Hello,
I have a N3K-C3064PQ-10GX and i have 350 Vlans on it and i am monitoring it from 2x server (one of them is observium and other one is prtg) and i have applied an acl on my uplink to drop snmp packets and snmptrap packet except for my trusted ips. but yesterday from 9.00PM my cpu load was at %70 and prolonged up to 11.00pm and when i execute 'show process cpu sort' the output was :
PID Runtime(ms) Invoked uSecs 1Sec Process
----- ----------- -------- ----- ------ -----------
3530 21871511 143773394 152 25.54% snmpd
3613 418 3372 124 20.79% netstack
3395 121460885 2147483647 45 11.85% bcm_usd
3534 124 66 1881 0.94% adjmgr
3569 211 2116 99 0.94% arp
3453 2884415 20002223 144 0.44% ipfib
3679 1522539 46052299 33 0.44% cfs
so the snmp was using the most cpu,
i have checked with debug snmp pkt to see if i am receiving many packets but i see only 3 packet per 30 second then i was checking my uplink mirror and I see there was no attack and everything was working properly and i did not find anything strange,
here is my sh ver output :
Software
BIOS: version 2.8.0
loader: version N/A
kickstart: version 6.0(2)U6(7)
system: version 6.0(2)U6(7)
Power Sequencer Firmware:
Module 1: version v1.0
SFP uC: version 2.12
BIOS compile time: 12/22/2015
kickstart image file is: bootflash:///n3000-uk9-kickstart.6.0.2.U6.7.bin
kickstart compile time: 7/28/2016 9:00:00 [07/28/2016 20:10:23]
system image file is: bootflash:///n3000-uk9.6.0.2.U6.7.bin
system compile time: 7/28/2016 9:00:00 [07/28/2016 20:42:38]
the only thing that i guess is that was a bug or procotol flap or such things, can anyone help me on this?
i have in mind to upgrade my switch to nxos.7.0.3.I7.8 as this is the cisco suggested nxos.
if this case happen again what thing should i check?
Thank you.
