2.5.1 2.5.2 loses connection

[jerothle]

Help.
I had 2.5.2 installed, but went back to 2.5.1 to fix connection dropping.
I have been using cron to reboot every hour, but at times that is not enough to get the connection up again.

I have searched on the web and here, but I might not have the right search phrasing.

In Gateway Logs:
Aug 31 19:32:31 dpinger 29997 WAN_DHCP ##.##.##.#: sendto error: 64

Intel nics em0, em1, etc. and Supermicro onboard nics. The Intel card is a 4 port.
It seems if alot is being downloaded the connection dies. A reboot of Pfsense fixes it.
Comcast cable to
Motorola modem to
pfsense to
Netgear M4100-24G-POE+ ProSafe switch.
I have no rules set other then the default setup.

Sometimes I cant connect to pfsense, when the connection dies, other times I can but not out to the internet.
I tried pinging 8.8.8.8 from within pfsense and it works, but a computer connected to the switch that is connected to pfsense cant get to the outside.
Nothing has changed in the hardware, it has been used for several years.
I have 2 identical setups and the same thing seems to happen on both.
I had VLAN's setup on the switch and pfsense, but cleaned all that out (it all worked previously)
There is only 2 connections on pfsense right now: WAN and LAN.

I have had all connections on the Intel nic or some on Intel and some on the motherboard nic.... no difference.

Supermicro motherboard, Intel nic.
New DOCSIS 3.1 modem
The problem started with the previous DOCSIS 3 modem also.

[jerothle]

@jerothle

Aug 31 19:36:31 dpinger 28952
send_interval 500ms loss_interval 2000ms time_period 60000ms report_interval 0ms data_len 1 alert_interval 1000ms latency_alarm 500ms loss_alarm 20% dest_addr ##.##.##.1 bind_addr ##.##.##.49 identifier "WAN_DHCP "
Aug 31 19:36:32 dpinger 35290
send_interval 500ms loss_interval 2000ms time_period 60000ms report_interval 0ms data_len 1 alert_interval 1000ms latency_alarm 500ms loss_alarm 20% dest_addr ##.##.##.1 bind_addr ##.##.##.49 identifier "WAN_DHCP "