E250 temperatures

[Skeezics Boondoggle-2]

On Mon, 14 Apr 2008, "r.stricklin" wrote:

> On Apr 14, 2008, at 3:02 PM, Joshua D. Boyd wrote:
> > On Mon, 2008-04-14 at 14:44 -0700, Skeezics Boondoggle wrote:
> >> Whoa.  I just caught this.  You're running Darwin on an 8-way 486 box
> >> with 8GB of RAM?  What make and model?  That's very curious.
> >
> > I'm going to hazard a guess that it is a Mac Pro.
>
> It is, indeed, Leopard on a Mac Pro. Xeon 5400.

Ah.  Of course.  Mach reporting it as an 80486 threw me.  I thought
maybe you had compiled up the open sourced Darwin release on an old
Dell Poweredge or something. :-)

-- Skeez
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[r.stricklin]

On Apr 15, 2008, at 10:37 AM, Charles Monett wrote:

> Full diag boot? I'm unsure on that one, but my RS/6000's (7012-39H  
> and the 7011-220** being the main offenders, followed by the  
> 7044-270, and the 7044-170) would come close.

I used to run a Sun E10k system which took well over 45 minutes to  
POST, even at its least thorough diagnostics setting.

We had a run of problems with EMC Powerpath and VxVM around that  
time... I remember on several occasions being on the phone with  
Veritas support, and having them recommend a reboot as part of a  
troubleshooting procedure. They were always strangely resistant to  
finding an alternative procedure not involving a reboot, so we would  
say, "okay, we'll reboot, but you have to wait on the phone with us  
until it completes." The best only took two reboots to come around to  
our way of thinking. The worst took us through six reboots over about  
ten hours.

ok
bear
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Charles Monett]

Sridhar Ayengar wrote:
>
> Come on.  The 390H doesn't take long to boot.  It's somewhere between 5
> and 10 minutes.
>
> My ES/9000 takes an hour.  Maybe slightly more.
Consider me humbled by far on that - just that it is the longest of the
machines I have, and that it's in a relatively small/dense package. Not
long, but far off from the PC's of its time(in more than just time).
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Mike Meredith-3]

On Tue, 15 Apr 2008 19:01:52 -0700, r.stricklin wrote:
> On Apr 15, 2008, at 10:37 AM, Charles Monett wrote:
>
> > Full diag boot? I'm unsure on that one, but my RS/6000's (7012-39H  
> > and the 7011-220** being the main offenders, followed by the  
> > 7044-270, and the 7044-170) would come close.
>
> I used to run a Sun E10k system which took well over 45 minutes to  
> POST, even at its least thorough diagnostics setting.

Whilst repeatedly running an E3500 through it's diag boot, I happened
to be talking to a Sun engineer about how long systems take to boot. He
said he'd recently been looking at an E10k system which took 8 hours to
POST with diags turned on
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Robert Darlington]

I'd have to start asking why a system built in the last decade would
take any more than a few minutes to POST.  What could the system
possibly be doing that takes that kind of time?  I was extremely
annoyed at the 5 minutes my old dell server would take.  None of my
big iron systems (alpha, or sgi) ever took more than a few minutes to
come up.

On Wed, Apr 16, 2008 at 11:29 AM, Mike Meredith <very@...> wrote:

> On Tue, 15 Apr 2008 19:01:52 -0700, r.stricklin wrote:
>  > On Apr 15, 2008, at 10:37 AM, Charles Monett wrote:
>  >
>  > > Full diag boot? I'm unsure on that one, but my RS/6000's (7012-39H
>  > > and the 7011-220** being the main offenders, followed by the
>  > > 7044-270, and the 7044-170) would come close.
>  >
>  > I used to run a Sun E10k system which took well over 45 minutes to
>  > POST, even at its least thorough diagnostics setting.
>
>  Whilst repeatedly running an E3500 through it's diag boot, I happened
>  to be talking to a Sun engineer about how long systems take to boot. He
>  said he'd recently been looking at an E10k system which took 8 hours to
>  POST with diags turned on
>
>
> _______________________________________________
>  rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Mike Meredith-3]

On Wed, 16 Apr 2008 11:58:05 -0600, Robert Darlington wrote:
> I'd have to start asking why a system built in the last decade would
> take any more than a few minutes to POST.  What could the system
> possibly be doing that takes that kind of time?  I was extremely
> annoyed at the 5 minutes my old dell server would take.  None of my
> big iron systems (alpha, or sgi) ever took more than a few minutes to
> come up.

a) An E10k isn't that modern.

b) The key is 'diags turned on'. Your Dell doesn't have that level of
   diagnostics on board. POST times are slightly more reasonable with
   diags turned off.

c) These systems are a bit bigger than your Dell. A later example
   (M8000) can take 32 cores, and 512Gbytes of memory. A PC memory
   'test' would probably take more than 5 minutes to count that memory.

d) There's a difference between the time taken to do a proper power on
   test. I don't have figures to hand, but I'd guestimate that some of
   the E6900 domains I admin (16 cores, 32Gbytes memory) would take
   5 minutes to test during a reboot, and something like 30minutes
   during power on.

e) These systems aren't supposed to be powered off.

f) If these systems have a slightly wonky CPU board or some memory with
   a bit of a problem, you *really* want to know about it early. Sure
   most can 'degrade' a CPU board during operation but it's better to
   degrade it early.

I'm pretty sure the bigger SGI stuff takes a while to get going too ...
at least the O2K I once worked on did.
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Robert Darlington]

I ran the biggest sgi cluster on the planet for a while (6144 cpus).
It didn't take 30 minutes to boot an O2K with diags on.  I don't have
experience with the little E10k systems (little by my standards), but
even our big alpha cluster that took up a football field didn't take
that long to boot up -at least if you don't count the fact that we
brought it up in sections.   I not disputing that systems take that
long, I just can't see why.  What exactly are they doing when they
POST that takes hours or days?

-Bob

On Wed, Apr 16, 2008 at 12:41 PM, Mike Meredith <very@...> wrote:

> On Wed, 16 Apr 2008 11:58:05 -0600, Robert Darlington wrote:
>  > I'd have to start asking why a system built in the last decade would
>  > take any more than a few minutes to POST.  What could the system
>  > possibly be doing that takes that kind of time?  I was extremely
>  > annoyed at the 5 minutes my old dell server would take.  None of my
>  > big iron systems (alpha, or sgi) ever took more than a few minutes to
>  > come up.
>
>  a) An E10k isn't that modern.
>
>  b) The key is 'diags turned on'. Your Dell doesn't have that level of
>    diagnostics on board. POST times are slightly more reasonable with
>    diags turned off.
>
>  c) These systems are a bit bigger than your Dell. A later example
>    (M8000) can take 32 cores, and 512Gbytes of memory. A PC memory
>    'test' would probably take more than 5 minutes to count that memory.
>
>  d) There's a difference between the time taken to do a proper power on
>    test. I don't have figures to hand, but I'd guestimate that some of
>    the E6900 domains I admin (16 cores, 32Gbytes memory) would take
>    5 minutes to test during a reboot, and something like 30minutes
>    during power on.
>
>  e) These systems aren't supposed to be powered off.
>
>  f) If these systems have a slightly wonky CPU board or some memory with
>    a bit of a problem, you *really* want to know about it early. Sure
>    most can 'degrade' a CPU board during operation but it's better to
>    degrade it early.
>
>  I'm pretty sure the bigger SGI stuff takes a while to get going too ...
>  at least the O2K I once worked on did.
>
>
> _______________________________________________
>  rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Bill Bradford]

On Wed, Apr 16, 2008 at 07:41:03PM +0100, Mike Meredith wrote:
> a) An E10k isn't that modern.

GA of 3/97, last shipment date of 1/04.  They've only been EOLed for
four years and three months - maybe not "cutting edge" but I bet there are
still a few in service.

Bill

--
Bill Bradford
Houston, Texas
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Bill Green]

On Apr 16, 2008, at 1:58 PM, Robert Darlington wrote:
> I'd have to start asking why a system built in the last decade would
> take any more than a few minutes to POST.  What could the system
> possibly be doing that takes that kind of time?  I was extremely
> annoyed at the 5 minutes my old dell server would take.  None of my
> big iron systems (alpha, or sgi) ever took more than a few minutes to
> come up.
>

My Sun Blade 2000 (8 GB of RAM) takes around 15-20 minutes to POST
with diag-level = max.  On normal settings however it's less than a  
minute.
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Jonathan C. Patschke]

On Wed, 16 Apr 2008, Robert Darlington wrote:

> I'd have to start asking why a system built in the last decade would
> take any more than a few minutes to POST.  What could the system
> possibly be doing that takes that kind of time?

It's a full-system checkout, along the lines of what a field engineer
would run to re-certify a system.  This means different bit-patterns
through cache and real memories, bus checkouts, adapter checkouts, etc.

I have it on good authority that many largish IBM systems have more
code in their firmware to do diagnostics than there is total code (in
terms of instruction count) in the operating system kernel.

The idea is that if the system comes up, it's going to -stay- up because
it was all-but-recertified at last power-on.

--
Jonathan Patschke | "There is no such thing as a short of reserves...
Elgin, TX         |  one bank can have a problem...the Fed can print
USA               |  money, there is no shortage."
.                 |     --Jim Glassman, US Economist, JPMorgan Chase
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Shannon Hendrix]

On Apr 15, 2008, at 11:30 , Sridhar Ayengar wrote:

> Bill Bradford wrote:
>> On Mon, Apr 14, 2008 at 02:44:21PM -0700, Skeezics Boondoggle wrote:
>>> I'd love to fire that beastie up again and fill up the empty SIMM
>>> slots, maybe even add in the power supplies and 7 more system boards
>>> from the other machine... The I/O cabinet has 7 disk trays in it,  
>>> with
>>> ~ 3 drives on each of 14 of those SCSI buses.  And lots of fans.
>>> Enough upward airflow to put on some goggles and jumpsuit and hover
>>> over the machine.  Can't do THAT with a T1000. :-)
>> Yeah, I never feel like a machine is REALLY beefy unless it takes  
>> more than
>> 10-15 minutes to run through a full diag boot (like the big Alpha  
>> boxes at
>> $JOB-1).
>
> I have a machine or two that take 45+ minutes to POST.

I do too.

And one day, real soon now, I'm going to fix the broken script that  
causes it to hang like that... :)


--
Shannon Hendrix
shannon@...
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Sridhar Ayengar]

Shannon Hendrix wrote:

> On Apr 15, 2008, at 11:30 , Sridhar Ayengar wrote:
>
>> Bill Bradford wrote:
>>> On Mon, Apr 14, 2008 at 02:44:21PM -0700, Skeezics Boondoggle wrote:
>>>> I'd love to fire that beastie up again and fill up the empty SIMM
>>>> slots, maybe even add in the power supplies and 7 more system boards
>>>> from the other machine... The I/O cabinet has 7 disk trays in it, with
>>>> ~ 3 drives on each of 14 of those SCSI buses.  And lots of fans.
>>>> Enough upward airflow to put on some goggles and jumpsuit and hover
>>>> over the machine.  Can't do THAT with a T1000. :-)
>>> Yeah, I never feel like a machine is REALLY beefy unless it takes
>>> more than
>>> 10-15 minutes to run through a full diag boot (like the big Alpha
>>> boxes at
>>> $JOB-1).
>>
>> I have a machine or two that take 45+ minutes to POST.
>
> I do too.
>
> And one day, real soon now, I'm going to fix the broken script that
> causes it to hang like that... :)

I meant it takes 45+ to *POST*.  *Before the OS starts to bootstrap*
Just the POST.

Peace...  Sridhar
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue