15% lost packets pinging pfSense

CurtisThe

Hello,

I have an issue that I'm not sure how to proceed on debugging. When I ping any client in the same subnet as my PC, I get 100% successful packets. However, when I ping my pfSense gateway, I get only about 85% successful packets, with the rest of them being Request timed out. I feel this significantly while gaming in the form of latency spikes, but it's only happening on this one PC.

My Setup

pfSense connected to ISP's router which is in bridge mode
My PC dual boots with Ubuntu which is having issues getting an IP address via DHCP - the 2 ethernet cards I'm trying out are both failing to configure properly despite having drivers that should work.
Unifi switch connecting pfSense to my Windows 10 PC and other devices
I have a few VLANs configured as outlined here: https://community.ui.com/questions/Client-Devices-Inconsistent-VLAN-Network/c727073e-c564-46e5-8d8b-d330ed6205f3 (I'm also having issues with my VLANS)

Summary Issues

I'm hoping someone can look at this and with their expertise point me in the right direction for investigation:

~15% of packets dropped when pinging pfsense interface gateway from my PC; however pinging any other host in the subnet is working 100%
Issues getting an IP address on the same PC's Ubuntu dualbooted partition
VLANs & subnets not what is assigned in Unifi

Maybe there is a smell here that points to something obvious I'm not seeing.

What I've tried

I've used the packet capture tool in pfSense to see the ICMP traffic of the pings, but I'm new to this and not sure exactly what I'm looking for
I've tried rebooting/switching between the 2 network cards on my PC.

Thanks in advance!

AndyRH

First steps:
Replace Ethernet cable for pfSense
Move pfSense to different switch port
If possible move pfSense LAN port to different port on pfSense.

After each step repeat the ping test.

These 3 steps, done 1 at a time, will either eliminate a HW problem or show you where it is.

CurtisThe

@AndyRH
Thanks for the advice! One thing I neglected to mention is that other machines in my network are able to ping pfSense without issue, though on a different VLAN/Subnet.

Results of your recommendations:

1. Replace Ethernet cable for pfSense

I noticed that when my TrueNAS box was down after replacing this cable, I had no issues. But then my TrueNAS box came back online and I have issues again. I'll investigate why that is. What is the best path forward for investigating why one machine hosting the following services would interfere with my PC's connection?

Unifi Controller
Syncthing
Plex

Is there a tool within pfSense that will help me with this? For now I'll just try taking these TrueNAS services down one at a time.

CurtisThe

@CurtisThe
Update: It seems it's none of the services running on TrueNAS but the TrueNAS server itself causing the dropped packets... how do I investigate what's specifically the cause of one machine causing dropped packets on another machine in the network?

AndyRH

@CurtisThe You can follow similar steps with the TrueNAS.
3 likely causes:

bad port
bad cable
bad NIC driver

If your switch has error counters that show the type of error that may help narrow it down.

stephenw10

Could also be a conflict of some sort if they are in the same subnet.

If it's just that one machine I would be looking at that. Especially if you are already seeing other network issues with it.

If it's dual booting do you still see loss in the other OS? Windows I assume?

CurtisThe

@stephenw10
I just got an email saying that my kea-dhcp server was restarted.... I switched back to the ISC server a few days ago via the System->Advanced->Networking UI. I guess I didn't properly terminate the Kea DHCP server before switching back to ISC? Do I need to switch back to Kea and then disable the DHCP server on each interface manually before switching back to ISC again?

As for the dual booting; I'm on Windows now and the Ubuntu partition is not able to even get an IP address.

stephenw10

No you shouldn't need to do anything to switch between kea and ISC dhcp servers.

Did you enable the service watchdog for it? That could cause problems if it somehow started both services. Or tried to.

CurtisThe

@stephenw10
Yes, the watchdog was monitoring the kea server, thank you. I disabled it in watchdog, should I do anything else to make sure it's not running?

stephenw10

You could check the output of: ps -auxwwd | grep kea

That will tell you if it's running but it shouldn't be. Nothing beyond setting dhcpd back to ISC should be required.

CurtisThe

@stephenw10

root 13579 0.0 0.0 13684 3100 - S 17:54 0:00.00 | | `-- sh -c ps -auxwwd | grep kea 2>&1

root 13954 0.0 0.0 13120 2544 - S 17:54 0:00.00 | | `-- grep kea

^ This is the result

stephenw10

Yup so that's only the grep command you're running. Kea is not running.