Dragon0x40
asked on
Should CEF be enabled?
It seems that my switch is process switching all packets.
Is that a bad thing?
sh cef state
CEF Status:
RP instance
common CEF disabled
IPv4 CEF Status:
CEF disabled/not running
dCEF disabled/not running
CEF switching disabled/not running
universal per-destination load sharing algorithm, id 83852352
IPv6 CEF Status:
CEF disabled/not running
dCEF disabled/not running
universal per-destination load sharing algorithm, id 83852352
sh int fa1/0/1 stats
FastEthernet1/0/1
Switching path Pkts In Chars In Pkts Out Chars Out
Processor 0 0 11260657 720682048
Route cache 0 0 0 0
Total 0 0 11260657 720682048
sh int gi1/0/1 stats
GigabitEthernet1/0/1
Switching path Pkts In Chars In Pkts Out Chars Out
Processor 0 0 294692675 18865399070
Route cache 0 0 0 0
Total 0 0 294692675 18865399070
sh proc cpu history
99395744471468293722483529 2927323623 2793733384 3392947221 7334739263 5273
92094981769421968853337621 0985870412 6390703929 1055967399 5016499454 4489
100 * * * * * * * *
90 ** * * * * * * * *
80 ** * * * * * * * * * * * * * * * * * * *
70 ** * * * * * * * * * * ** * * * * * * * * * *
60 ** * * * ** * * * * * * * * ** * * * * * * * * * *
50 ** **** ** ** * * * * * * * * ** * ** * *** * ** * * * *
40 ** ******* *** *** **** * * ** * ** * *** * *** * **** * * **
30 ********** ******** **** * ******* ****************** * ******* *** **
20 ************************** ********** ********** ********** ********** ****
10 ########################## ########## ########## ########## ########## ####
0....5....1....1....2....2 ....3....3 ....4....4 ....5....5 ....6....6 ....7.
0 5 0 5 0 5 0 5 0 5 0 5 0
CPU% per hour (last 72 hours)
* = maximum CPU% # = average CPU%
Is that a bad thing?
sh cef state
CEF Status:
RP instance
common CEF disabled
IPv4 CEF Status:
CEF disabled/not running
dCEF disabled/not running
CEF switching disabled/not running
universal per-destination load sharing algorithm, id 83852352
IPv6 CEF Status:
CEF disabled/not running
dCEF disabled/not running
universal per-destination load sharing algorithm, id 83852352
sh int fa1/0/1 stats
FastEthernet1/0/1
Switching path Pkts In Chars In Pkts Out Chars Out
Processor 0 0 11260657 720682048
Route cache 0 0 0 0
Total 0 0 11260657 720682048
sh int gi1/0/1 stats
GigabitEthernet1/0/1
Switching path Pkts In Chars In Pkts Out Chars Out
Processor 0 0 294692675 18865399070
Route cache 0 0 0 0
Total 0 0 294692675 18865399070
sh proc cpu history
99395744471468293722483529
92094981769421968853337621
100 * * * * * * * *
90 ** * * * * * * * *
80 ** * * * * * * * * * * * * * * * * * * *
70 ** * * * * * * * * * * ** * * * * * * * * * *
60 ** * * * ** * * * * * * * * ** * * * * * * * * * *
50 ** **** ** ** * * * * * * * * ** * ** * *** * ** * * * *
40 ** ******* *** *** **** * * ** * ** * *** * *** * **** * * **
30 ********** ******** **** * ******* ****************** * ******* *** **
20 **************************
10 ##########################
0....5....1....1....2....2
0 5 0 5 0 5 0 5 0 5 0 5 0
CPU% per hour (last 72 hours)
* = maximum CPU% # = average CPU%
ASKER CERTIFIED SOLUTION
membership
Create a free account to see this answer
Signing up is free and takes 30 seconds. No credit card required.
What kind of switch is this and what network devices do it connect to?
SOLUTION
membership
Create a free account to see this answer
Signing up is free and takes 30 seconds. No credit card required.
ASKER
On a 3750 should I run "ip cef distributed" as I don't see a <cr> after "ip cef ?"
right now my processes show
sh proc cpu
CPU utilization for five seconds: 9%/2%; one minute: 7%; five minutes: 7%
Does this mean that 2% of my cpu is being used for process switching?
right now my processes show
sh proc cpu
CPU utilization for five seconds: 9%/2%; one minute: 7%; five minutes: 7%
Does this mean that 2% of my cpu is being used for process switching?
IP Cef should be it.
ASKER
WS-C3750-48TS-S
#ip cef ?
distributed Distributed Cisco Express Forwarding
linecard CEF linecard commands
load-sharing Load sharing
optimize Optimizations
traffic-statistics Enable collection of traffic statistics
#ip cef ?
distributed Distributed Cisco Express Forwarding
linecard CEF linecard commands
load-sharing Load sharing
optimize Optimizations
traffic-statistics Enable collection of traffic statistics
Are you saying that is it not excepting just "IP CEF"? Have you tried just entering it. Just because <cr> is not listed doesn't mean it is not an option.
ip cef distributed
is what you want
is what you want
@ddiazp
Can you explain why?
Can you explain why?
SOLUTION
membership
Create a free account to see this answer
Signing up is free and takes 30 seconds. No credit card required.
Also, in addition, to perform any further CEF/dCEF functions, you need to have CEF/dCEF running at least:
•Enabling CEF or dCEF (Required)
•Configuring Load Balancing for CEF (Optional)
•Configuring Network Accounting for CEF (Optional)
•Configuring Distributed Tunnel Switching for CEF (Optional)
more info: http://www.cisco.com/en/US/docs/ios/12_1/switch/configuration/guide/xcdcefc.html
•Enabling CEF or dCEF (Required)
•Configuring Load Balancing for CEF (Optional)
•Configuring Network Accounting for CEF (Optional)
•Configuring Distributed Tunnel Switching for CEF (Optional)
more info: http://www.cisco.com/en/US/docs/ios/12_1/switch/configuration/guide/xcdcefc.html
@ddiazp
Thanks
Thanks
My only question is stated above. Have the author tried entering ip cef? I still don't see that not being an available command.
ASKER
I can't enter "ip cef" unless I get authorization.
I thought it would be easy just enter "ip cef distributed" and be done.
Wouldn't just that command help with not punting all packets to the CPU?
I thought it would be easy just enter "ip cef distributed" and be done.
Wouldn't just that command help with not punting all packets to the CPU?
SOLUTION
membership
Create a free account to see this answer
Signing up is free and takes 30 seconds. No credit card required.
can you paste the top 10 from this
show processes cpu sorted 5min
Greg
show processes cpu sorted 5min
Greg
ASKER
I will post sh processes tomorrow.
If we really should have cef on all our access layer switches how do I convince the people I work with?
If we really should have cef on all our access layer switches how do I convince the people I work with?
ASKER
show processes cpu sorted 5min
CPU utilization for five seconds: 5%/0%; one minute: 6%; five minutes: 6%
PID Runtime(ms) Invoked uSecs 5Sec 1Min 5Min TTY Process
212 66656092 277670662 240 0.15% 0.33% 0.34% 0 Spanning Tree
150 61978233 30163250 2054 0.15% 0.18% 0.20% 0 HRPC qos request
141 78542874 598514645 131 0.15% 0.18% 0.16% 0 Hulc LED Process
4 30834452 3033192 10165 0.00% 0.15% 0.12% 0 Check heaps
221 443 173 2560 0.00% 0.26% 0.10% 1 SSH Process
104 38443784 52361199 734 0.00% 0.11% 0.09% 0 hpm counter proc
105 24869316 108972972 228 0.15% 0.09% 0.09% 0 HRPC pm-counters
69 14395718 70120457 205 0.00% 0.02% 0.04% 0 hrpc <- response
306 1340675 716662 1870 0.00% 0.01% 0.02% 0 SNMP ENGINE
65 62802391344647164 4 0.00% 0.04% 0.01% 0 Fifo Error Detec
CPU utilization for five seconds: 5%/0%; one minute: 6%; five minutes: 6%
PID Runtime(ms) Invoked uSecs 5Sec 1Min 5Min TTY Process
212 66656092 277670662 240 0.15% 0.33% 0.34% 0 Spanning Tree
150 61978233 30163250 2054 0.15% 0.18% 0.20% 0 HRPC qos request
141 78542874 598514645 131 0.15% 0.18% 0.16% 0 Hulc LED Process
4 30834452 3033192 10165 0.00% 0.15% 0.12% 0 Check heaps
221 443 173 2560 0.00% 0.26% 0.10% 1 SSH Process
104 38443784 52361199 734 0.00% 0.11% 0.09% 0 hpm counter proc
105 24869316 108972972 228 0.15% 0.09% 0.09% 0 HRPC pm-counters
69 14395718 70120457 205 0.00% 0.02% 0.04% 0 hrpc <- response
306 1340675 716662 1870 0.00% 0.01% 0.02% 0 SNMP ENGINE
65 62802391344647164 4 0.00% 0.04% 0.01% 0 Fifo Error Detec