Stability issues with pfSense and UniFi devices
-
I originally posted my problem on the UniFi forums thinking that my issue was on the AP side, but I have come to believe that it's a misconfiguration with pfSense. Original post: https://community.ui.com/questions/Network-stability-issues-with-UniFi-Switch-and-AP/95abb5d5-690a-4d2d-a06e-895e0b82cbd2
Problem: UniFi devices and pfSense currently have a lot of stability issues with the UniFi devices. I can see from the Interfaces, that the LAN port that the UniFi devices are plugged into turn on/off constantly. The error I receive are:
<13>Apr 1 16:45:51 check_reload_status: Linkup starting e6000sw0port4 192.168.1.1 01/04 16:45:50.641 <13>Apr 1 16:45:52 check_reload_status: Reloading filter 192.168.1.1 01/04 16:45:51.684 <29>Apr 1 16:45:53 xinetd[48698]: Starting reconfiguration 192.168.1.1 01/04 16:45:51.895 <29>Apr 1 16:45:53 xinetd[48698]: Swapping defaults 192.168.1.1 01/04 16:45:51.897 <29>Apr 1 16:45:53 xinetd[48698]: readjusting service 19000-tcp 192.168.1.1 01/04 16:45:51.899 <29>Apr 1 16:45:53 xinetd[48698]: readjusting service 19001-tcp 192.168.1.1 01/04 16:45:51.901 <29>Apr 1 16:45:53 xinetd[48698]: Reconfigured: new=0 old=2 dropped=0 (services) 192.168.1.1 01/04 16:45:51.903 <29>Apr 1 16:45:54 xinetd[48698]: Starting reconfiguration 192.168.1.1 01/04 16:45:52.930 <29>Apr 1 16:45:54 xinetd[48698]: Swapping defaults 192.168.1.1 01/04 16:45:52.933 <29>Apr 1 16:45:54 xinetd[48698]: readjusting service 19000-tcp 192.168.1.1 01/04 16:45:52.935 <29>Apr 1 16:45:54 xinetd[48698]: readjusting service 19001-tcp 192.168.1.1 01/04 16:45:52.936 <29>Apr 1 16:45:54 xinetd[48698]: Reconfigured: new=0 old=2 dropped=0 (services) 192.168.1.1 01/04 16:45:52.938 <13>Apr 1 16:45:54 check_reload_status: Linkup starting e6000sw0port4 192.168.1.1 01/04 16:45:53.721 <5>Apr 1 16:45:54 kernel: e6000sw0port4: link state changed to UP 192.168.1.1 01/04 16:45:53.723 <13>Apr 1 16:45:55 check_reload_status: Linkup starting e6000sw0port4 192.168.1.1 01/04 16:45:54.723 <5>Apr 1 16:45:55 kernel: e6000sw0port4: link state changed to DOWN 192.168.1.1 01/04 16:45:54.725 <13>Apr 1 16:45:55 check_reload_status: Reloading filter 192.168.1.1 01/04 16:45:54.731 <13>Apr 1 16:45:56 check_reload_status: Reloading filter 192.168.1.1 01/04 16:45:55.738 <29>Apr 1 16:45:57 xinetd[48698]: Starting reconfiguration 192.168.1.1 01/04 16:45:55.996 <29>Apr 1 16:45:57 xinetd[48698]: Swapping defaults 192.168.1.1 01/04 16:45:56.000 <29>Apr 1 16:45:57 xinetd[48698]: readjusting service 19000-tcp 192.168.1.1 01/04 16:45:56.007 <29>Apr 1 16:45:57 xinetd[48698]: readjusting service 19001-tcp 192.168.1.1 01/04 16:45:56.013 <29>Apr 1 16:45:57 xinetd[48698]: Reconfigured: new=0 old=2 dropped=0 (services) 192.168.1.1 01/04 16:45:56.017 <29>Apr 1 16:45:58 xinetd[48698]: Starting reconfiguration 192.168.1.1 01/04 16:45:56.990 <29>Apr 1 16:45:58 xinetd[48698]: Swapping defaults 192.168.1.1 01/04 16:45:56.993 <29>Apr 1 16:45:58 xinetd[48698]: readjusting service 19000-tcp 192.168.1.1 01/04 16:45:56.994 <29>Apr 1 16:45:58 xinetd[48698]: readjusting service 19001-tcp 192.168.1.1 01/04 16:45:56.996 <29>Apr 1 16:45:58 xinetd[48698]: Reconfigured: new=0 old=2 dropped=0 (services) 192.168.1.1 01/04 16:45:56.997 <13>Apr 1 16:45:58 check_reload_status: Linkup starting e6000sw0port4 192.168.1.1 01/04 16:45:57.766 <5>Apr 1 16:45:58 kernel: e6000sw0port4: link state changed to UP 192.168.1.1 01/04 16:45:57.768
UniFi devices are reset to default. I plugged the UniFi devices into a DD-WRT network and do not experience this issue. Is there anyway I can post my pfSense configuration to spot obvious issues? Or any ideas on how to begin troubleshooting? I have no VLANs. Network at the moment is Netgate SG-3100 w/ 2.4.5-RELEASE plugged into a UniFi UAP-AC-LITE for simplicity. Ethernet cables have been tested and are good.
-
You should probably open a support ticket. You are plugging the APs directly into the switch ports of the 3100? This should be in the official hardware section, not the general questions, if there's a mod about.
-
That's correct, I am plugging the AP directly into the switch port of the 3100. Sorry for wrong section, if a mod would move the thread I'd appreciate it. I will also open a support ticket.
-
I just tried restoring previous backups from January and December and noticed that this behavior doesn't go away.
-
The SG-3100 became unresponsive and I ended up doing a power cycle. Additionally, I switched the ports around. It seems to be stable for now, I am not seeing the errors in syslog any longer. I'll wait a few days to see if this addresses the problem.
Does this sound like a hardware issue? I'm still within my 12 month warranty -- need to understand if need to escalate to RMA.
-
I don't have much experience with the Arm units, but if you open a support ticket they can probably give you some suggestions to test the hardware.
-
So I was told I only have Community Support. I've closed the ticket and moved towards this forum. Additionally, I've requested permission for RMA.
I attempted to recreate the behavior this morning. Switching to Port 4 caused the up/down behavior again. Switching to Port 3 made the issue go away. Maybe a bad port/hardware issue?
-
It sounds like hardware to me, but I don't know much about how the switch port work on those. If you have the ability to backup your config, reset to defaults, and see if the port goes up and down on a default config, that might help to see if it is hardware or configuration.
-
Yep, I tried that as well yesterday. I restored previous known good configurations from December 2019 and January 2020 and behavior was still present.
-
Yes, but that's still your config. It's easier to make a case that there is a hardware issue if you can replicate it on the default config.
-
@dchiang1987 said in Stability issues with pfSense and UniFi devices:
e6000sw0port4: link state changed to DOWN
.....
e6000sw0port4: link state changed to UPThis breaks everything.
Ripping out cables is considered equal. -
I will try this evening after the workday and report back.