11-09-2014 08:00 PM - edited 03-03-2019 07:39 AM
Hi All,
I have a Cisco 2851 router which placed on one of our company node. Currently it's handling 80M of traffic. Recently ?I have monitor increased in CPU due to SNMP. Appreciate if someone can help me with this ??
Following are the Logs taken from the router.
Router Version:-
"flash:c2800nm-adventerprisek9-mz.151-4.M8.bin"
2851# show proc cpu sorted
CPU utilization for five seconds: 84%/64%; one minute: 86%; five minutes: 75%
PID Runtime(ms) Invoked uSecs 5Sec 1Min 5Min TTY Process
378 5408556 11245357 480 5.83% 6.46% 3.89% 0 IP SNMP
380 4739792 5616341 843 5.43% 6.49% 3.94% 0 SNMP ENGINE
142 3982236 8455561 470 3.67% 3.93% 2.80% 0 IP Input
379 936308 5615860 166 1.11% 1.25% 0.77% 0 PDU DISPATCHER
366 747504 735117 1016 0.71% 0.48% 0.49% 0 MRIB Trans
116 11196 14099948 0 0.55% 0.57% 0.55% 0 Ethernet Msec Ti
371 170552 782130 218 0.39% 0.27% 0.26% 0 MFIB_mrib_read
203 791500 983787 804 0.31% 0.38% 0.39% 0 ADJ resolve proc
370 10484 5557602 1 0.31% 0.27% 0.26% 0 MFIB_mrib_write
364 232568 2761087 84 0.23% 0.26% 0.25% 0 PIM Process
2 70272 22341 3145 0.07% 0.06% 0.07% 0 Load Meter
189 9672 290274 33 0.07% 0.08% 0.07% 0 TCP Timer
313 920 558364 1 0.07% 0.02% 0.00% 0 Atheros LED Ctro
262 20168 140738 143 0.07% 0.01% 0.00% 0 OSPF-10 Router
106 568 111651 5 0.07% 0.00% 0.00% 0 COLLECT STAT COU
177 9160 151130 60 0.07% 0.07% 0.07% 0 CEF: IPv4 proces
289 6344 3487573 1 0.07% 0.07% 0.07% 0 MMON MENG
282 59712 223334 267 0.07% 0.06% 0.07% 0 BGP Router
372 8436 13966 604 0.07% 0.01% 0.00% 0 MFIB_rate
363 7600 1132583 6 0.07% 0.06% 0.07% 0 IGMP Input
set-2851#show snmp stats oid
time-stamp #of times requested OID
09:16:28 LK Nov 10 2014 327 ifDescr
09:16:27 LK Nov 10 2014 398 ifName
09:16:26 LK Nov 10 2014 396 ifPhysAddress
09:16:26 LK Nov 10 2014 2 ifLastChange
09:16:26 LK Nov 10 2014 2 ifOperStatus
09:16:25 LK Nov 10 2014 548 ipAddrEntry.2
09:16:23 LK Nov 10 2014 528 ifType
09:16:09 LK Nov 10 2014 1 internet.6.3.12.1.1
09:16:09 LK Nov 10 2014 1 internet.6.3.11.2.1.3
09:16:09 LK Nov 10 2014 1 internet.6.3.11.2.1.2
09:16:09 LK Nov 10 2014 1 internet.6.3.11.2.1.1
09:16:09 LK Nov 10 2014 1 internet.6.3.10.2.1.4
09:16:09 LK Nov 10 2014 1 internet.6.3.10.2.1.3
09:16:09 LK Nov 10 2014 1 internet.6.3.10.2.1.2
09:16:09 LK Nov 10 2014 1 internet.6.3.10.2.1.1
09:16:09 LK Nov 10 2014 2 internet.6.3.1.1.6.1
09:16:09 LK Nov 10 2014 1 cbQosCMPrePolicyByte
09:16:09 LK Nov 10 2014 2 cbQosCMPrePolicyByteOverflow
09:16:09 LK Nov 10 2014 1 cbQosCMPrePolicyByteOverflow
09:16:09 LK Nov 10 2014 1 cbQosCMPrePolicyByteOverflow
09:16:09 LK Nov 10 2014 5 cbQosCMPrePolicyPkt64
09:16:09 LK Nov 10 2014 1 cbQosPolicyMapDesc
09:16:09 LK Nov 10 2014 3 cbQosPolicyMapName
09:16:09 LK Nov 10 2014 12 cbQosParentObjectsIndex
09:16:09 LK Nov 10 2014 10 cbQosObjectsType
09:16:09 LK Nov 10 2014 1 cbQosObjectsType
09:16:09 LK Nov 10 2014 1 cbQosObjectsType
09:16:09 LK Nov 10 2014 13 cbQosConfigIndex
11-11-2014 08:50 PM
Any Help ???
11-12-2014 06:20 AM
Hi there,
looking at your process stats of the 84% util, 64% of that is due to CPU interupts. The SNMP process being high up the list is a red herring. You should investigate what traffic is being sent to the CPU for processing.
Following the steps in these documents should aid in troubleshooting:
http://www.cisco.com/c/en/us/support/docs/routers/10000-series-routers/15095-highcpu.html
http://www.cisco.com/c/en/us/support/docs/routers/7500-series-routers/41120-highcpu-interrupts.html
cheers,
Seb.
11-12-2014 09:44 AM
Thanks Seb !!!
But when I disable snmp-server community XXXX RO CPU utilization get down to 35 -40%... I suspected that one of my servers which using snmp to collect data may causing this problem... Any Ideas ????
12-03-2014 12:28 PM
Taking a look you can see that the OIDs being polled the most are:
09:16:25 LK Nov 10 2014 548 ipAddrEntry.2
09:16:23 LK Nov 10 2014 528 ifType
So you can either check the monitoring tool using that community and increase the polling intervals or set a view to those OIDs.
Those are huge OIDs so it is expected that being polled a lot cause high cpu, on this case it is weird as they only have 5%, that is not high cpu at all, you can check the show process cpu history, I think about a bug showing wrong information on the processes.
Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: