Fuel CPU Disabled

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
xiri
Donor
Donor
Posts: 54
Joined: Wed Dec 12, 2012 2:40 am
Location: Norway

Re: Fuel CPU Disabled

Unread postby xiri » Sun Jul 30, 2017 2:11 pm

L2 Emulator ? The L2-controller was a ppc-based hardware-appliance running Linux.
I assume you mean the L3-controller (System-controller) which was connected to all of the L2-controllers in the racks and
ran on Red Hat Linux on a SGI 230. 3 weeks ago I installed Red Hat Linux 9 (2003) and the L3-controller software in a vm,
had no difficulties connecting to my Fuel.
Attachments
l3.jpeg
l3.jpeg (57.68 KiB) Viewed 310 times
:Indigo: :Indigo2: :Indigo2IMP: :Indy: :O2: :O2+: :Octane: :Octane2: :Fuel: :O3x04R: :A3503L: :PrismDT: :Tezro:

mazzinia
Posts: 97
Joined: Thu Nov 18, 2010 1:14 pm
Location: Italy

Re: Fuel CPU Disabled

Unread postby mazzinia » Sun Jul 30, 2017 2:28 pm

yes, I meant that, sorry.

How big is the resulting vm ? ( vmware ? )
Some streamlined vm to distribute around would be interesting
:O2: :Fuel: HP c8000 Digital PWS 500au (x2) , IBM 9402-P03

kubatyszko
Donor
Donor
Posts: 456
Joined: Sat Nov 12, 2011 3:18 am
Location: Los Angeles
Contact:

Re: Fuel CPU Disabled

Unread postby kubatyszko » Sun Jul 30, 2017 2:46 pm

Uh oh, I am very curious about this now, I actually broke my Fuel's PROM (it froze when flashing), my understanding is that the L3 can reflash prom even if it's broken.
Do you have any information on what's required to have that L3 software running ?

@Dodoid - one more thing came to mind, Fuel that had PIMM of certain type installed (say 800MHz) will refuse to boot with PIMM of lower specs (less than 800Mhz in this case) - now, I *don't think* that it manifests itself with "cpu disabled" message but who knows...
Designer of Fuel ATX PSU adapter board : viewtopic.php?f=3&t=16732341
(now ships from USA), In stock!

User avatar
Dodoid
Posts: 603
Joined: Mon Jul 04, 2016 1:36 pm
Location: Ottawa, Canada
Contact:

Re: Fuel CPU Disabled

Unread postby Dodoid » Sun Jul 30, 2017 3:33 pm

There's another possibility I have been wondering about. The power supply I'm using with you Fuel to ATX adapter has in the past had problems with more power consuming graphics cards in PCs. env says the voltages are fine, but maybe that's not to be trusted if the adapter is being used, or maybe the PSU is having harder to detect problems. I could try to find and buy another PSU for it to test that, if it doesn't work I have a useful PSU sitting around.
:Onyx: :O2000: :Fuel: :Octane: :Octane: :Octane: :O2: :O2: :Indigo2: :Indigo2: :Indy: :Indy:
and a small army of Image

kubatyszko
Donor
Donor
Posts: 456
Joined: Sat Nov 12, 2011 3:18 am
Location: Los Angeles
Contact:

Re: Fuel CPU Disabled

Unread postby kubatyszko » Sun Jul 30, 2017 6:13 pm

Nah, that's unlikely, I've successfully run Fuel on a very crappy 400W PSU (it wasn't stable, voltages all over the place but it did run).
Designer of Fuel ATX PSU adapter board : viewtopic.php?f=3&t=16732341
(now ships from USA), In stock!

mazzinia
Posts: 97
Joined: Thu Nov 18, 2010 1:14 pm
Location: Italy

Re: Fuel CPU Disabled

Unread postby mazzinia » Sun Jul 30, 2017 6:18 pm

That L3 software can be downloaded by 2 of the links in these posts (other links are down)

viewtopic.php?f=3&t=16719038&start=30
:O2: :Fuel: HP c8000 Digital PWS 500au (x2) , IBM 9402-P03

mopar5150
Posts: 548
Joined: Tue Apr 24, 2012 6:02 pm
Location: Palm Springs, CA
Contact:

Re: Fuel CPU Disabled

Unread postby mopar5150 » Mon Jul 31, 2017 8:17 am

If you changed a graphics card you may have accidently pushed the bottom metal shield into the motherboard pulling down one of the supplies.

I have also seen the small copper colored EFI springs get knocked off and fall onto the mother board. Since this worked before the graphics card swap I would look there first.
If the thing isn't on fire it's a software problem.

:Tezro: :O3x0: :A350:

mazzinia
Posts: 97
Joined: Thu Nov 18, 2010 1:14 pm
Location: Italy

Re: Fuel CPU Disabled

Unread postby mazzinia » Mon Jul 31, 2017 8:32 am

Forgot
The iso of the 2nd cd of redhat 9 is faulty in all the mirrors I checked.

This link has the dvd version. I would recommend to install the kernel sources + the 2 dev libraries while installing, otherwise just find the 3 rpm and add them from the command line ( the package manager, after the installation, tries to ask for a cd2, using that dvd, and fails )
http://www.fis.unipr.it/pub/linux/redha ... /iso/i386/

Another requisite is KSH

Get it from here ( for that old version of redhat )
http://rpm.pbone.net/index.php3/stat/4/ ... 6.rpm.html

The result is around 3.7gb , but I kept the workstation default selections.
:O2: :Fuel: HP c8000 Digital PWS 500au (x2) , IBM 9402-P03

User avatar
xiri
Donor
Donor
Posts: 54
Joined: Wed Dec 12, 2012 2:40 am
Location: Norway

Re: Fuel CPU Disabled

Unread postby xiri » Mon Jul 31, 2017 9:09 am

mazzinia wrote:
Another requisite is KSH

That is right, during the installation of the L3-controller software, the routine asks for either ksh or pdksh to be installed.
After installing a suitable version of ksh the installation failed with less errors than before, but was still asking for either ksh or pdksh.
After I had installed pdksh as well, the installation of the L3 went flawlessly.
The L3 software definitely installs (and works) on 2.6 kernels as well, some years ago I installed it on Fedora 14.
:Indigo: :Indigo2: :Indigo2IMP: :Indy: :O2: :O2+: :Octane: :Octane2: :Fuel: :O3x04R: :A3503L: :PrismDT: :Tezro:

User avatar
Dodoid
Posts: 603
Joined: Mon Jul 04, 2016 1:36 pm
Location: Ottawa, Canada
Contact:

Re: Fuel CPU Disabled

Unread postby Dodoid » Mon Jul 31, 2017 12:40 pm

I don't think the fuel did worked before the graphics card swap, as apparently they cannot run without a graphics card and I did not have one. I was never able to access the PROM from the first serial port, only the L1, so nothing seems to have changed in that respect.
:Onyx: :O2000: :Fuel: :Octane: :Octane: :Octane: :O2: :O2: :Indigo2: :Indigo2: :Indy: :Indy:
and a small army of Image

User avatar
johnnym
Donor
Donor
Posts: 176
Joined: Sun Sep 04, 2016 9:53 pm

Re: Fuel CPU Disabled

Unread postby johnnym » Tue Aug 01, 2017 3:39 am

Dodoid wrote:I don't think the fuel did worked before the graphics card swap, as apparently they cannot run without a graphics card and I did not have one.

Is that a fact (e.g. from SGI documentation)? Why should a Fuel not run without a graphics card?
:Indy: :O2: :Octane: :Octane2: :O200: = :O200: - :O200: = :O200: (O200 cluster w/2 GIGAchannel cabinets)
[ ( hp ) ] 712/80 c3000 (dead) :hpserv: (J5600) c3700 c3750 c8000 rp2470 :rx2600: (rx2620) rx4640
| d | i | g | i | t | a | l | AXPpci33 AlphaStation 200 AlphaStation 255 PWS 500au AlphaServer DS20E AlphaServer DS25
C O B A L T Qube 2 Qube 3 RaQ RaQ 2 RaQ 4r RaQ XTR

User avatar
miod
Posts: 508
Joined: Fri Oct 09, 2009 2:44 am
Location: Clermont-Ferrand (France)
Contact:

Re: Fuel CPU Disabled

Unread postby miod » Wed Aug 02, 2017 11:53 pm

johnnym wrote:
Dodoid wrote:I don't think the fuel did worked before the graphics card swap, as apparently they cannot run without a graphics card and I did not have one.

Is that a fact (e.g. from SGI documentation)? Why should a Fuel not run without a graphics card?

It is not documented, but experimentally verified:
http://forums.nekochan.net/viewtopic.php?f=5&t=16729657&p=7380380#p7380380
:Indigo:R3000 (alas, dead) :Indigo:R4000 x4 :Indigo2:R4400 :Indigo2IMP:R4400 x2 :Indigo2:R8000 :Indigo2IMP:R10000 :Indy:R4000PC :Indy:R4000SC :Indy:R4400SC :Indy:R4600 :Indy:R5000SC :O2:R5000 x3 :O2:RM7000 :Octane:2xR10000 :Octane:R12000 :O200:2xR12000 :O200: - :O200:2x2xR10000 :Fuel:R16000 :O3x0:4xR16000 :A350:
among more than 150 machines : Apollo, Data General, Digital, HP, IBM, MIPS before SGI, Motorola, NeXT, SGI, Solbourne, Sun...

User avatar
johnnym
Donor
Donor
Posts: 176
Joined: Sun Sep 04, 2016 9:53 pm

Re: Fuel CPU Disabled

Unread postby johnnym » Thu Aug 03, 2017 1:06 am

miod wrote:
johnnym wrote:
Dodoid wrote:I don't think the fuel did worked before the graphics card swap, as apparently they cannot run without a graphics card and I did not have one.

Is that a fact (e.g. from SGI documentation)? Why should a Fuel not run without a graphics card?

It is not documented, but experimentally verified:
http://forums.nekochan.net/viewtopic.php?f=5&t=16729657&p=7380380#p7380380

OK, now that's unfortunate. Is that true for all Origin 3x0 related SGI machines then, too? Older SGI gear seems to not have this "problem".

UPDATE:
Wait a moment, in Dodoid's video about his Fuel he didn't have a graphics card installed originally (as his Fuel came without one) and it could be powered up and down from the L1 console. But his Fuel also has an older firmware image installed ("Revision 1.18.8, Built 02/07/2003 [...]" instead of "1.38.9 (Image B), Built 07/26/2005" on yours). Is this the L1 firmware or the machine's firmware?

As it's unknown what parts are working OK in Dodoid's Fuel, my recommendation was to remove everything that's not needed for operation from the machine before further testing (i.e. going down to two memory modules and check with all available memory modules if the behaviour changes and also removing the graphics card again (assuming the Fuel would work without it) and then even try with the NVRAM removed (one of my Origin 200s came up without NVRAM and tried to "reset" its configuration then). I assume the system board might be OK, as L1 and environmental readings seem to work OK. But the status of the remainder of the hardware is unclear.

viewtopic.php?f=3&t=16728549&p=7368328&hilit=fuel+fumes#p7368337 "suggests" that the PROM might come up even without memory modules installed and then complain about the missing memory modules. Even if it cannot continue later, it would have started at least. Can someone with a Fuel confirm this? I'll check that with my Origin 200 and update this post later on.

EDIT:
Indeed, my Origin 200 came up starting the PROM without memory modules installed:

Code: Select all

IP27 PROM SGI Version 6.156  built 11:27:56 AM Nov 18, 2003
*** No memory found.
*** Warning: MSC debug (dbg) switches are non-zero
*** Diag level set to None (2)
using BaseIO nic
Testing/Initializing memory ...............             DONE
*** No useable RAM installed. Need working and enabled memory in bank 0 or 1
*** Add working and enabled memory present in bank 0 or 1 and reset the system


Of course this does not say something about the Fuel, but it might behave similarly.

If this would work, Dodoid could even rule out the memory modules as source of error and try without them.

A possible next step in diagnosing Dodoid's machine might be to compare the L1 log messages of Dodoids Fuel with the L1 log messages of a confirmed working Fuel. Dodoid, can you record the console output somehow? Because it's hard to follow it in your video.
:Indy: :O2: :Octane: :Octane2: :O200: = :O200: - :O200: = :O200: (O200 cluster w/2 GIGAchannel cabinets)
[ ( hp ) ] 712/80 c3000 (dead) :hpserv: (J5600) c3700 c3750 c8000 rp2470 :rx2600: (rx2620) rx4640
| d | i | g | i | t | a | l | AXPpci33 AlphaStation 200 AlphaStation 255 PWS 500au AlphaServer DS20E AlphaServer DS25
C O B A L T Qube 2 Qube 3 RaQ RaQ 2 RaQ 4r RaQ XTR

mazzinia
Posts: 97
Joined: Thu Nov 18, 2010 1:14 pm
Location: Italy

Re: Fuel CPU Disabled

Unread postby mazzinia » Thu Aug 03, 2017 4:38 am

miod wrote:
johnnym wrote:
Dodoid wrote:I don't think the fuel did worked before the graphics card swap, as apparently they cannot run without a graphics card and I did not have one.

Is that a fact (e.g. from SGI documentation)? Why should a Fuel not run without a graphics card?

It is not documented, but experimentally verified:
http://forums.nekochan.net/viewtopic.php?f=5&t=16729657&p=7380380#p7380380


but if I read correctly the referenced thread, they tried with the env monitoring enabled.
One of the i2c chips that L1 uses for env monitoring is on the V10/V12 board itself...

Or was it tried also with the env monitoring disabled ?
:O2: :Fuel: HP c8000 Digital PWS 500au (x2) , IBM 9402-P03

User avatar
miod
Posts: 508
Joined: Fri Oct 09, 2009 2:44 am
Location: Clermont-Ferrand (France)
Contact:

Re: Fuel CPU Disabled

Unread postby miod » Thu Aug 03, 2017 1:00 pm

mazzinia wrote:but if I read correctly the referenced thread, they tried with the env monitoring enabled.
One of the i2c chips that L1 uses for env monitoring is on the V10/V12 board itself...

Or was it tried also with the env monitoring disabled ?

Ah, that would explain the i2c complains.

I don't remember if this particular Fuel has environment enabled. I'll check this and will try to run it without the graphics board again if it was enabled.
And of course, the ability to power on without the VPro might depend on the L1 firmware revision.
:Indigo:R3000 (alas, dead) :Indigo:R4000 x4 :Indigo2:R4400 :Indigo2IMP:R4400 x2 :Indigo2:R8000 :Indigo2IMP:R10000 :Indy:R4000PC :Indy:R4000SC :Indy:R4400SC :Indy:R4600 :Indy:R5000SC :O2:R5000 x3 :O2:RM7000 :Octane:2xR10000 :Octane:R12000 :O200:2xR12000 :O200: - :O200:2x2xR10000 :Fuel:R16000 :O3x0:4xR16000 :A350:
among more than 150 machines : Apollo, Data General, Digital, HP, IBM, MIPS before SGI, Motorola, NeXT, SGI, Solbourne, Sun...


Return to “SGI: Hardware”

Who is online

Users browsing this forum: Bing [Bot] and 2 guests