The system is experiencing high %sys CPU usage, primarily due to IBM Tivoli Monitoring processes (klzagent and k08agent) handling a large volume of incoming TCP packet processing
Issue
- The system is showing relatively high %sys CPU usage.
- Perf data captured during the issue window implies that IBM Tivoli Monitoring processes (klzagent and k08agent) together, are consuming ~44.20+% of total sampled CPU cycles:
- klzagent: ~24.37% Total
- k08agent: ~19.82% Total
Samples: 160K of event 'cpu-clock:pppH', Event count (approx.)
Overhead Command Shared Object
- 46.47% swapper [kernel.kallsyms] ◆
+ 42.91% native_safe_halt ▒
+ 2.38% finish_task_switch ▒
- 17.83% klzagent [kernel.kallsyms] ▒
- 4.50% _raw_spin_unlock_irqrestore ▒
- 3.88% __wake_up_common_lock ▒
- 3.86% sock_def_readable ▒
- 3.86% tcp_rcv_established ▒
tcp_v4_do_rcv ▒
tcp_v4_rcv ▒
ip_protocol_deliver_rcu ▒
ip_local_deliver_finish ▒
ip_local_deliver ▒
ip_rcv ▒
__netif_receive_skb_core ▒
process_backlog ▒
__napi_poll ▒
net_rx_action ▒
__softirqentry_text_start ▒
do_softirq_own_stack ▒
do_softirq.part.16 ▒
__local_bh_enable_ip ▒
ip_finish_output2 ▒
ip_output ▒
__ip_queue_xmit ▒
__tcp_transmit_skb ▒
tcp_write_xmit ▒
__tcp_push_pending_frames ▒
tcp_sendmsg_locked ▒
tcp_sendmsg ▒
__sock_sendmsg ▒
__sys_sendto ▒
__x64_sys_sendto ▒
do_syscall_64 ▒
entry_SYSCALL_64_after_hwframe ▒
0x153ef7f1bb66 ▒
+ 1.35% do_syscall_64 ▒
+ 0.75% finish_task_switch ▒
- 14.39% k08agent [kernel.kallsyms] ▒
- 3.45% _raw_spin_unlock_irqrestore ▒
- 2.79% __wake_up_common_lock ▒
+ 2.78% sock_def_readable ▒
+ 1.20% do_syscall_64 ▒
+ 0.64% finish_task_switch ▒
+ 3.24% klzagent libpthread-2.28.so ▒
+ 2.65% k08agent libpthread-2.28.so ▒
...
Environment
- Red Hat Enterprise Linux 8.10.z - kernel-4.18.0-553.46.1.el8_10.el8
- IBM Tiboli Monitoring (klzagent / k08agent)
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.