4506 High Utilisation peaks

[wizo]

Hi Guys

I have 2 4506's that are running a collapsed core at a smaller site, one of my colleagues advised me that there were a significant number of cpu peaks of 60%, so I wanted to find out what was causing the peaks when the average cpu utilisation was only around 25%.

I started out with a show proc cpu history and sure enough the peaks were there, 5 second cpu bursts every 2 minutes, and even higher spikes of 80-90% every 4 hours (not sure of duration)

so just looking at the 60% issue, i did the following

Switch#show processes cpu

31 26534564 9274947 2860 43.19% 8.21% 6.95% 0 Cat4k Mgmt LoPri

this appeared to be the cause of the 60% 5 second bursts so following the cisco docs i then did a

Switch#show platform health

K2AclCamMan Audit re 1.00 42.70 10 5 100 500 0 0 0 16:48

and this little bugger appears to be the culprit of the 60% but for the life of me I cannot find any info as to what the process is or does....

breaking it apart...

Acl...access control list?
Cam...Cam table?
Man...Management?
Audit..ok.
re ....????

is it auditing cam tables for aged entries??? auditing acl's?????

the next in the process is Switch#show platform cpu packet statistics but i dont seem to get anything that stands out to me..

Switch#sh platform cpu packet statistics
Packets Dropped In Hardware By CPU Subport (txQueueNotAvail)

CPU Subport TxQueue 0 TxQueue 1 TxQueue 2 TxQueue 3
------------ --------------- --------------- --------------- ---------------
0 67 0 0 7822676

Packets Dropped In Processing Overall

Total 5 sec avg 1 min avg 5 min avg 1 hour avg
-------------------- --------- --------- --------- ----------
118385 0 0 0 0

Packets Dropped In Processing by CPU event

Event Total 5 sec avg 1 min avg 5 min avg 1 hour avg
----------------- -------------------- --------- --------- --------- ----------
Control Packet 118227 0 0 0 0
Input Acl 112 0 0 0 0
SA Miss 23 0 0 0 0

Packets Dropped In Processing by Priority

Priority Total 5 sec avg 1 min avg 5 min avg 1 hour avg
----------------- -------------------- --------- --------- --------- ----------
Normal 23 0 0 0 0
Medium 46 0 0 0 0
High 112 0 0 0 0
Crucial 118227 0 0 0 0

Packets Dropped In Processing by Reason

Reason Total 5 sec avg 1 min avg 5 min avg 1 hour avg
------------------ -------------------- --------- --------- --------- ----------
SrcAddrTableFilt 12 0 0 0 0
L2DstDrop 13 0 0 0 0
L2DstDropInAcl 118227 0 0 0 0
NoFloodPorts 133 0 0 0 0

Total packet queues 16

Packets Received by Packet Queue

Queue Total 5 sec avg 1 min avg 5 min avg 1 hour avg
---------------------- --------------- --------- --------- --------- ----------
Esmp 69491565 226 197 157 145
Control 2818251 9 0 1 0
Host Learning 70099 0 0 0 0
L3 Fwd Medium 5 0 0 0 0
L3 Fwd Low 2378131 1 0 0 0
L2 Fwd High 377 0 0 0 0
L2 Fwd Low 7867002 22 13 14 6
L3 Rx Low 195474 9 0 0 0
ACL fwd(snooping) 4915362 15 8 5 0

Packets Dropped by Packet Queue

Queue Total 5 sec avg 1 min avg 5 min avg 1 hour avg
---------------------- --------------- --------- --------- --------- ----------
Host Learning 663 0 0 0 0



hope someone can point me in the right direction....

cheers

Greg

 

[vipergg]

doesn't even sound like a problem if its just short peaks . The cpu will attempt to use all available resources to do tasks if it is available. You browse the cisco guide if you feel its a problem.

http://www.cisco.com/en/US/products/hw/switches/ps663/products_tech_note09186a00804cef15.shtml

 

[wizo]

I personally didnt feel it to be a problem, but others did, so I just wanted to understand exactly what the processes are that are running

 

[burtsbees]

Run it through Output Interpreter on Cisco.com (requires CCO account).

/

tim@tim-laptop ~ $ sudo apt-get install windows
Reading package lists... Done
Building dependency tree
Reading state information... Done
E: Couldn't find package windows...Thank Goodness!