CPU utilization of SRX branch devices is high

[shingane]

Dear All

I am facing  CPU untilization problem in SRX 220.  I am not able take web access and my cli is also very slow. i am not aware about the flowd_octeon_hm / rtlogd process and what it do. On the J-Web Dashboard tab, CPU utilization shows over 90%.for referance find the attach copy. 

How to kill this process.  (need the command)  

Router# run show system processes extensive 
last pid: 49580; load averages: 1.00, 1.02, 1.00 up 62+01:41:12 14:50:35
133 processes: 18 running, 104 sleeping, 11 waiting

Mem: 163M Active, 134M Inact, 514M Wired, 127M Cache, 112M Buf, 31M Free
Swap:


PID USERNAME THR PRI NICE SIZE RES STATE C TIME WCPU COMMAND
1153 root 5 76 0 498M 48536K select 0 1629.2 96.92% flowd_octeon_hm
1174 root 1 139 0 11884K 5384K RUN 0 353.8H 84.96% rtlogd
22 root 1 171 52 0K 16K RUN 0 937.7H 0.00% idle: cpu0
24 root 1 -20 -139 0K 16K RUN 0 975:17 0.00% swi7: clock
5 root 1 -84 0 0K 16K rtfifo 0 250:24 0.00% rtfifo_kern_recv
1180 root 1 76 0 10628K 4608K select 0 244:50 0.00% license-check
1142 root 1 76 0 17432K 6904K select 0 172:37 0.00% l2ald
1148 root 1 76 0 19264K 8504K select 0 161:47 0.00% kmd
23 root 1 -40 -159 0K 16K WAIT 0 154:10 0.00% swi2: net
1139 root 1 76 0 17796K 10576K select 0 115:41 0.00% snmpd
1149 root 1 76 0 9012K 3348K select 0 105:33 0.00% ppmd
1146 root 1 76 0 11580K 4952K select 0 92:40 0.00% rmopd

Thanks

Regards

Hemant Shingane

[coolblue]

Can you look in the rtlogd log file, if there is something wrong?

[Flannigan]

That is normal cpu usage for flowd but your rtlogd is abnormally high.  Can you post the configurations for both the system and security logging facilities?  And if you're running an earlier build of JUNOS I would strongly suggest that you investigate 10.4R7.5.

Thanks.

[Duga]

Hi,

Do you have some traceoptions or packet capture activated ?

If you are not debugging, deactivate all logging configuration.