weird wild Octane error at bootup

SGI hardware problems, solutions, tips, hacks, etc.
Forum rules
Any posts concerning pirated software or offering to buy/sell/trade commercial software are subject to removal.
User avatar
OSWebman
Posts: 42
Joined: Wed Mar 26, 2003 5:20 pm
Location: Brooklyn, NY
Contact:

weird wild Octane error at bootup

Unread postby OSWebman » Thu Apr 01, 2004 8:44 pm

Recently, I've had to remount my gfx boards every day or two because they just wouldn't work if I didn't. About a week ago I started getting a problem that I can't solve. I tried multiple systems, cleaned the compressors and the crossbar and under the heatsinks, but no luck. When I turn the system on I get the solid red lightbar of graphics doom. This even though i have a green LED for QA and moving my l3w3 adaptor around changes the colors on the screen. I know its not the adaptor and I know it's not the monitor because I've used em on this system for ages and I just tested the monitor with a linux box. What could be causing this problem on the system I've used for so long? My local radioshack doesnt have any serial cables, and the monitor displays a solid blue color if the adaptor isnt touched.
Any help is appreciated
Thank you all in advance,
Eugene Redko

User avatar
nvukovlj
Posts: 356
Joined: Mon Jun 09, 2003 8:27 am
Location: London, UK
Contact:

Unread postby nvukovlj » Fri Apr 02, 2004 1:44 am

What gfx card do you have ?

Nik.

User avatar
OSWebman
Posts: 42
Joined: Wed Mar 26, 2003 5:20 pm
Location: Brooklyn, NY
Contact:

Unread postby OSWebman » Fri Apr 02, 2004 6:36 pm

The main card is an SE but changing main to any of my SI's doesn't help.
hinv is here:
viewtopic.php?t=1977
I don't understand how a particular crossbar input (QA) can send power to, but not input a signal to a board. My connections are clean and this is the first time this problem has happened.
Thanks again,
Eugene Redko

User avatar
whiter
Posts: 975
Joined: Tue Apr 29, 2003 2:02 pm
Location: Melbourne, Australia
Contact:

Unread postby whiter » Sat Apr 03, 2004 6:23 am

Power is not given through the XIO compression connector but through a seperate channel pretty close to it. Looks pretty much the same as the power connectors next to the GIO64 busses in an Indigo2. So it might very well be that the XIO compression connector is broken but the power (and some other signals too) connector is not.
Shall I describe it to you? Or do you want me to get you a box?

User avatar
OSWebman
Posts: 42
Joined: Wed Mar 26, 2003 5:20 pm
Location: Brooklyn, NY
Contact:

Unread postby OSWebman » Sun Apr 04, 2004 6:42 am

So your saying the crossbar's female XIO compression connector is damaged or broken? Any suggestions short of replacing the crossbar? the connecter is very clean with no damage within it, and the other side of the crossbar with the capacitators and all seems perfect.
Thanks again,
Eugene Redko

User avatar
The Keeper
Posts: 1632
Joined: Mon Nov 17, 2003 9:23 pm
Location: Buffalo, NY
Contact:

Unread postby The Keeper » Sun Apr 04, 2004 10:11 am

OSWebman wrote:So your saying the crossbar's female XIO compression connector is damaged or broken? Any suggestions short of replacing the crossbar? the connecter is very clean with no damage within it, and the other side of the crossbar with the capacitators and all seems perfect.
Thanks again,
Eugene Redko


If you're up to a drive to Buffalo, I can give you a chassis with a 1.3XBow frontplane, to replace the one that's obviously not really happy anymore... I have a few spares, but they're really not worth the cost of the packing materials and the effort to pack...


Chris

User avatar
OSWebman
Posts: 42
Joined: Wed Mar 26, 2003 5:20 pm
Location: Brooklyn, NY
Contact:

Unread postby OSWebman » Mon Apr 05, 2004 8:58 am

In a change for the even weirder..... No matter how hard I try.... I can't get any serial console output from port 1. Nothing, not Mincom nor hyperterminal do the trick. This can only be explained by unproper motherboard pin to crossbar port, right?
The green led's show:

x
x o
o x
o x

Where x signifies the particular part is recieveing a marked usable load of power. Which means a down serial connection doesnt make any sense at all. If anyone has any suggestions on how to fix this, please suggest them.
Thanks,
Eugene Redko

User avatar
OSWebman
Posts: 42
Joined: Wed Mar 26, 2003 5:20 pm
Location: Brooklyn, NY
Contact:

Unread postby OSWebman » Mon Apr 05, 2004 9:14 am

I also just noticed that setting the pin option of Alt. PROM on the motherboard to the alternative value puts the system in a state where it doesnt even boot to the point of displaying to the lightbar. Is this normal? Is there another way to get into this system other then DB9 to DB9 ? What is that port in the crossbar for? Sorry for all the questions, it's just when my SGI goes down it becomes a huge hobby for me to fix it.
thanks again,
Eugene Redko

User avatar
The Keeper
Posts: 1632
Joined: Mon Nov 17, 2003 9:23 pm
Location: Buffalo, NY
Contact:

Unread postby The Keeper » Mon Apr 05, 2004 11:17 am

OSWebman wrote:I also just noticed that setting the pin option of Alt. PROM on the motherboard to the alternative value puts the system in a state where it doesnt even boot to the point of displaying to the lightbar. Is this normal? Is there another way to get into this system other then DB9 to DB9 ? What is that port in the crossbar for? Sorry for all the questions, it's just when my SGI goes down it becomes a huge hobby for me to fix it.
thanks again,
Eugene Redko


It's normal for an Octane to sit there, braindead, for a few minutes if the NVRAM defaults jumper is installed. I'm wondering if it's a security measure on SGI's part, so you can't just go in and quickly reset the NVRAM...

Anyway, just give it time, and it will eventually come up (if the system is working in the first place, that is).

Chris

User avatar
nvukovlj
Posts: 356
Joined: Mon Jun 09, 2003 8:27 am
Location: London, UK
Contact:

Unread postby nvukovlj » Mon Apr 05, 2004 11:52 am

There is no Alternative PROM on the Octane MB. There's a space for it, but there isn't one there. I think it was only used during the design of the motherboard, as even the SGI engineers have no access to these proms.
I remember looking for it when the main prom on my motherboard got fried during an IRIX upgrade... Ended up buying another motherboard...

Anyway, this is a little OT, and not really helping with the main issue you're experiencing.

Nik.

User avatar
OSWebman
Posts: 42
Joined: Wed Mar 26, 2003 5:20 pm
Location: Brooklyn, NY
Contact:

Unread postby OSWebman » Fri Apr 16, 2004 10:32 pm

After a lot of tinkering the system boots and passes diagnostics, but all version of Irix that I have on hardisks give me Kernel Failures on mainboards that have been used for a while. Short of tossing the system and/or components how can I fix this:

PANIC: KERNEL FAULT PC
0xa8000000202e7d40 ep: 0xffffffffffffbc48 EXC
code:128 'software detected SEGV'
bad addr: 0x3e8 cause: 0x8<CE=0

The mainboard connectors are perfectly clean.
Please help out, any suggestions are useful...I feel like I'm so close to the finish line/login screen.
Thanks again,
Eugene Redko


Return to “SGI: Hardware”

Who is online

Users browsing this forum: No registered users and 1 guest