SRX Services Gateway
SRX Services Gateway

process extensive

‎09-05-2019 11:41 PM

Hi all,

from https://www.juniper.net/documentation/en_US/junos/topics/reference/command-summary/show-system-proce...

I couldn't get what is difference between "SIZE" and "RES"  from the junos-cli?

 

CAn anyone explain this pls...

And also what thing really should be concerned on output of the the following junos-cli?

 

>show system processes extensive | except 0.00
last pid: 50884; load averages: 0.27, 0.20, 0.17 up 156+21:17:26 16:33:10
152 processes: 4 running, 128 sleeping, 20 waiting

Mem: 424M Active, 54M Inact, 98M Wired, 135M Cache, 110M Buf, 270M Free
Swap:


PID USERNAME THR PRI NICE SIZE RES STATE TIME WCPU COMMAND
11 root 1 171 52 0K 16K RUN 3084.7 86.67% idle
1306 root 2 44 -52 72068K 18500K select 213.1H 4.05% sfid
1910 root 1 8 0 86840K 43384K nanslp 187.4H 2.73% pfem
1836 root 1 106 0 28840K 13892K RUN 45.6H 0.68% mib2d
1305 root 1 4 0 67844K 11616K kqread 79.6H 0.05% chassism
40 root 1 76 0 0K 16K sleep 30.7H 0.05% netdaemon

Thanks,

A

2 REPLIES 2
SRX Services Gateway

Re: process extensive

‎09-06-2019 12:54 AM

When dealing with CPU issues, we need to correlate information from different sources. Do you see high CPU utilization when running "show chassis routing-engine"? Do you see any weird logs when running "show log messages"?

 

The processes showing CPU utilization are:

 

sfid

 

The sfid process is the software forwarding infrastructure daemon. This process handles all traffic that the switch cannot handle in hardware alone. Common examples of such traffic are:

  • GRE encapsulation
  • ARP resolution
  • management traffic

High CPU usage for the sfid process might indicate a problem, and certainly could impact performance when you have legitimate traffic that ought to be software switched. I would suggest checking what traffic causes this problem in the first place.

Ref: https://networkengineering.stackexchange.com/questions/51294/whats-the-purpose-of-junos-sfid-process

 

  • Are you performing intensive SNMP polling or experiencing an SSH attack maybe? Can you stop any of them for testing if the CPU utilization will go down?

 

pfem

Forwarding process: Defines how routing protocols operate on the switch. The overall performance of the switch is largely determined by the effectiveness of the forwarding process.

Ref: https://www.juniper.net/documentation/en_US/release-independent/junos/topics/concept/ex-series-softw...

 

  • Are you running any dynamic routing protocols?

 

mib2d

This process has to do with SNMP polling via the kernel. some SNMP MIB data are maintained by the kernel or Packet Forwarding Engine. For such data to be available for the network management system, the kernel has to provide the required information to the SNMP subagent in mib2d.

Ref: https://www.juniper.net/documentation/en_US/junos/topics/task/configuration/snmp-best-practices-moni...

 

  • Can you stop SNMP polling for testing?

 

Try the following shell command:

 

> start shell

% top -H

 

SRX Services Gateway

Re: process extensive

‎09-07-2019 10:10 PM

Hello 

 

RES refers to actual RAM usage by the process, while SIZE also includes files mapped by the process. So if large size files are mapped by the process SIZE would be much higher than RES.

 

I do not see anything of concern in the output you shared.

 

Regards,

 

Vikas