UP2000+ Problems

[Matt Turner]

Hey Jeff,

On Fri, Feb 19, 2010 at 3:30 PM, jmd <jmd8800 at charter.net> wrote:
> hello...
>
> i have a up2000+ board in an original Powerac 320 enclosure. i am
> assuming, but not positive, this computer is all original. there are 2
> cpus @ 833mhz with 4mb cache. 512MB ram. i have 3 different video cards
> to choose from: matrox mill 8mb, ati mach 64 2mb, and elsa gloria
> synergy-II 32mb. 1 DEC network card. 1 cdrom and 1 floppy drive. I have
> removed 3 scsi adapters from the machine that were in it when i first
> bought it, as i don't need them. i can install linux with the onboard
> scsi enabled, but i must turn it off to install tru64 unix. SRM is the
> latest image available: 5.8-81
>
> jumper J31 is now closed for 21264B cpus which i have. it was not set
> previously. this controls the cpu voltage regulator.
>
> the problem is: this machine will randomly reboot. it may reboot in 3
> hours or it may reboot as SRM is initializing. i never know.
>
> *show power* command in SRM shows the uppermost line of 3.3V as FAIL.
> the lower reference to 3.3V shows OK. everything else is OK. (i don't
> have the full output in front of me, but i can get it.)

As we told you on the IRC channel, you should first try another power
supply. One known to be good.

> the diagnostic LEDS on the board show the power is good. the LEDS on the
> power supply show the power as good.
>
> also, most of the time, i have to hit the reset button once to gain
> access to keyboard, mouse and monitor. the console is set to graphics
> now.
>
> diagnostic testing from SRM passes, including memory.
>
> i don't really have a use for this computer and will sell it, but before
> i do i would like to try and resolve these issues. from the user manual
> i see 2 possible options:
>
> 1- use the jumper pins on J29 for:
> a)normal recovery of the ROM   pins 1/2
> b)restore factory defaults   pins 15/16
>
> 2- reflash from SRM using the ldr command (? correct command)
>
> if i do either of these will i automatically default to *console
> serial*? i do not have a serial cable here and would have to obtain one.
>
> any other suggestions?
>
> thanks
> jeff duncan

If the problem persists with a working power supply, then we might
consider that the CPUs or motherboard are somehow faulty. The CPUs
were run with a higher (read: wrong) voltage for a non-trivial amount
of time after all.

Matt Turner