Sorry to hear about the borkification.

Silly question (and you don't have to reply to the list on this...) 
but... Did you follow the entire readme of the firmware, and have you 
done a complete A/C powercycle of the box? (ie: Remove the A/C power cord)

It's baroque, I know, and it's something they have pretty much addressed 
in the later versions of firmware... but... I'm just relating my 
*experience*, and pulling the A/C is the only way to get *everything* 
back to being 100% clean, including things like the FPGA and volatile 
areas of memory that are not actually part of the sp host memory (which 
is reset when you 'resetsc'.

I recall from a while ago, that going from some versions of firmware to 
other versions of firmware, there were some funny things that might have 
happened, usually as a result of what the technical community calls 
'hanging chads'. :)  Something that's previously not been used, or *was* 
set by an old version of firmware that we use differently now...

For the most part, these are complete un-reproducable, and in my 
experience, sometimes only cleared through an AC poweroff.

But - before you go getting all medieval on the platform make sure you 
have run the collectDebuginfo.sh script on the SP (the services guys 
should tell you how to do that) so we get a current view of the 
b0rkedness, and en explorer on the control domain.

And - You owe me a steak if an A/C power cycle fixes the issue.

Let me know how you go.

Cheers!

Nathan - NOT speaking for services on this one - Kroenert. :)


On  8/10/08 12:58 PM, Malcolm Herbert wrote:
> I have an open support call about this, however I was curious to know
> whether anyone had tried applying this firmware patch to fix the Time Of
> Day issues mentioned in Sun Alert 242106 ... 
> 
> So far it has not been all beer and skittles for me - although the
> firmware appears to have applied OK, ldmd will now coredump at OS boot
> and consequently I can't start any of the LDOMs on the host ... 
> 
> Initially the host was locking up solid after login in multi-user mode,
> however I discovered that this was due to not having applied kernel
> patch 127127-11 (and dependancies) ... this meant I can now get the OS
> to multi-user mode without hanging, but ldmd is still b0rk3d ... 
> 
> *sigh*
> 
> if anyone has any experience with it I'd love to hear from you - still
> awaiting a response from Sun about the issue ... 
> 

-- 


//////////////////////////////////////////////////////////////////
// Nathan Kroenert              nathan.kroenert at sun.com              //
// Senior Systems Engineer      Phone:  +61 3 9869 6255         //
// Global Systems Engineering   Fax:    +61 3 9869 6288         //
// Level 7, 476 St. Kilda Road                                  //
// Melbourne 3004   Victoria    Australia                       //
//////////////////////////////////////////////////////////////////

Reply via email to