cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
3084
Views
0
Helpful
7
Replies

6509 high CPU utilization

luisamato
Level 1
Level 1

Hello everybody, I have a situation, and y wanted to investigate if this is normal or not.

On a 6509 router ( 720 supervisor ) I start a "ping X.X.X.X size 5000 repeat 100000" from the supervisor to any host, and the CPU is incremented to 100%.

Can anyone affirm that this is usual ? I read this article and I think it is! (http://www.cisco.com/en/US/products/hw/switches/ps708/products_tech_note09186a00804916e0.shtml)

Here is the output of a show proccess with the ping running.

Without the ping it is at 5%.

...................................

#show processes cpu

CPU utilization for five seconds: 100%; one minute: 100%; five minutes: 99%

PID 5Sec 1Min 5Min Process

1 0.0% 0.1% 0.3% kernel

3 0.0% 0.0% 0.0% devc-pty

4 0.0% 0.0% 0.0% devc-mistral.proc

5 0.0% 0.0% 0.0% pipe

4102 0.0% 0.0% 0.0% dumper.proc

4103 0.0% 0.0% 0.0% pcmcia_driver.proc

4104 0.0% 0.0% 0.0% bflash_driver.proc

12297 0.0% 0.0% 0.0% mqueue

12298 0.0% 0.0% 0.0% flashfs_hes.proc

12299 0.0% 0.0% 0.0% dfs_bootdisk.proc

12300 0.0% 0.0% 0.0% ldcache.proc

12301 0.0% 0.0% 0.0% watchdog.proc

12302 0.0% 0.0% 0.0% syslogd.proc

12303 0.0% 0.0% 0.0% name_svr.proc

12304 0.0% 0.0% 0.0% wdsysmon.proc

12305 0.0% 0.0% 0.0% sysmgr.proc

16386 0.0% 0.0% 0.0% chkptd.proc

16402 0.0% 0.0% 0.0% sysmgr.proc

16403 0.0% 0.0% 0.0% syslog_dev.proc

16404 0.0% 0.0% 0.0% itrace_exec.proc

16405 0.0% 0.0% 0.0% packet.proc

16406 0.0% 0.0% 0.0% installer.proc

16407 88.4% 88.3% 86.8% ios-base

16408 0.0% 0.0% 0.0% fh_fd_rf.proc

16409 0.0% 0.0% 0.0% fh_fd_oir.proc

16410 0.0% 0.0% 0.0% fh_fd_env.proc

16411 0.0% 0.0% 0.0% fh_fd_cli.proc

16412 0.0% 0.0% 0.0% fh_metric_dir.proc

16413 0.0% 0.0% 0.0% fh_fd_test.proc

16414 0.0% 0.0% 0.0% fh_fd_snmp.proc

16415 0.0% 0.0% 0.0% fh_fd_none.proc

16416 0.0% 0.0% 0.0% fh_fd_intf.proc

16417 0.0% 0.0% 0.0% fh_fd_gold.proc

16418 0.0% 0.0% 0.0% fh_fd_timer.proc

16419 0.0% 0.0% 0.0% fh_fd_ioswd.proc

16420 0.0% 0.0% 0.0% fh_fd_counter.proc

16421 0.0% 0.0% 0.0% fh_server.proc

16422 0.0% 0.0% 0.0% fh_policy_dir.proc

16423 0.0% 0.0% 0.0% call_home.proc

16424 0.0% 0.0% 0.0% ipfs_daemon.proc

16425 0.0% 0.0% 0.0% fh_fd_config.proc

16426 2.2% 2.0% 2.0% raw_ip.proc

16427 0.0% 0.0% 0.0% inetd.proc

16428 8.4% 8.7% 8.6% tcp.proc

16429 0.0% 0.0% 0.0% udp.proc

16430 0.0% 0.0% 0.0% iprouting.iosproc

16431 0.5% 0.4% 0.4% cdp2.iosproc

1 Accepted Solution

Accepted Solutions

Edison Ortiz
Hall of Fame
Hall of Fame

The packet is too large and needs to be fragmented which requires CPU cycles.

What you are seeing is normal, avoid packet fragmentation at all cost.

HTH,

__

Edison.

View solution in original post

7 Replies 7

Edison Ortiz
Hall of Fame
Hall of Fame

The packet is too large and needs to be fragmented which requires CPU cycles.

What you are seeing is normal, avoid packet fragmentation at all cost.

HTH,

__

Edison.

Thank you for your soon answer.

Now please look at this command. Is this CPU consumption normal, why. Now we haven't got any fragmentation.

This is the command:

#ping 172.17.8.66 repeat 10000

This is the status of the CPU wile the ping runs:

#sh processes cpu history

99988877765432

3208529504788592223334543222233333334444566766554455673222

100

90 *****

80 ********

70 *********

60 ***********

50 ************

40 *************

30 **************

20 **************

10 *************** * ******** ****

0....5....1....1....2....2....3....3....4....4....5....5....

0 5 0 5 0 5 0 5 0 5

CPU% per second (last 60 seconds)

* = maximum CPU% # = average CPU%

G100603RT000#sh processes cpu

CPU utilization for five seconds: 99%; one minute: 37%; five minutes: 15%

PID 5Sec 1Min 5Min Process

1 0.0% 0.1% 0.1% kernel

3 0.0% 0.0% 0.0% devc-pty

4 0.0% 0.0% 0.0% devc-mistral.proc

5 0.0% 0.0% 0.0% pipe

4102 0.0% 0.0% 0.0% dumper.proc

4103 0.0% 0.0% 0.0% pcmcia_driver.proc

4104 0.0% 0.0% 0.0% bflash_driver.proc

12297 0.0% 0.0% 0.0% mqueue

12298 0.0% 0.0% 0.0% flashfs_hes.proc

12299 0.0% 0.0% 0.0% dfs_bootdisk.proc

12300 0.0% 0.0% 0.0% ldcache.proc

12301 0.0% 0.0% 0.0% watchdog.proc

12302 0.0% 0.0% 0.0% syslogd.proc

12303 0.0% 0.0% 0.0% name_svr.proc

12304 0.0% 0.0% 0.0% wdsysmon.proc

12305 0.0% 0.0% 0.0% sysmgr.proc

16386 0.0% 0.0% 0.0% chkptd.proc

16402 0.0% 0.0% 0.0% sysmgr.proc

16403 0.0% 0.0% 0.0% syslog_dev.proc

16404 0.0% 0.0% 0.0% itrace_exec.proc

16405 0.0% 0.0% 0.0% packet.proc

16406 0.0% 0.0% 0.0% installer.proc

PID 5Sec 1Min 5Min Process

16407 65.0% 23.9% 9.8% ios-base

16408 0.0% 0.0% 0.0% fh_fd_rf.proc

16409 0.0% 0.0% 0.0% fh_fd_oir.proc

16410 0.0% 0.0% 0.0% fh_fd_env.proc

16411 0.0% 0.0% 0.0% fh_fd_cli.proc

16412 0.0% 0.0% 0.0% fh_metric_dir.proc

16413 0.0% 0.0% 0.0% fh_fd_test.proc

16414 0.0% 0.0% 0.0% fh_fd_snmp.proc

16415 0.0% 0.0% 0.0% fh_fd_none.proc

16416 0.0% 0.0% 0.0% fh_fd_intf.proc

16417 0.0% 0.0% 0.0% fh_fd_gold.proc

16418 0.0% 0.0% 0.0% fh_fd_timer.proc

16419 0.0% 0.0% 0.0% fh_fd_ioswd.proc

16420 0.0% 0.0% 0.0% fh_fd_counter.proc

16421 0.0% 0.0% 0.0% fh_server.proc

16422 0.0% 0.0% 0.0% fh_policy_dir.proc

16423 0.0% 0.0% 0.0% call_home.proc

16424 0.0% 0.0% 0.0% ipfs_daemon.proc

16425 0.0% 0.0% 0.0% fh_fd_config.proc

16426 4.5% 1.7% 0.6% raw_ip.proc

16427 0.0% 0.0% 0.0% inetd.proc

16428 28.9% 10.1% 3.5% tcp.proc

16429 0.0% 0.0% 0.0% udp.proc

16430 0.0% 0.0% 0.0% iprouting.iosproc

PID 5Sec 1Min 5Min Process

16431 0.0% 0.0% 0.0% cdp2.iosproc

114738 0.0% 0.0% 0.0% tftp_fs.proc

Don't ping from the box as the packet will be processed by the CPU.

Do your test from devices sitting behind the box, don't stress the box with pings or any other communication tool.

In some circumstances, even a simple show run can cause a spike in the CPU.

HTH,

__

Edison.

Many MLS devices, such as a 6500, use special hardware to deliver their high perfomance for moving traffic through the device. Any processing that's not supported by their special hardware is processed on some kind of general processor which often can not deliver the same level of high performance. Using a network device as a host device, such as generating pings or responding to pings, are often functions not supported by their special hardware and may cause their CPU to spike.

Hello everybody,

I have a high cpu utilization problem too. I have a Cat6509 with Sup32-PISA-10GE and when I upgraded from 12.2(18)ZY2 to 12.2(18)ZYA and switch booted, the following process was utilizing CPU at 99%:

140 3426368 21182 161758 99.92% 99.53% 99.53% 0 Xscale Sibyte Co

I reloaded again and now all seems to be OK.

Does anyone know what could have happened?

You may be hitting a bug, can you backrev to ZY2 and see if you can reproduce it?

__

Edison.

Gotta be a bug. We installed brand new switches and everyone is doing the same thing. Rebooting clears it up. Same Version, Same Process.

Getting Started

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: