Forum Discussion

rafaelbn_176840's avatar
rafaelbn_176840
Icon for Altocumulus rankAltocumulus
May 25, 2018

Viprion 2400 with 2xB2150. 20-30 active connections. CPU0 at 100%

Hello Devs! Hows everybody?

We have a chassi running v12.1.2 (HF1). The entire system shows an average of 20-30 active connections during the day. But CPU-0 of both 2150 blades inserted on this system are very close to 100%.

show /sys tmm-info confirms that both CPU-0 are under heavy load.

---------------------------
Sys::TMM: 1.0           
---------------------------
Global
  TMM Process Id      14649
  Running TMM Id          4
  TMM Count               1
  CPU Id                  0

Memory (bytes)     
  Available           13.7G
  Used               908.2M


CPU Usage Ratio (%)
  Last 5 Seconds      86.92
  Last 1 Minute       84.32
  Last 5 Minutes      84.04


  ---------------------------
Sys::TMM: 2.0           
---------------------------
Global
  TMM Process Id      14839
  Running TMM Id          0
  TMM Count               1
  CPU Id                  0

Memory (bytes)     
  Available           13.7G
  Used               940.6M


CPU Usage Ratio (%)
  Last 5 Seconds      97.10
  Last 1 Minute       93.90
  Last 5 Minutes      93.93

All other CPUs are below 10%. Is this normal? Could this be a bad load distribution (DAG problem)? This system is a VPN box.

Thanks! Rafael

  • That certainly does not look correct.

     

    You would be better to raise a support ticket, so that the QKView can be checked by a support engineer to see what is using the CPU (either traffic through a virtual, an irule, or something else).