[weblacky]

With my Fuel _with_ a V10 installed, when I look at env before booting, I see:

Code:

env
Environmental monitoring is enabled and running.

... (omitted power) ...

Description    State       Warning RPM  Current RPM
-------------- ----------  -----------  -----------
FAN 0  EXHAUST   Wait Pwr          920            0
FAN 1       HD   Wait Pwr         1560            0
FAN 2      PCI   Wait Pwr         1120            0
FAN 3    XIO 1   Wait Pwr         1600            0
FAN 4    XIO 2   Wait Pwr         1600            0
FAN 5       PS   Wait Pwr         1600            0

                           Advisory   Critical   Fault      Current
Description    State       Temp       Temp       Temp       Temp
-------------- ----------  ---------  ---------  ---------  ---------
NODE 0           Wait Pwr   60C/140F   65C/149F   70C/158F   21C/ 69F
NODE 1           Wait Pwr   60C/140F   65C/149F   70C/158F   20C/ 68F
NODE 2           Wait Pwr   60C/140F   65C/149F   70C/158F   21C/ 69F
PIMM             Wait Pwr   60C/140F   65C/149F   70C/158F   21C/ 69F
ODYSSEY          Wait Pwr   60C/140F   65C/149F   70C/158F   21C/ 69F
BEDROCK          Wait Pwr  Not currently available

Note the "Not currently available" on BEDROCK.  So that's normal.  My guess is the monitoring chip for BEDROCK just doesn't have power applied until the system is fully running.

I can pull my V10 temporarily if you'd like, but I'm not sure what it would tell you; the system would boot fine, like it did before.  I guarantee this system ran fine without a graphics board, and I can't image why SGI would want to limit the possibility to do that -- it would be a valuable option in case of graphics card damage when data needs to be recovered or similar.  There's no reason to explicitly forbid it.  I'm hesitant to upgrade my firmware because, well, things are working fine.

However I don't see what in your logs actually prevents things from booting; I see "ERROR: 001a01 auto power up error." but nothing beyond that.  I see from an earlier log something like "system failed pre-power check"; is that the current issue?  Are you seeing anything from the PROM after you start booting?  Issue a "pwr up" and then when the L1 prompt comes back, hit Control-D.

If that wasn't stated in my posting, let me see if I can describe it better.  To be 100% clear, since I cannot power up...I can't get verbose debug info...you have to "power up" for that :-)  So we cannot see anything because there's nothing to see (we're at too early a stage of "power on "to have any real info as a consumer).

When my card is inserted (after replacing the DS1780 and shorting out the 12V XIO short on the V10), I get dual white LED flashing on the front.  The entire ENV system claims, running, and ready - the bedrock temps DO APPEAR (they don't say unavailable anymore!)...the auto power-on proceeds to the next stage where my fans are running and RPM is correct (my fans don't run now with card removed...as I cannot go onto next power stage).

At this point I have an L1 that loves me, ENV filled, temps filled, fans RPMs filled & running, Ready to for pwr up....so I type pwr up.


Nothing happens...a few seconds later I get that "pre-power check failed" and when I say "pwr down", the L1 responds with "aren't I already powered down?" (or something like that, it's in the logs).  I cannot get to the stage where I can attempt power on without the V10 installed (as I showed), as then will NOT go to the next stage (fans and ready to power-on) and will complain of I2C missing.  So I HAVE to have a working graphics card to get to the stage I get to now...just ready for power up to PROM.

Now the original errors on my logs (posted earlier) where "Error reading monitor ODYSSEY interrupt status 1: no acknowledge" and sensor alerts.  Upon replacement of the V10 DS1780 I only got ALERT XIO 12V Bias low ~2.68v...which is how I found out I had a short on the card.  I accidently shorted the card out with a 1V DC 1.5A power injection on the short to try to find the short with my thermal imager on the V10 outside the system on my bench, so something went from dragging down the 12V XIO line and originally taking the DS1780 with it to now...but being open circuit...so it's disconnected inside the V10.


However I no longer get the Acknowledge error either...but I shorted something open on the 12V XIO rail on the V10...so that rail isn't fully connected (yet to figure it out).

My working theory at this point is the original owner had the system die by the V10 shorting out, also killing the DS1780 (the thing that monitors voltages for the XIO).  With the DS1780 now working the env is happy but alerting to the original short (now that the DS1780 is there to MONITOR the XIO 12V line :-)).

So that got shorted open by me while trying to find it (my bad)...so my assumption (that's the biggie) at this time is...the mechanism that shorted likely powers the 12V line of the card on/off or is otherwise included in signaling...that's disconnected inside the V10.

So the mainboard is ready for power-on...sends the power-on signal to the card...nothing happens...I believe (for lack of other evidence and damage outside the V10) this is why I get this pre-power on error. Because the V10 cannot be started because part of it's 12V internally have now become disconnected...here I sit.

Since I'm 100% certain the original failure was the V10 in this case (as I found real damage) I believe it's isolated to the V10.

You're suggesting that there is in fact more damage and it's the mainboard as well as the V10 (because I already have proof of V10 Damage)...well we both have zero way of knowing that until I try another graphics card because from what I can tell right now, when I complete the env i2C bus the L1 is happy and ready to go...the dead card is holding me back as the originally shorted 12V could have been a simple cap (not likely, I didn't get that lucky) more than likely it was a semiconductor and/or something important to the startup/operation signalling of the V10.

I do not see another way to break the stalemate other than by trying another card.