04-14-2003 04:32 PM - edited 03-02-2019 06:39 AM
Hi,
We have a number of Catalyst 5500 switches interconnected via ATM backbone.
On one of the core switches, on the ATM module, a sh proc cpu shows a 94% on a 5sec CPU utilization (as below). I need help understanding what processes that causes the CPU utilization to spike. This sh proc cpu was captured when the users were experiencing slow network response...
#sh proc cpu
CPU utilization for five seconds: 94%/0%; one minute: 69%; five minutes: 66%
PID Runtime(ms) Invoked uSecs 5Sec 1Min 5Min TTY Process
1 697484 7535900 92 0.00% 0.00% 0.00% 0 Load Meter
2 1782692 5427614 328 0.00% 0.00% 0.00% 0 SSCOP Input
3 898100256 8055076 111495 0.00% 2.07% 2.37% 0 Check heaps
4 4 1 4000 0.00% 0.00% 0.00% 0 Pool Manager
5 4 2 2000 0.00% 0.00% 0.00% 0 Timers
6 257456 631072 407 0.00% 0.00% 0.00% 0 ARP Input
7 161376 20942 7705 0.00% 0.00% 0.00% 0 ATM ILMI Input
8 115628 46981 2461 0.00% 0.00% 0.00% 0 ILMI Process
9 4 1 4000 0.00% 0.00% 0.00% 0 SERIAL A'detect
10 0 2 0 0.00% 0.00% 0.00% 0 Cat5K VTP HELPER
11 0 3 0 0.00% 0.00% 0.00% 0 Cat5K ATM DUALPHY
12 4 2 2000 0.00% 0.00% 0.00% 0 ATM OAM Input
13 354171656 607141381 583 1.30% 0.82% 0.78% 0 C5K ATM IPC
14 132045624 112885881 1169 1.96% 0.24% 0.05% 0 IP Input
15 20 10 2000 0.00% 0.00% 0.00% 0 ATMSIG Timer
16 0 2 0 0.00% 0.00% 0.00% 0 IPC DL Deamon
17 29587676 56708694 521 0.49% 0.06% 0.01% 0 TCP Timer
18 588 111 5297 0.00% 0.00% 0.00% 0 TCP Protocols
19 21464691483155729759 680 2.53% 47.87% 53.59% 0 EMANATE SUBAGENT
20 0 1 0 0.00% 0.00% 0.00% 0 Critical Bkgnd
21 124 5 24800 0.00% 0.00% 0.00% 0 Net Background
PID Runtime(ms) Invoked uSecs 5Sec 1Min 5Min TTY Process
22 1808 5649 320 0.00% 0.00% 0.00% 0 Logger
23 7228612 37594502 192 0.00% 0.01% 0.00% 0 TTY Background
24 2915960 37594524 77 0.00% 0.00% 0.00% 0 Per-Second Jobs
25 30623596 37594523 814 0.08% 0.09% 0.07% 0 Net Periodic
26 0 2 0 0.00% 0.00% 0.00% 0 Net Input
27 490804 7535903 65 0.00% 0.01% 0.00% 0 Compute load avgs
28 20141220 627869 32078 0.00% 0.07% 0.05% 0 Per-minute Jobs
29 28552636801127731705 2531 6.04% 6.96% 7.50% 0 LANE Client
30 4192 1838 2280 0.00% 0.00% 0.00% 0 LES/BUS input
31 46719380 362850808 128 0.08% 0.19% 0.15% 0 Cat5K ATM LED
32 1784760 947798 1883 0.00% 0.00% 0.00% 0 SSCOP Output
33 1964380 4710059 417 0.00% 0.00% 0.00% 0 SSCOP Timer
34 30718244 861555 35654 0.00% 0.17% 0.09% 0 ATMSIG Input
35 1754336 709415 2472 0.00% 0.00% 0.00% 0 ATMSIG Output
36 80 98 816 0.00% 0.00% 0.00% 0 LECS Finder
37 8396 2939 2856 0.00% 10.15% 2.67% 1 Virtual Exec
Could the above causes the problem below ?
We have a problem most days where the terminal service client and pcanywhere were disconnected from the hosts. We also have a Microsoft 2000 server running roaming profile. Every so often, users connected (when this happen not all users connected to the server have the problem) have a message saying they were disconnected from the server and had to resynchronize the files on the server to be able to access the server again. On the window explorer, you will see red crosses against the drives connected to the server. Once the files were synchronized, the users will be able to access the server again.
Any help is much appreciated.
TIA.
PF
04-18-2003 09:09 AM
Hi,
I feel the output of "Show proc cpu " has got nothing to do with the problem of
the requiered time and again re-synchronization. There must be some other
reason for that problem.
04-29-2003 06:31 AM
I copied this output from the output interpreter on Cisco'c website:
==========================================================================
SHOW PROCESS CPU NOTIFICATIONS (if any)
==========================================================================
INFO: Total CPU Utilization is comprised of process and interrupt percentages.
Total CPU Utilization: 94%
Process Utilization: 94%
Interrupt Utilization: 0%
These values are found on the first line of the output:
CPU utilization for five seconds: x%/y%; one minute: a%; five minutes: b%
Total CPU Utilization: x%
Process Utilization: (x - y)%
Interrupt Utilization: y%
Process Utilization is the difference between the Total and Interrupt (x and y).
The one and five minute utilizations are exponentially decayed averages (rather
than an arithmetic average), therefore recent values have more influence on the
calculated average.
ERROR: Total CPU Utilization is at 94% for the past 5 seconds, which is very high (>90%).
This can cause the following symptoms:
- Input queue drops
- Slow performance
- Slow response in Telnet or unable to Telnet to the router
- Slow response on the console
- Slow or no response to ping
- Router doesn't send routing updates
The following processes are causing excessive CPU usage:
PID CPU Time Process
19 47.87% EMANATE SUBAGENT
37 10.15% Virtual Exec
INFO: The virtual EXEC process handles virtual type terminal (vty) lines,
such as Telnet sessions on the router. Issuing commands that generate long
output (such as 'show tech-support'), or if the debug output has been redirected
to the vty (using the terminal monitor or the no logging console command), the
amount of CPU resources used by the virtual EXEC process increases.
TRY THIS: If there is no indication of any problem in logged messages, then the
problem could possibly be caused by a bug in the IOS. Using the Bug Navigator,
run a search for the specified process to see if any bugs have been reported.
If this is not caused by a bug, this device may be overloaded. Investigate
upgrading this device or moving some CPU intensive tasks to a second device.
INFO: If you need help from a Customer Support Engineer in the Cisco TAC,
capture the 'show tech-support' command output (from enable mode) before
contacting Cisco TAC. Also, if the high CPU utilization is caused by a process,
please capture the 'show stacks {pid}' command output (where pid is the process
ID of the process causing the high CPU utilization). If the problem is caused by
a bug in IOS, please relay the bug ID to the Cisco Customer Support Engineer
handling the case.
REFERENCE: For troubleshooting information, please visit Troubleshooting High CPU Utilization on Cisco Routers
WARNING: 77.04% of CPU Utilization is from process switching.
Process switching is the most basic way of handling a packet.
TRY THIS: Consider another form of switching:
- Fast switching, using the 'ip route-cache' command
- Same-interface switching, using the 'ip route-cache same-interface' command
- Autonomous switching (7000 platforms only), using the 'ip route-cache cbus'
command
- Silicon switching (7000 platforms with an SSP installed only), using the
'ip route-cache sse' command
- Distributed switching (VIP-capable platforms only), using the 'ip route-cache
distributed' command
- Optimum switching (high-end routers only), using the 'ip route-cache optimum'
command
- Netflow switching, using the 'ip route-cache flow' command
- Cisco Express Forwarding (CEF), using the 'ip cef' command
- Distributed CEF, using the 'ip cef distributed' command
NOTE: When TCP Intercept is enabled process switching is required for all
packets during session set-up.
REFERENCE: For more information see Performance Tuning Basics
You have to be registered on their website to use it, please let me know whether this helps?
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: