New - R9000 Firmware Version 1.0.5.42 Released

I upgraded from V1.0.5.40 and had no end of issues requiring frequent reboots to return full speed.  My setup is as follows,

• COMCAST 1.2Gbps down 45 up (I get 1.425Gbps on speedtest)
• Motorola MB8611 Modem
• WAN Port is using a 10Gbps SFP+ -T port with CAT7 cable between R9000 and Modem - negotiated at 2.5Gbps
• LAG (Only static works for some reason) to my Office
• GC728X switch running LAG to router
• Many other (76 at last count) devices on Wifi 2.4 and 5 and Ethernet
• 802.11ad (60Ghz) is turned off
• R9000 handling DHCP (this will change soon)
• PiHole running DNSfwd as the primary
• A few other switches connected
• VoIP Services
• No other firewalls or significant NW hardware on the net, but a whole lot of clients

I know it's a pretty "weird" setup but I'm a career NW engineer and SW developer so my work is also my hobby to an extent.  I'd recently changed a number of things at the house including new equipment and the SFP+ connection so quite a bit changed within a few weeks of my "upgrade".  This happened right after I upgraded to 1.0.5.42 which worked well for a few days before the rest of it and I'd put that firmware update off as "not the issue".

The symptoms were a general slowdown or disconnection primarily of the ethernet connected devices.  I saw slowdowns in several "steps" the first being the 70Mbps range for download the next being ~20Mbps and the last being Kbps.  Sometimes WAN connectivity was essentially off but for some reason pings usually got through.  Anything more significant like web page loading was not happening.  Finally ALL devices (Windows and Apple) that have an internet heartbeat would appear "Disconnected from Internet" in that last state.

I tried nearly everything to increasing FAN speed, disabling LAG, setting up VLANS for chatty home automation devices, re-terminating connections, replacing cables, turning off XBoX On Demand and putting it on WiFi and nearly everything else I could find in the forums.  Some things seemed to make it less terrible but nothing "fixed' the situation.  The logs off the main interface didn't provide any hints and the debug logs didn't have anything that would lead to the issue.

One thing of note, I have a Ryzen 5900x workstation with a 10Gbps SFP+ network connection it.  for some reason, booting this machine and logging in seemed to be a pretty regular and reliable event that would cause the R9000 to choke. I pulled the 10G connection and was using both a PCIe card and the built in 1Gbps network to no avail.  If that machine booted (Win10) and started doing heavy network traffic it would cause the R9000 to flake.

After all this (two weeks), I decided to downgrade to V1.0.5.40 and so far (past few hours) I've been trouble free.  I realize that I should have downgraded sooner and not wasted the last week trying everything else under the sun.  I wist Netgear would add specifics on the "security  updates" to allow us to choose if the risk of the security issue is less than the risk of what I just went through. I understand the rationale of not publishing it as a misguided sense of additional protection for users that haven't upgraded but it makes it really tough to make an educated decision.  I'll post a follow up in a few days on 1.0.5.40 and if it's 100% solid, I'm saying .42 has a bug in it.  At least with my configuration.