are in slots J0100
    > J0202 J0304 J0406.
    > This is all of Group 0 which is every other DIMM starting closest to the
    > CPUs.
    > If that fixes the problem, then at least you'll know it's one of those 4.
    > Flashing to the latest OBP firmware and running max diags with the memory
    > back in at that point may find which is the fault one.
    >
    > If it's not the memory, then try removing CPU1 since that seems to be the
    > one called out as far as processors go.
    >


    after removing CPU1 and DIMM J100 J0202 J0304 J0406, the machine accept
    to boot normally!
    very thanks for your help,

    gerard

    PS: i have the (false?) impression that E280r aren't the best machine,
    because since i bought it six years ago, i had many problems with it...

  5. Re: e280r: panic[cpu1] UE WDU Error(s)

    On 2007-12-18, jdh13 wrote:
    > Trinean wrote:
    >
    >
    >>
    >> If nothing fails POST, then try removing whatever modules are in slots J0100
    >> J0202 J0304 J0406.
    >> This is all of Group 0 which is every other DIMM starting closest to the
    >> CPUs.
    >> If that fixes the problem, then at least you'll know it's one of those 4.
    >> Flashing to the latest OBP firmware and running max diags with the memory
    >> back in at that point may find which is the fault one.
    >>
    >> If it's not the memory, then try removing CPU1 since that seems to be the
    >> one called out as far as processors go.


    [ ... ]

    > after removing CPU1 and DIMM J100 J0202 J0304 J0406, the machine accept
    > to boot normally!


    Did you try just removing the DIMMs and leaving the CPU in
    place? Or removing the CPU and leaving the DIMMs in place? it would be
    interesting to determine which is bad.

    My wife's Sun Blade 1000 crashed the CDE window manager a couple
    of times, and the diagnostic information pointed me to a specific DIMM
    (one of four of the 256 MB ones -- I also had four 128 MB DIMMs there.
    I pulled the 256 MB ones and it worked fine. I then got a set of four
    512 MB ones and put them into my machine, freeing four 256 MB DIMMs to
    go into my wife's machine, and now everyone is happy. (She does not do
    image processing, so she does not need as much RAM as I do. Now, all I
    need to do is to pick up one more 256 MB DIMM and her total RAM can go
    up from 1.5 GB to 2.0 GB. I've carefully marked the one which it
    complained about. The diagnostics in that were excellent.

    And when you have serious problems, the best thing to do is to
    strip the system down to the minimum (one CPU, one bank of DIMMs, and no
    PCI cards, and boot it with a serial terminal on TTYA. Once that is
    known to work, start adding things back, one at a time, until the
    failure returns.

    [ ... ]

    > PS: i have the (false?) impression that E280r aren't the best machine,
    > because since i bought it six years ago, i had many problems with it...


    Is the E280r the same as the Sun Fire 280R? Same system board
    as a Sun Blade 2000, with two 900 MHz Cu CPUs, RSC board (and in my case
    a low-end framebuffer a dual LVD SCSI card, a dual HVD SCSI card, and a
    Crypto card?

    I have one of those (Sun Fire 280R), which I picked up at a
    hamfest for $250.00, and it has been serving quite well -- with 4GB
    total RAM, a couple of 147 GB disks internally (separate purchase), and
    a lot of 18 GB disks formed into two zfs filesystems.

    Ive got no complaints about mine. Perhaps you have had a flakey
    DIMM or CPU for quite a while, and it has finally gotten bad enough to
    require removal or replacement.

    The only complaint about mine might be that the fans are a bit
    noisy. :-)

    Good Luck,
    DoN.

    --
    Email: | Voice (all times): (703) 938-4564
    (too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
    --- Black Holes are where God is dividing by zero ---

  6. Re: e280r: panic[cpu1] UE WDU Error(s)

    DoN. Nichols wrote:

    >
    > Is the E280r the same as the Sun Fire 280R? Same system board
    > as a Sun Blade 2000, with two 900 MHz Cu CPUs, RSC board (and in my case
    > a low-end framebuffer a dual LVD SCSI card, a dual HVD SCSI card, and a
    > Crypto card?
    >


    yes, i think it's the same, but i haven't a crypto card.
    Very thanks for your suggstions, but now, the machine is again in
    production, so i'll do the tests later

    gerard

+ Reply to Thread