10 Apr
2012
10 Apr
'12
9:47 a.m.
Greetings;
Long-story-short, a couple of years ago I picked up six Onyx2 racks and
have been moving them around with me without ever actually firing them up.
I've finally got myself sorted and have been slowly working through
bringing things up and having some successes, but every step closer has me
finding a new problem.
My set-up right now has one graphics head and five compute nodes cabled
together in a daisy-chain (not enough CrayLinks for anything else). It
appears all but one of the MMSCs are shot, so I'm doing manual start-ups
using the keys.
My current confusion is how to nominate which system becomes the Global
Master. For some odd reason whenever I bring up three racks the machine
I've "picked" as the master (keyboard/mouse/gfx head) comes up just fine
and boots into IRIX, but whenever I add two more nodes things get a bit
more fuzzy and the Global Master appears to migrate around.
I had initially believed that the last rack in the power-up sequence would
always become the Global Master, since it goes and finds all the rest, but
this apparently is not the case... or perhaps there are corollaries I'm
unaware of.
The more times I turn this thing on and off the more hardware is failing
on me, not unexpectedly. I've lost a PSU, a node board and now one of the
racks has started making a worryingly hot-electrical smell. I'd really
like to get it all working together just once before I get old and grey.
Cheers;
- JP
10 Apr
10 Apr
10:28 a.m.
Just a few pointers.
Check out http://forums.nekochan.net. The site is down at the moment but
usualy quite alive and has more than a few members with Onyx2 racks
(myself included).
Check out sgi documentation (I'm sure you have seen it):
http://techpubs.sgi.com/library/dynaweb_docs/hdwr/SGI_Admin/books/Onyx2_Rac…
When it comes to spares.. well, the PSU seems to go and I'd like to keep
my spares for when I bring my own rack up. I could probably separate
with a few node boards (uncertain what I actually have, will have to
check).
The MMSC fails in the display if I'm not misstaken, perhaps a
replacement LCD could be found? But you should be able to fire up the
whole system with just one.
And finally, I love your youtube channel! keep the videos comming :)
Cheers,
Pontus.
On Tue, Apr 10, 2012 at 08:47:33AM -0500, JP Hindin wrote:
Greetings;
Long-story-short, a couple of years ago I picked up six Onyx2 racks and
have been moving them around with me without ever actually firing them up.
I've finally got myself sorted and have been slowly working through
bringing things up and having some successes, but every step closer has me
finding a new problem.
My set-up right now has one graphics head and five compute nodes cabled
together in a daisy-chain (not enough CrayLinks for anything else). It
appears all but one of the MMSCs are shot, so I'm doing manual start-ups
using the keys.
My current confusion is how to nominate which system becomes the Global
Master. For some odd reason whenever I bring up three racks the machine
I've "picked" as the master (keyboard/mouse/gfx head) comes up just fine
and boots into IRIX, but whenever I add two more nodes things get a bit
more fuzzy and the Global Master appears to migrate around.
I had initially believed that the last rack in the power-up sequence would
always become the Global Master, since it goes and finds all the rest, but
this apparently is not the case... or perhaps there are corollaries I'm
unaware of.
The more times I turn this thing on and off the more hardware is failing
on me, not unexpectedly. I've lost a PSU, a node board and now one of the
racks has started making a worryingly hot-electrical smell. I'd really
like to get it all working together just once before I get old and grey.
Cheers;
- JP
10:50 a.m.
On Tue, 10 Apr 2012, Pontus Pihlgren wrote:
Just a few pointers.
Check out http://forums.nekochan.net. The site is down at the moment but
usualy quite alive and has more than a few members with Onyx2 racks
(myself included).
The NekoChan guys are great - but, as you say, dead in the water recently
or I would've posted there first. I'm being impatient :)
Check out sgi documentation (I'm sure you have
seen it):
http://techpubs.sgi.com/library/dynaweb_docs/hdwr/SGI_Admin/books/Onyx2_Rac…
The documentation seems to be pretty sparse on these sorts of specifics -
I've been vaguely wondering if an SGI tech generally taught owners this
sort of thing.
When it comes to spares.. well, the PSU seems to go
and I'd like to keep
my spares for when I bring my own rack up. I could probably separate
with a few node boards (uncertain what I actually have, will have to
check).
Believe me, I hear you. And the PSUs are unbelievably heavy, so shipping
would kill. And complicated. I was delighted when I popped it out and
refreshed my memory on its output though - 3.45v at 375 Amps. 375!
The MMSC fails in the display if I'm not
misstaken, perhaps a
replacement LCD could be found? But you should be able to fire up the
whole system with just one.
My understanding is the MMSCs are far more prone to fail than the screens.
The MMSCs will bring up the NIC lights on start, so I've been using that
as an indicator and only one seems to work.
More importantly, unfortunately, every rack needs an MMSC to fire. The
MMSC connects to the MSC to co-ordinate the power sequencing of the
modules and each MMSC can only talk to two modules at once.
And finally, I love your youtube channel! keep the
videos comming :)
Thank you!
I'm trying to put up something w/all these damned SGIs for a joke but it
keeps getting further delayed because of all the freakin' problems I'm
having. I'll get there in the end. Three racks gave me 24 processors the
other day, which blew me away. But I'd sure like to go for 40 (I've got a
couple spare nodeboards).
- JP
On Tue, Apr 10, 2012 at 08:47:33AM -0500, JP Hindin
wrote:
Greetings;
Long-story-short, a couple of years ago I picked up six Onyx2 racks and
have been moving them around with me without ever actually firing them up.
I've finally got myself sorted and have been slowly working through
bringing things up and having some successes, but every step closer has me
finding a new problem.
My set-up right now has one graphics head and five compute nodes cabled
together in a daisy-chain (not enough CrayLinks for anything else). It
appears all but one of the MMSCs are shot, so I'm doing manual start-ups
using the keys.
My current confusion is how to nominate which system becomes the Global
Master. For some odd reason whenever I bring up three racks the machine
I've "picked" as the master (keyboard/mouse/gfx head) comes up just fine
and boots into IRIX, but whenever I add two more nodes things get a bit
more fuzzy and the Global Master appears to migrate around.
I had initially believed that the last rack in the power-up sequence would
always become the Global Master, since it goes and finds all the rest, but
this apparently is not the case... or perhaps there are corollaries I'm
unaware of.
The more times I turn this thing on and off the more hardware is failing
on me, not unexpectedly. I've lost a PSU, a node board and now one of the
racks has started making a worryingly hot-electrical smell. I'd really
like to get it all working together just once before I get old and grey.
Cheers;
- JP
11:39 a.m.
I was delighted when I popped [a power supply] out and
refreshed my
memory on its output though - 3.45v at 375 Amps. 375!
Er. Wow. That's .0092 ohm of load.
Talk about energy hazard....
/~\ The ASCII Mouse
\ / Ribbon Campaign
X Against HTML mouse at rodents-montreal.org
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B
12:32 p.m.
In article <Pine.LNX.4.44.1204100945160.17750-100000 at localhost>,
JP Hindin <jplist2008 at kiwigeek.com> writes:
More importantly, unfortunately, every rack needs an
MMSC to fire.
Every rack with nodes, yes?
I have a 4-rack Onyx2 Reality Monster and the only rack with an MMSC
is the rack with the compute nodes. The other 3 racks with graphics
don't have an MMSC.
I'm trying to put up something w/all these damned
SGIs for a joke but it
keeps getting further delayed because of all the freakin' problems I'm
having. I'll get there in the end. Three racks gave me 24 processors the
other day, which blew me away. But I'd sure like to go for 40 (I've got a
couple spare nodeboards).
Please keep us posted as to your progress. I have never had
sufficient power to fire up my Reality Monster and I suspect that the
longer it sits gathering dust, the more little problems I'm
accumulating...
--
"The Direct3D Graphics Pipeline" -- DirectX 9 version available for download
<http://legalizeadulthood.wordpress.com/the-direct3d-graphics-pipeline/>
Legalize Adulthood! <http://legalizeadulthood.wordpress.com>
12:40 p.m.
On Apr 10, 2012, at 12:32 PM, Richard wrote:
Please keep us posted as to your progress. I have
never had
sufficient power to fire up my Reality Monster and I suspect that the
longer it sits gathering dust, the more little problems I'm
accumulating...
Especially with the caps (as the PDP-10 conservationists here have
already expounded upon at length). They're the most likely
components to "rot" in storage after mechanical media.
- Dave
1:13 p.m.
In article <200DD58A-8C8C-4EDE-B342-BFA89AFF3837 at gmail.com>,
David Riley <fraveydank at gmail.com> writes:
On Apr 10, 2012, at 12:32 PM, Richard wrote:
Yeah, I'm guessing that's what's taking a toll on the PSUs? They've
got big caps in them, I'm sure.
My previous warehouse had *no* power outlets in the portion I rented.
The current warehouse has power outlets, but the building is woefully
underpowered. I feel like I'd be on the edge of burning a
fuse/popping a breaker to plug in anything that needs a high amperage
plug. So it's individual workstations and terminals only, at least
for a while. I never expected to be able to power on my big SGI
machines until I had some sort of machine room type environment with
controlled cooling and beefy power.
--
"The Direct3D Graphics Pipeline" -- DirectX 9 version available for download
<http://legalizeadulthood.wordpress.com/the-direct3d-graphics-pipeline/>
Legalize Adulthood! <http://legalizeadulthood.wordpress.com>
Please keep us posted as to your progress. I
have never had
sufficient power to fire up my Reality Monster and I suspect that the
longer it sits gathering dust, the more little problems I'm
accumulating...
Especially with the caps (as the PDP-10 conservationists here have
already expounded upon at length). They're the most likely
components to "rot" in storage after mechanical media.
1:04 p.m.
On Tue, 10 Apr 2012, Richard wrote:
Willdo. Most of the effort was involved in running a new 6-3 line, new
outlets/plugs (mismatching 20A and 30A twistloks combined with
horrifically expensive outlets), and building my own power-strip to allow
me to run so many racks side-by-side.
Now I just need to work out how to say "YOU ARE THE MASTER" and get the
idiotic thing all up with all five racks.
- JP
In article
<Pine.LNX.4.44.1204100945160.17750-100000 at localhost>,
JP Hindin <jplist2008 at kiwigeek.com> writes:
Half. The Gfx modules definitely don't need the sequencing, beyond
avoiding massive in-rush current popping breakers. When the Gfx modules
come up they display a basic screen and then, when the KTOWNs they're
connected to are powered on, the Gfx module reinitialises anyway. But
flipping on six Gfx modules by hand, and then letting the MMSC handle the
compute nodes is a bit odd, to my mind.
But the Gfx modules (or, rather, the MSCs they contain) are capable of
connecting to the MMSCs. In my setup, each rack is mixed Gfx
module/Compute module, and both upper/lower bays was connected to the
MMSC. I don't know how many racks they usually had connected together,
however.
More importantly, unfortunately, every rack needs
an MMSC to fire.
Every rack with nodes, yes?
I have a 4-rack Onyx2 Reality Monster and the only rack with an MMSC
is the rack with the compute nodes. The other 3 racks with graphics
don't have an MMSC. I'm trying
to put up something w/all these damned SGIs for a joke but it
keeps getting further delayed because of all the freakin' problems I'm
having. I'll get there in the end. Three racks gave me 24 processors the
other day, which blew me away. But I'd sure like to go for 40 (I've got a
couple spare nodeboards).
Please keep us posted as to your progress. I have never had
sufficient power to fire up my Reality Monster and I suspect that the
longer it sits gathering dust, the more little problems I'm
accumulating...
3:42 p.m.
On 04/10/2012 04:50 PM, JP Hindin wrote:
I'm mixing things up here. I think the MMSC PSU tends to fail, have you
tried to swap them around?
http://forums.nekochan.net/viewtopic.php?f=5&t=16725089
Regards,
Pontus.
The MMSC fails in the display if I'm not
misstaken, perhaps a
replacement LCD could be found? But you should be able to fire up the
whole system with just one.
My understanding is the MMSCs are far more prone to
fail than the screens.
The MMSCs will bring up the NIC lights on start, so I've been using that
as an indicator and only one seems to work.
More importantly, unfortunately, every rack needs an MMSC to fire. The
MMSC connects to the MSC to co-ordinate the power sequencing of the
modules and each MMSC can only talk to two modules at once. 
4:46 p.m.
On 10/04/2012 20:42, Pontus wrote:
I'm mixing things up here. I think the MMSC PSU
tends to fail, have you
tried to swap them around?
That's quite plausible, and the good news is that those SMPSUs are an
industry-standard type that's fairly inexpensive and easy to replace.
Not that hard to repair, either, if you're so inclined.
--
Pete Peter Turnbull
Network Manager
University of York
4:10 p.m.
On 04/10/2012 09:50 AM, JP Hindin wrote:
The documentation seems to be pretty sparse on these
sorts of specifics -
I've been vaguely wondering if an SGI tech generally taught owners this
sort of thing.
That was certainly my experience in the latter half of the 90s. SGI liked
to sell support deals - we had a couple of tame (and probably very
expensive) SGI techs on-call who would either help via phone/net or come in
for site visits, and they were always great for imparting information that
wasn't in the released manuals.
There were quite a few times when they couldn't find what they needed in
whatever documentation they had back at base, either, and had to bounce
things back to someone deep within the bowels of SGI who could replicate
the setup as close as possible and work out what was going on.
cheers
Jules
12:57 p.m.
On Tuesday 10 April 2012, Pontus Pihlgren wrote:
Agreed...
Just a few pointers.
Check out http://forums.nekochan.net. The site is down at the moment but
usualy quite alive and has more than a few members with Onyx2 racks
(myself included).
Nekochan is very good.
I have brought up SGIs including Elans, Maximum Impacts, Octanes, etc. up to Onyx3s. I
have found that the following site(s) by Ian Mapleson are great sources of SGI info:
England:
http://www.sgidepot.co.uk/sgi.html
U.S. Mirror
http://vintagecomputers.info/sgi.html
If you get stuck, you can email Ian and as you can imagine, he knows SGI. He also has
spare parts (if you don't mind paying shipping costs from the UK).
Check out sgi documentation (I'm sure you have
seen it):
http://techpubs.sgi.com/library/dynaweb_docs/hdwr/SGI_Admin/books/Onyx2_Rac…
Agreed, techpubs.sgi.com is still a great source. I'd download all you can from there,
as the "new" SGI doesn't do MIPS and has been slowly deleting MIPS related
information from there websites and Supportfolio.
Speaking of which, you might want to sign up with Supportfolio:
https://support.sgi.com/
IIRC, you can still sign up free - and get certain IRIX patches and other software which
is downloadable free.
When it comes to spares.. well, the PSU seems to go
and I'd like to keep
my spares for when I bring my own rack up. I could probably separate
with a few node boards (uncertain what I actually have, will have to
check).
Indeed, most of us who collect SGIs have a good number of spares we've collected over
time - as they have been getting more and more difficult to find. Scrappers love MIPS
based SGIs because their backplanes and boards have a high level of gold content. Sigh...
The MMSC fails
in the display if I'm not misstaken, perhaps a
replacement LCD could be found?
But you should be able to fire up the
whole system with just one.
And finally, I love your youtube channel! keep the videos comming :)
Cheers,
Lyle
On Tue, Apr 10, 2012 at 08:47:33AM -0500, JP Hindin
wrote:
--
Lyle Bickley, AF6WS
Bickley Consulting West Inc.
http://bickleywest.com
"Black holes are where God is dividing by zero"
Greetings;
Long-story-short, a couple of years ago I picked up six Onyx2 racks and
have been moving them around with me without ever actually firing them up.
I've finally got myself sorted and have been slowly working through
bringing things up and having some successes, but every step closer has me
finding a new problem.
My set-up right now has one graphics head and five compute nodes cabled
together in a daisy-chain (not enough CrayLinks for anything else). It
appears all but one of the MMSCs are shot, so I'm doing manual start-ups
using the keys.
My current confusion is how to nominate which system becomes the Global
Master. For some odd reason whenever I bring up three racks the machine
I've "picked" as the master (keyboard/mouse/gfx head) comes up just fine
and boots into IRIX, but whenever I add two more nodes things get a bit
more fuzzy and the Global Master appears to migrate around.
I had initially believed that the last rack in the power-up sequence would
always become the Global Master, since it goes and finds all the rest, but
this apparently is not the case... or perhaps there are corollaries I'm
unaware of.
The more times I turn this thing on and off the more hardware is failing
on me, not unexpectedly. I've lost a PSU, a node board and now one of the
racks has started making a worryingly hot-electrical smell. I'd really
like to get it all working together just once before I get old and grey.
Cheers;
- JP
4464
days inactive
4464
days old
11 comments
8 participants
participants (8)
-
fraveydank＠gmail.com -
jplist2008＠kiwigeek.com -
jules.richardson99＠gmail.com -
lbickley＠bickleywest.com -
legalize＠xmission.com -
mouse＠Rodents-Montreal.ORG -
pete＠dunnington.plus.com -
pontus＠Update.UU.SE