Switch Fabric Resources Peak 100%

Unanswered Question
Jan 19th, 2010
User Badges:

I have been looking around at our 6509 and noticed that the peak utilization on Module 3 was at 100%. I thought this may be a one off and returned later and found that the recorded time for the peak was the current time. It almost saying that it always the peak however if I go to our other 6509 this just shows peak well in the past like module 5.


MY6509#show fabric utilization detail
  Fabric utilization:     Ingress                    Egress
    Module  Chanl  Speed  rate  peak                 rate  peak              
    3       0        20G    1%  100% @10:05 19Jan10    1%  100% @10:05 19Jan10
    3       1        20G    1%    0%                   1%    0%              
    4       0        20G    0%    0%                   0%    0%              
    4       1        20G    1%    0%                   1%    0%              
    5       0        20G    0%  100% @20:12 22Sep08    2%  100% @20:12 22Sep08
~~~

OR


MY6509#show platform hardware capacity fabric
Switch Fabric Resources
  Bus utilization: current: 0%, peak was 8% at 15:15:15 GMT Sat Dec 5 2009
  Fabric utilization:     Ingress                    Egress
    Module  Chanl  Speed  rate  peak                 rate  peak              
    3       0        20G    0%  100% @10:21 19Jan10    0%  100% @10:21 19Jan10
    3       1        20G    0%    0%                   1%    0%              
    4       0        20G    0%    0%                   0%    0%              
    4       1        20G    1%    0%                   0%    0%              
    5       0        20G    0%  100% @20:12 22Sep08    1%  100% @20:12 22Sep08

~~~

IOS: s72033-entservicesk9_wan-vz.122-18.SXF15.bin


MY6509#show platform hardware capacity
System Resources
  PFC operating mode: PFC3B
  Supervisor redundancy mode: administratively sso, operationally sso
  Switching resources: Module   Part number               Series      CEF mode
                       3        WS-X6708-10GE             CEF720          dCEF
                       4        WS-X6708-10GE             CEF720          dCEF
                       5        WS-SUP720-3B          supervisor           CEF
                       6        WS-X6708-10GE             CEF720          dCEF
                       7        WS-X6708-10GE             CEF720          dCEF
                       8        WS-X6708-10GE             CEF720          dCEF
                       9        WS-X6748-GE-TX            CEF720           CEF


Question: Should I be concerned, do I have an isue which needs to be investigated further?



  

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 0 (0 ratings)
Loading.
Giuseppe Larosa Tue, 01/19/2010 - 15:26
User Badges:
  • Super Silver, 17500 points or more
  • Hall of Fame,

    Founding Member

Hello Alan,

how many 10GE ports are in use on that WS-6708?

channel fabric is 20G + 20G and cannot support 8 tengiga full rate.


you should start to look at interface statistics looking for ignored packets or output drops.

If interface stats are clear you should be fine.


module 5 is the supervisor and I would see that a peak there is more problematic, but it happened in the past.


I see that you have installed 5 WS-6708 what power supplies are you using?



Hope to help

Giuseppe

alanjbrown Thu, 01/21/2010 - 00:23
User Badges:

Giuseppe,

Of the 8 x Te, 4 are not connected, 3 are trunked and 1 routing. There a few output drops but very low in compared to the amount of traffic. let's say around 60 over a couple of weeks which is acceptable. I do however have 50,000 input drops/flushes over a 2 week period on the routed interface. All interfaces are running at very low capacity.


We are using 2 x WS-CAC-4000W-INT.


Alan

Actions

This Discussion

Related Content