08-05-2021 01:20 PM
Hello,
Several weeks I've found unusually high CPU usage on my cisco nexus 6001.
It's up to 30% which leads to slow ssh cli, but console connection works normally.
How to find the issue?
CPU utilization for five seconds: 30%/3%; one minute: 31%; five minutes: 28% PID Runtime(ms) Invoked uSecs 5Sec 1Min 5Min TTY Process ----- ----------- -------- ----- ------ ------ ------ --- ----------- 3814 2080832470 2042661456 1 0.31% 0.32% 0.27% - snmpd 3804 1165904730 3482137267 0 0.25% 0.18% 0.18% - fwm 3862 310 2407098321 0 0.53% 0.21% 0.18% - pktmgr 3733 394130 303550778 0 0.13% 0.13% 0.13% - sysinfo 4089 1095081660 3302435989 0 0.13% 0.13% 0.12% - dhcp_snoop 3613 868045800 135990837 6 0.11% 0.11% 0.10% - pfma 3919 370 3942619880 0 0.11% 0.11% 0.10% - netstack 3974 230 3329496747 0 0.05% 0.06% 0.06% - igmp 3763 2147483560 534849954 4 0.03% 0.04% 0.03% - bigsurusd 4091 247308700 2783399895 0 0.03% 0.03% 0.02% - interface-vlan 4087 161992480 741203171 0 0.01% 0.02% 0.02% - lacp 4069 236861100 2147466773 0 0.01% 0.02% 0.01% - stp 3742 156547440 2080626658 0 0.03% 0.02% 0.01% - pfstat 4727 180 736849811 0 0.01% 0.01% 0.01% - bgp 3670 135685250 2102246259 0 0.01% 0.01% 0.01% - feature-mgr
show system internal processes cpu top - 23:13:25 up 1151 days, 12:05, 1 user, load average: 3.04, 2.27, 2.36 Tasks: 247 total, 1 running, 239 sleeping, 0 stopped, 7 zombie Cpu(s): 5.1%us, 3.6%sy, 0.0%ni, 90.1%id, 0.0%wa, 0.0%hi, 1.1%si, 0.0%st Mem: 8238120k total, 3989296k used, 4248824k free, 160k buffers Swap: 0k total, 0k used, 0k free, 1857572k cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 3862 root 20 0 255m 33m 23m S 7.6 0.4 77237:04 pm 29231 admin 20 0 3604 1544 1148 R 5.7 0.0 0:00.06 top 3613 root 20 0 209m 29m 20m S 1.9 0.4 14496:59 pfm 3857 root 20 0 332m 72m 22m S 1.9 0.9 6224:06 arp 3919 root 20 0 367m 76m 29m S 1.9 1.0 62608:48 netstack 4089 root 20 0 271m 31m 22m S 1.9 0.4 18285:45 dhcp_snoop 1 root 20 0 1988 664 584 S 0.0 0.0 20:12.62 init 2 root 15 -5 0 0 0 S 0.0 0.0 0:00.01 kthreadd 3 root RT -5 0 0 0 S 0.0 0.0 0:56.43 migration/0 4 root 15 -5 0 0 0 S 0.0 0.0 506:10.35 ksoftirqd/0 5 root RT -5 0 0 0 S 0.0 0.0 19:26.09 watchdog/0 6 root RT -5 0 0 0 S 0.0 0.0 0:43.71 migration/1 7 root 15 -5 0 0 0 S 0.0 0.0 1005:53 ksoftirqd/1
No traffic loops exists.
Cpu and used memory looks like this:
Thanks in advance.
08-05-2021 11:17 PM
Presumably similar commands will apply to your platform too.
M.
08-14-2021 03:56 AM
Yes I know this document.
Unfortunately this don't help ((
08-14-2021 04:48 AM
Hello,
this is just a wild guess, but there is a bug where unused 1GB SFPs cause high CPU and switch latency. What is the output of:
sh hardware internal bigsur logs runtime
Here is the link to the bug:
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide