How to determine whether CPU usages of the system and processes are high for S series switches

10

For S series switches (except the S1700), the system CPU usage is within a normal range when the following conditions are met:
- The CPU usage does not exceed 80% when the system runs for a long period.
- The CPU usage does not exceed 95% in a short period and does not keep increasing.
- No high CPU usage alarm is generated.
In the following situations, the system may generate an alarm when CPU usage becomes high instantly and then clear this alarm rapidly: Cards just start. Information about all optical modules is queried at a time. Traffic volume increases instantly. This situation is a normal situation and does not affect the device operation.

CPU usage of a CPU task process may become low or high, depending on the service volume and processing time. CPU usage is within a normal range as long as the system CPU usage does not exceed 80% and no high CPU usage alarm is generated.

For possible causes and common troubleshooting procedures for high CPU usage, see the following:
- For the S12700, see "High CPU Usage > Common Information" in S12700 Troubleshooting - Preliminary.
- For the S7700 and S9700, see "High CPU Usage > Common Information" in S7700&S9700 Troubleshooting - Preliminary.
- For the S1720, S2700, S3700, S5700, and S6700, see "High CPU Usage > Common Information" in S1720&S2700&S3700&S5700&S6700 Troubleshooting - Preliminary.
- For the S9300, see "High CPU Usage > Common Information" in S9300 Troubleshooting - Preliminary.
- For the S600-E, see "High CPU Usage > Common Information" in S600-E Troubleshooting - Preliminary.

Other related questions:
How to determine whether the memory usage of an S series switch is high
For S series switches (except the S1700), the memory usage of a switch is within the normal range when the following conditions are met: - The memory usage does not exceed 80%. - The memory usage does not keep increasing or fluctuate. - No high memory usage alarm is generated.

Meaning of VRPI process in CPU usage on S series switches
On S series switches, the VRPI process loads configurations. When a switch starts, you need to load the configuration, which causes high CPU usage of this process. After the configuration is loaded, the CPU usage is restored to normal state.

CE series switches have high CPU usage, then how to quickly locate the cause and collect CPU information
If the CPU usage of the device is high, perform the following steps to locate and collect information, and then send the information to Huawei technical support personnel. 1. Run the display cpu command to view the services with the high CPU usage. display cpu CPU utilization statistics at 2016-02-02 02:07:22 366 ms System CPU Using Percentage : 6% CPU utilization for five seconds: 6%, one minute: 6%, five minutes: 6%. Max CPU Usage : 38% Max CPU Usage Stat. Time : 2016-02-01 12:29:26 821 ms State: Non-overload Overload threshold: 90%, Overload clear threshold: 75%, Duration: 480s --------------------------- ServiceName UseRate --------------------------- SYSTEM 6% AAA 0% ARP 0% CMF 0% ...... 2. Run the display system service service-name command repeatedly in the diagnostic view to check the service components with the high CPU usage. system-view [~HUAWEI] diagnose [~HUAWEI-diagnose] display system service SYSTEM Service : SYSTEM -------------------------------------------------------------------------------- Component CID CGID Process HaState CpuUsage MemUsage -------------------------------------------------------------------------------- DFS 0x81DE271D 0x1DE271B 10003 PRIMARY 0% 4006312 TELC 0x8091271F 0x91271D 10001 PRIMARY 0% 19000 SSHC 0x80922720 0x92271E 10001 PRIMARY 0% 41208 SSHS 0x80932723 0x932721 10001 PRIMARY 0% 48844 3. Run the display system thread process process-id command repeatedly in the diagnostic view to check the processes with the high CPU usage. system-view [~HUAWEI] diagnose [~HUAWEI-diagnose] display system thread process 10003 Info: This operation needs several seconds. Info: Operating, please wait for a moment......done. ------------------------------------------------------------------------------- Process ID Thread ID Thread Type BindComp BindCpu BindFlag Usage ------------------------------------------------------------------------------- 10003 1210525456 main thread Bind all Free 0% 10003 1227637888 DefSch0200 Free all Free 0% 10003 1227506816 IPC0000 Free all Free 0% 10003 1227768960 DefSch0300 Free all Free 0% 10003 1237476480 DefSch0100 Free all Free 0% 10003 1237607552 DefSch0101 Free all Free 0% 10003 1226847360 DMS_PIPE_RECV_TASK Free all Free 0% 10003 1218028672 TICK Free all Free 0% 10003 1226716288 DMS_TIPC_SEND Free all Free 0% 10003 1217766528 BOX_Out Free all Free 0% 10003 1217897600 VCLK Free 0 Free 0% 10003 1237738624 CliGetThreadCpu Free all Free 0% ------------------------------------------------------------------------------- Total = 12 4. Run the display system thread callstack process process-id thread-id command in the diagnostic view to check process call stack information. system-view [~HUAWEI] diagnose [~HUAWEI-diagnose] display system thread callstack process 10003 1210525456 Thread 1210525456 (Thread MainThread): #00 libc.so.6(epoll_wait) #01 location(Frame_FdMainThread) #02 location(Frame_Main) #03 location(main) #04 libc.so.6() #05 libc.so.6()

View CPU usage and memory usage on s series switch
For S series switches except S1700 switches, you can run commands to display CPU and memory usage. - display cpu-usage [ slave | slot slot-id ]: displays CPU usage statistics. - display cpu-usage configuration [ slave | slot slot-id ]: displays CPU usage configuration. - display memory-usage [ slave | slot slot-id ]: displays memory usage statistics. - display memory-usage threshold [ slot slot-id ]: displays the alarm threshold for memory usage. For S series switches and E series switches except S1700 switches, you can check CPU and memory usage through the related MIB OIDs. - hwEntityCpuUsage 1.3.6.1.4.1.2011.5.25.31.1.1.1.1.5: provides CPU usage. - hwEntityCpuUsageThreshold 1.3.6.1.4.1.2011.5.25.31.1.1.1.1.6: provides the alarm threshold for CPU usage. - hwEntityMemUsage 1.3.6.1.4.1.2011.5.25.31.1.1.1.1.7: provides memory usage. - hwEntityMemUsageThreshold 1.3.6.1.4.1.2011.5.25.31.1.1.1.1.8: provides the alarm threshold for memory usage.

If you have more questions, you can seek help from following ways:
To iKnow To Live Chat
Scroll to top