Unusual ping times to machine
-
I am seeing ping times jumping around and the machine is a P4 2ghz with 1gb ram. I am not seeing any processor load over 5 %. The nics are:
fxp (Intel built-in on the D845GB board)
rl (Realtek 8139 based card)
de (Digital 21140A Fast Ethernet)The Digital card is the outside interface and is having no problems. I have tried to switch the internal interface between the Intel and Realtek nics. The result was the same.
I am running 1.0.1-SNAPSHOT-02-27-2007 from Mon Mar 5 12:08:17 EST 2007
Reply from 10.10.10.3: bytes=32 time=40ms TTL=64
Reply from 10.10.10.3: bytes=32 time<1ms TTL=64
Reply from 10.10.10.3: bytes=32 time<1ms TTL=64
Reply from 10.10.10.3: bytes=32 time<1ms TTL=64
Reply from 10.10.10.3: bytes=32 time=72ms TTL=64
Reply from 10.10.10.3: bytes=32 time<1ms TTL=64
Reply from 10.10.10.3: bytes=32 time<1ms TTL=64
Reply from 10.10.10.3: bytes=32 time<1ms TTL=64
Reply from 10.10.10.3: bytes=32 time=163ms TTL=64
Reply from 10.10.10.3: bytes=32 time<1ms TTL=64
Reply from 10.10.10.3: bytes=32 time<1ms TTL=64
Reply from 10.10.10.3: bytes=32 time<1ms TTL=64
Reply from 10.10.10.3: bytes=32 time=250ms TTL=64
Reply from 10.10.10.3: bytes=32 time<1ms TTL=64
Reply from 10.10.10.3: bytes=32 time<1ms TTL=64
Reply from 10.10.10.3: bytes=32 time<1ms TTL=64
Reply from 10.10.10.3: bytes=32 time=317ms TTL=64
Reply from 10.10.10.3: bytes=32 time=139ms TTL=64
Reply from 10.10.10.3: bytes=32 time<1ms TTL=64 -
it might be that 10.10.10.3 is busy doing something else… i've seen this on computers infected with spyware, but it could be something totaly legitimate too.
-
If you are coming from lan, do you experience the same problems when pinging the wan IP of the pfSense? Also what infrastructure do you have in front of your wan? Is there the possibility that something is wrong there?
-
Sorry, 10.10.10.3 is the LAN interface of a pfSense machine. lol
Other machines on the LAN I can reach with no problems, also I am on the same subnet, so I am not hoping through any routers.
I guess you have to assume I am a moron, until I show you I ain't! ;)
-
Try with a crossovercable and a client directly connected to the LAN interface just to make sure it's not a switch or a cable acting up.
-
I am going to guess that it will do it then as well if I generate any traffic - as I have tried the following:
changed the cable 4 times (twice for each nic) - tried 4 different ports on 3 different switches (Cisco 2924 and 3500) and tried both the network cards.I have never seen anything like this before. Normally I would see packet loss, or errors on the nic and/or switchport; but I see neither. Or perhaps high processor load caused by some process on the machine, which I am also not seeing. It corresponds to outbound traffic on the firewall. Ie. when there is no traffic going outbound, then the ping is sub 1ms.
Roy
-
Just started turning things off…
traffic shaper was causing it!
-
Guess you don't have ICMP set for high priority. My pingties go up a bit too on extreme load but they won't freak out like what you showed us.
-
No this was not a mis-configuration issue. I am on the LAN interface, pinging the LAN interface, that traffic should never even enter a queue. I did have ICMP in the high queue, even if the code was trying to stick it there. But there was not a congestion issue that was causing the problem, it was any traffic going through the firewall!
-
if you shape wan - lan
then if you ping the lan of pfsense then it will enter en exit the shaper -
Is there now 2 threads on this topic!?
-
I'll lock this thread. Please continue posting to http://forum.pfsense.org/index.php/topic,3954.0.html as this one covers the same topic and is in the correct board. Thanks.