Hi!
I have two pfSense boxes connected to the same net segment where we have 4-10Mb/s (up to 5kps) stream of small udp multicast packets. pfSense1 is connected to this segment via VLAN (em3) - this pfSense is overloaded. pfSense2 is connected to this segment via LAN interface (em0) - this pfSense is ok. I am not very good at FreeBSD internals so my question is why do I have that drastic difference in performance. Basically all em3's load is caused by this udp stream because if it stops the load goes to normal 'almost 100% free'. Is VLAN tagging affecting performance that bad?

---- pfSense1 -----
last pid: 7245; load averages: 0.20, 0.28, 0.31 up 0+05:23:40 15:20:24
105 processes: 5 running, 81 sleeping, 19 waiting
CPU:  0.2% user,  0.0% nice,  6.5% system,  0.2% interrupt, 93.2% idle
Mem: 52M Active, 32M Inact, 58M Wired, 308K Cache, 24M Buf, 3365M Free
Swap: 8192M Total, 8192M Free

 PID USERNAME  THR PRI NICE   SIZE    RES STATE  C   TIME   WCPU COMMAND
13 root 1 171 ki31 0K 8K CPU1 1 315:09 100.00% idle: cpu1 11 root 1 171 ki31 0K 8K CPU3 3 312:52 100.00% idle: cpu3
  12 root        1 171 ki31     0K     8K CPU2   2 282:52 96.29% idle: cpu2
  14 root        1 171 ki31     0K     8K RUN    0 231:47 84.38% idle: cpu0
  34 root        1 -68    -     0K     8K -      0  90:24 16.26% em3 taskq
  31 root        1 -68    -     0K     8K -      2  36:51  7.28% em0 taskq
29 root 1 8 - 0K 8K - 3 9:19 1.37% thread taskq
  32 root        1 -68    -     0K     8K -      0   1:35  1.17% em1 taskq
37 root 1 -68 - 0K 8K WAIT 2 1:42 0.10% irq25: bge0
  18 root        1  44    -     0K     8K -      2   0:48  0.00% yarrow
16 root 1 -32 - 0K 8K WAIT 3 0:46 0.00% swi4: clock s
3633 root        1   4    0 46120K 20760K accept 1   0:27  0.00% php
  33 root        1 -68    -     0K     8K -      2   0:22  0.00% em2 taskq

----- pfSesnse2 -----
last pid: 84582; load averages: 0.15, 0.18, 0.16 up 1+22:51:13 15:20:24
92 processes:  5 running, 65 sleeping, 22 waiting
CPU states: 0.3% user, 0.0% nice, 0.3% system, 2.4% interrupt, 97.0% idle
Mem: 32M Active, 11M Inact, 39M Wired, 19M Buf, 3426M Free
Swap: 8192M Total, 8192M Free

 PID USERNAME  THR PRI NICE   SIZE    RES STATE  C   TIME   WCPU COMMAND
  10 root        1 171   52     0K     8K CPU3   0   0:00 99.02% idle: cpu3
  12 root        1 171   52     0K     8K CPU1   0   0:00 99.02% idle: cpu1
  13 root        1 171   52     0K     8K RUN    0  45.3H 93.36% idle: cpu0
  11 root        1 171   52     0K     8K CPU2   2  43.0H 86.28% idle: cpu2
  14 root        1 -44 -163     0K     8K WAIT   2  65:26  9.47% swi1: net
  29 root        1 -68 -187     0K     8K WAIT   2   7:20  0.68% irq65: em3
  25 root        1 -68 -187     0K     8K WAIT   0   5:14  0.05% irq39: em0
1144 root        1   4    0  6048K  5408K kqread 0   1:35  0.05% lighttpd
81785 root        1   4    0 15896K 13324K accept 0   0:04  0.05% php
15 root 1 -32 -151 0K 8K WAIT 0 4:58 0.00% swi4: clock s 33 root 1 -64 -183 0K 8K WAIT 0 2:54 0.00% irq19: uhci1 30 root 1 -64 -183 0K 8K WAIT 2 1:45 0.00% irq16: uhci0
  26 root        1 -68 -187     0K     8K WAIT   2   1:41  0.00% irq36: em1

Thank you!

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Commercial support available - https://portal.pfsense.org

Reply via email to