PfSense crashing on certain conditions
I was hoping someone might be able to help me identify an issue with my pfSense configuration - I seem to be having some sort of failure under certain conditions. When a failure occurs, my devices seem to lost the ability to obtain a DHCP address. If I connect to the console physically, I can still ping the interface addresses and ping the outside world.
I'm running pfSense 2.3.4-RELEASE amd64 full on a custom built server.
Intel Pentium G620
16GB ECC RAM
HP NC360T NIC
I have a managed switch with the following VLANS configured.
10 - WEB
20 - MGT
30 - DMZ
40 - WIFI
50 - WIFIGUEST
60 - VPN
Port 1 on the switch is an access port for LAN.
Port 2 on the switch is a trunk port for pfSense which is allowed on VLANS 10,20,30,40,50,60.
Port 10 on the switch is my Wifi AP which is allowed VLAN 40.
NIC1 is the onboard NIC1 and is connected to my Fibre Modem.
NIC2 is the onboard NIC2 and is connected to Port 1 on the switch (LAN)
NIC3 is the add-on NIC port 1 and is connected to the trunk port on the switch
I have an OpenVPN client also running. I have no DNS server specified under General > DNS Servers and use the DNS Resolver on with the outgoing interface of WAN_VPN (the interface assigned to ovpnc1).
This seems to work until I push a large volume of traffic. At which point the server becomes unresponsive from a client.
I'm not sure if it is related, but I also noticed if I stop the OpenVPN client, the CPU pegs at 100%. It seems to be the line below that pegs the CPU.
75924 root 94 0 14516K 2316K CPU0 0 2:08 71.68% /usr/sbin/syslogd -s -c -c -l /var/dhcpd/v
I caught that at 71% before it spiked to 100%. It sits there like that until I restart the OpenVPN client.
I have swapped the NIC out (same type) and re-installed. I have also changed the hard disk and re-installed. I have also updated to 2.3.4_1 to no avail. I have since reverted back to 2.3.4.
Any help would be much appreciated. Happy to provide any more detail if it helps. I've been pulling my hair out for days!