[ofa-general] Cisco 7008 Internal/External Port Mappings

[Charles Taylor]

Nope.   Looks pretty much the same as ibnetdiscover without the "- 
g".    It would be really useful if it *did* work.  :)

Thanks for the suggestion though.     I think we just may be stuck  
mapping it empirically.    I'm sure the pattern will become obvious  
quickly enough.

Charlie Taylor
UF HPC Center

On Jul 10, 2008, at 5:36 PM, Hal Rosenstock wrote:

> On Thu, 2008-07-10 at 17:27 -0400, Charles Taylor wrote:
>> We've got a couple of Cisco 7008/Topspin 270 core switches.    Much  
>> of
>> the diagnostic and topology information we get from the ib diag tools
>> in OFED refer to the internal ports rather than the actual external
>> slot/port.
>
> Does ibnetdiscover -g work ? There was some work done to do this  
> mapping
> for Cisco but it may be out of date.
>
> -- Hal
>
>>    We could deduce the mappings, I suppose, by tracing
>> cables and using ibtracert between nodes but that seems like a real
>> pain.    I was wondering if anyone on this list knows how that  
>> mapping
>> works.     The chassis itself consists of 8 "ib port" blades in the
>> front labelled as slots 1 - 8.   There are then six "fabric
>> controller" slots in the back of the chassis labelled 9 - 14.
>> Slots 11 and 12 are "special" as they hold the "master" and "standby
>> master" fabric controllers.
>>
>> Seems like the FC's in slots 11 and 12 connect the the FCs in slots
>> 9,10,13,14 which in turn connect to the external port blades in slots
>> 1 - 8.
>>
>> Again, all of the diagnostic information refers to the internal FC
>> ports and mapping those to the external ports (to which we actually
>> connect leaf switches) seems non-trivial.    I think there was even a
>> document somewhere that described the mapping.   We just can't find  
>> it
>> now.
>>
>> Thanks,
>>
>> Charlie Taylor
>> UF HPC Center
>> _______________________________________________
>> general mailing list
>> general at lists.openfabrics.org
>> http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
>>
>> To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
>