pa performance analyze

[minow]

hey

i am trying to analyse if the PA is under load regarding to the PA specs,

the customer is having sometimes disconnects on the network, i can see that the CPU have peaks sometimes but mainly is OK

netcom@PA-IL-ACTIVE(active)> show running resource-monitor hour

Resource monitoring sampling data (per hour):

CPU load (%) during last 24 hours:

core    0       1       2       3

     avg max avg max avg max avg max

       0   0   8  24  29  68  29  68

       0   0   8  23  28  75  28  75

       0   0   7  23  24  61  24  61

       0   0   6  19  20  50  20  49

       0   0   9  28  29  76  29  77

       0   0   6  21  21  64  21  64

       0   0   7 100  20 100  20 100

       0   0   5  15  16  44  17  44

       0   0   6  17  19  59  19  60

       0   0   7  26  21  73  21  73

       0   0   5  17  18  47  18  46

       0   0   5  12  17  40  18  40

       0   0   7  16  22  55  22  55

       0   0   7  24  21  68  22  69

       0   0   7  31  23  80  24  79

       0   0   9  26  31  83  31  83

       0   0  11  76  36 100  36 100

       0   0  10  61  35  94  35  94

       0   0  15 100  50 100  51 100

       0   0  13  25  45  87  45  87

       0   0  11  27  41  83  41  83

       0   0  13  33  42  84  42  83

       0   0  13  45  46  99  46  99

       0   0  12  33  41  83  41  83

The output of this command

show system statistics session

Is:

Device is up : 71 days 23 hours 16 mins 49 sec

Packet rate : 7464/s

Throughput : 27602 Kbps

Total active sessions : 11693

Active TCP sessions : 7649

Active UDP sessions : 3966

Active ICMP sessions : 42

Does this mean that the PA whole throughput is: 27602 Kbps (27 Mbps)

So the PA-2050 max threat prevention throughput is 500Mbps so there is no way the PA is under load?

Is this the same parameter the one from the command and the one from the datasheet?

do you have some scripts that i can run in real time that will show me the relevant output to show what is the utilization of the PA ?

thanks

dor

[stewart · Accepted solution]

You could run the show system resources follow command.

Look at the load average and the wa%

top - 16:57:15 up 14 days, 23:37,  1 user, load average: 1.00, 1.18, 1.38

Tasks:  95 total,  3 running,  92 sleeping,  0 stopped,  0 zombie

Cpu(s): 50.2%us,  0.0%sy,  0.0%ni, 49.8%id, 0.0%wa,  0.0%hi,  0.0%si,  0.0%st

Mem:  2575620k total,  2389496k used,  186124k free,    44344k buffers

Swap:    7992k total,    7992k used,        0k free,  939372k cached

You can also check the CPU% (I ran this on a PA200 so it will always show 100%)

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND                                          

2317 root      20  0 48536  23m 2676 R  100  0.9  21575:07 pan_task                                        

2510 root      20  0  180m  82m  68m S    0  3.3  6:16.29 useridd                                          

2525 root      20  0  759m 572m  9m S    0 22.7 131:46.22 mgmtsrvr

You can run the show session meter command which will show if any session throttling occurred due to lack of resources.