[weblacky]

Yes to the above...with additional explanation.

None of the SGIs have rechargeable batteries, and it's exceedly rare in a PC anyway (today most of those uses are using super caps to keep time/setting between battery changes (for like 5-20 mins max).

Also, yeah, depending on the PSU design, leaving it plugged in wears the main HV side of the supply standby power to the SGI mainboard. Modern SGIs (Fuel, Tezro, and that same Origin Arch) ALL have soft-power on switches. They run their L1s ALL THE TIME (Tezro even runs some fans ALL THE TIME) and the board will of course use/produce (in different cases) internal voltages to run stuff (onboard small VRM-like sections to make different DC voltages). So it's often gobbling low-voltage DC 24/7. Remember SGIs were designed to be left on to work a lot to earn back their value in a couple years. That's one reason they don't have any sleep/hibernation option in Irix!

Just because the computer isn't on doesn't mean the PSU isn't doing something. Now some PSU designs have a totally separate standby transformers or primitive voltage dropping resistor/diode to run the PSU brain and standby power until the main transformer is operational. I'm unsure what these really do (each generation is different).

But always assume worst case scenario, the PSU is running (in some fashion) if the board is drawing power.

The total opposite (never using your SGI) can actually cause FASTER degradation if you keep it disconnecting about 2 years or more. Caps were designed to be used. So occasional usage does keep them in the best shape (though that's from a new/factory fresh start). You cannot start that after being in the closet for 7 years.

So there are both sides of the coin.

Also I won't say the risk of risk of fire is high for just leaving it plugged in for standby, but we have certainly seen explosions, smoke, and some fires here from people that suddenly took an aging SGI and asked it to run a triathlon...it went with a fireworks display.

PSU work is something we want to get to in the future, for now it's risky to run them like they are new. Newer SGI machines have a few board issues but with a new PSU, you still have tread left on the tires right now to drive them. In about 10 years, they'll also need massive cap replacements as well, as SGI used advanced polymer caps, which are ok for now, but whose lifespan is coming to a predicted end.

[jwhat]

Hi Stewart,

to your request of l1 power response on known good fuel:

Code:

# l1cmd power
Supply          State Voltage    Margin  Value
--------------  ----- ---------  ------- -----
           12V     on   12.063V      N/A
        12V IO     NC   12.063V      N/A
            5V     NC    5.044V      N/A
          3.3V     NC    3.320V  default     0
          2.5V     on    2.470V  default     0
          1.5V     NC    1.466V  default     0
        5V AUX     NC    5.096V      N/A
      3.3V AUX     NC    3.285V      N/A
 PIMM 12V BIAS     NC   12.063V      N/A
          SRAM     NC    2.509V  default     0
          VCPU     on    1.593V  default   128
     PIMM 1.5V     NC    1.495V  default     0
 PIMM 3.3V AUX     NC    3.268V      N/A
   PIMM 5V AUX     NC    5.070V      N/A
  XIO 12V BIAS     NC   12.000V      N/A
        XIO 5V     NC    5.018V      N/A
      XIO 2.5V     on    2.457V  default     0
  XIO 3.3V AUX     NC    3.285V      N/A
pink 2#

and leds:

Code:

pink 2# l1cmd leds
CPU  A: 0x03: PLED_RUNTLB:  Switch to mapped mode.
        0x02: PLED_TESTCP1:  Test processor COP1 registers.

and env

Code:

pink 5# l1cmd env
Environmental monitoring is enabled and running.

Description    State      Warning Limits    Fault Limits      Current
-------------- ----------  -----------------  -----------------  -------
          12V    Enabled  10%  10.80/ 13.20  20%  9.60/ 14.40  12.063
        12V IO    Enabled  10%  10.80/ 13.20  20%  9.60/ 14.40  12.063
            5V    Enabled  10%  4.50/  5.50  20%  4.00/  6.00    5.044
          3.3V    Enabled  10%  2.97/  3.63  20%  2.64/  3.96    3.320
          2.5V    Enabled  10%  2.25/  2.75  20%  2.00/  3.00    2.470
          1.5V    Enabled  10%  1.35/  1.65  20%  1.20/  1.80    1.466
        5V AUX    Enabled  10%  4.50/  5.50  20%  4.00/  6.00    5.096
      3.3V AUX    Enabled  10%  2.97/  3.63  20%  2.64/  3.96    3.285
PIMM 12V BIAS    Enabled  10%  10.80/ 13.20  20%  9.60/ 14.40  12.063
          SRAM    Enabled  10%  2.25/  2.75  20%  2.00/  3.00    2.509
          VCPU    Enabled  10%  1.44/  1.76  20%  1.28/  1.92    1.593
    PIMM 1.5V    Enabled  10%  1.35/  1.65  20%  1.20/  1.80    1.495
PIMM 3.3V AUX    Enabled  10%  2.97/  3.63  20%  2.64/  3.96    3.268
  PIMM 5V AUX    Enabled  10%  4.50/  5.50  20%  4.00/  6.00    5.070
  XIO 12V BIAS    Enabled  10%  10.80/ 13.20  20%  9.60/ 14.40  11.938
        XIO 5V    Enabled  10%  4.50/  5.50  20%  4.00/  6.00    5.018
      XIO 2.5V    Enabled  10%  2.25/  2.75  20%  2.00/  3.00    2.457
  XIO 3.3V AUX    Enabled  10%  2.97/  3.63  20%  2.64/  3.96    3.285

Description    State      Warning RPM  Current RPM
--------------- ----------  -----------  -----------
FAN  0  EXHAUST    Enabled          920        1214
FAN  1      HD    Enabled        1560        2393
FAN  2      PCI    Enabled        1120        1548
FAN  3    XIO 1    Enabled        1600        2360
FAN  4    XIO 2    Enabled        1600        2205
FAN  5      PS    Enabled        1349        42187

                              Advisory  Critical  Fault      Current     
Description      State      Temp      Temp      Temp      Temp     
----------------- ----------  ---------  ---------  ---------  --------- 
0 NODE 0            Enabled    [Autofan Control]    80C/176F  43C/109F
1 NODE 1            Enabled    [Autofan Control]    80C/176F  43C/109F
2 NODE 2            Enabled    [Autofan Control]    80C/176F  39C/102F
3 PIMM              Enabled    [Autofan Control]    80C/176F  54C/129F
4 ODYSSEY          Enabled    [Autofan Control]    80C/176F  43C/109F
5 BEDROCK          Enabled    [Autofan Control]    85C/185F  45C/113F

pink 6#

Let me know if you want to see any other L1 info or configuration details.

NOTE: this machine has original SGI power supply and is powered on ...

on "L1 debug" I have put Fuel into Debug: "debug 0x10d" which stops it doing diagnostics and going to PROM, and do reset of logs.

I have not tried "4".
What does debug 4 do ?

Cheers from Oz,

jwhat/John

[weblacky]

I just happened upon this thread while doing some other searching and noticed it had not been finally answered.

The preposterous time warning when you boot your machine indicates that you have a dead Snaphat battery. If you have a dead Snaphat battery you will never be able to boot from cold start. The way to test this is to simply plug in the system and wait for about a minute or two after auto-start (fans running) then hit the reset button on the front panel and the system should start normally. If you have this behavior it’s because at first cold boot the NVRAM data is invalid and becomes reset with defaults but when this happens the system immediately goes into power on diagnostics mode and does not proceed to boot normally. The reset allows it to boot normally now that it has new defaults because you didn’t lose power between the first boot and the reboot.

Jwhat, if you look at the prom man page at the bottom they disclose several debug parameters.

The manual claims you can OR these together to do multiple options. But I’ve had very bad luck doing it. I believe an OR is simply an addition but when I try to add let’s say verbose output with factory testing it always thinks it’s something else. Technically I believe verbose is also additional command so it’s not required. Basically normally you use the command 10 just to get it to stop in power on diagnostics mode. But you can also cause it to simply do more rigorous testing and it will stop naturally in power on diagnostics mode if there’s any significant errors. So you can either boot your system with more rigorous diagnostics normally, boot with no diagnostics normally, stop at POD without startup diags, or try combinations (which always seem to select testing =2 then whatever’s left).

So that’s how I’ve been using it.


1 No testing.
2 Heavy testing.
3 Manufacturing-level testing.

Diagnostic Output Level
4 Verbose. Information level is set to verbose.

Boot Stop Point
0 Normal. Normal setting, do not stop.
8 Global POD. Global master stops in POD mode; slaves enter slave
loop.
10 Local POD. Boot stop requested at local POD. All local masters and
CPUs with console access enter POD mode; the rest enter the slave
loop.
18 Memoryless POD. Boot stop requested at no memory POD. All CPUs
enter POD mode after memory is probed, but before it is tested or
initialized.

Default Environment
20 Ignores environment variable.

Override Disabling
100 Overrides CPUs and memory disabled with the environment variables in
POD mode. It is useful for getting out of the situation in which
all CPUs or memory in the system have accidentally been disabled
simultaneously.