[fm-discuss] Proposal: libtopo eumeration of power supplies and fans via IPMI

[Eric Schrock]

On Tue, Jan 08, 2008 at 12:43:36PM -0800, Cynthia McGuire wrote:
> 
> 
> Eric Schrock wrote:
> 
> >It would certainly be nice to move 'motherboard' under the chassis as
> >well as the 'bay' nodes, but I know in the past it has been inferred
> >that doing so will break existing diagnoses.  Is it allowable to
> >rearrange the topology in this fashion?
> 
> Interface-wise, you may restructure the topology.  The path elements and 
> their construction is not considered an interface (or ARC'd) according 
> to the FMRI specification.  However, (as you noted) making such a change 
> could break diagnosis if the DE makes assumptions about the hc path.
> 
> For x64: CPU, mem and PCIE should all be fine as they use eft rules that 
> elide the 'motherboard' portion anyway.  (Check with Gavin and Stephen 
> Hanson to be sure).  I'm not sure about the disk diagnosis engine.

The disk DE doesn't do true diagnosis in any real sense, it just does a
1:1 translation between ereports and diagnosis.  So there is no impact
on active cases.  You would have the case where the resource would no
longer exist in the system and 'fmadm faulty' wouldn't show it.  But it
would immediately diagnose the same failure and generate a fault using
the new FMRI.

> On SPARC: I'm almost certain there is software (DE or other) that 
> depends on the path construction.

Does SPARC use the 'motherboard=0' root?  I assumed that all or some
systems used 'chassis' since that exists in the HC topology already.

> Are there (silk-screened) labels assigned on the physical machines?  Is 
> it possible to track the components by some serial ID or other identity 
> information?

On systems shipped by Sun, yes.  I don't know about other systems.  IPMI
does support FRU locator records, and the fan modules do have stickers
on them with serial numbers, but I doubt there is any way to read that
information.  If there is, the current systems don't currently export it
(at least that I've seen).  I have seen FRU locator records for fan
boards, but not fan modules or fans themselves.

I do have systems that report FRU records for power supplies, so we
should be able to generate serial numbers for these.  From ipmitool
output:

FRU Device Description : ps0.fru (ID 31)
 Board Product         : PWR SUPPLY, A217, 1050W 
 Board Serial          : 0000000-0727A301EF
 Board Part Number     : 300-1897-03

I'll look into incorporating FRU locator data in to the topology.

Thanks,

- Eric

--
Eric Schrock, FishWorks                        http://blogs.sun.com/eschrock