Broadcom Driver issue?

[radiant surge]

My HA server started becoming unresponsive almost daily, or sometimes multiple times per day, with only a hard reset resolving the issue. Looking through the host logs, I see issues that suggest to me it may be either a failing NIC or a broadcom driver issue and I figured I'd ask y'all's opinion before I go buy an external NIC or similar.

In my host logs, I'm receiving errors like tg3_stop_block timed out, Link is Down, NETDEV WATCHDOG ... transmit queue 0 timed out ... resetting, and issues showing where it fails to query external DNS servers due to an i/o timeout.

I can post a full log if needed. Mostly just sanity checking myself that I may need a new NIC.

I should note that this is running on an old optiplex 5055 and it is using the motherboard's nic

[radiant surge]

Short term, I'm going to create an automation with uptime kuma to automatically reboot my system if it seems to go down but this is not an ideal solution. Aside from buying a new NIC, I'm not sure what other solutions I may have. Likely needs to be a usb NIC since I do not have any pcie slots available

[buoyant ocean]

you can try and disable IPv6 in network settings as I had this on auto and my dashboard / webui was super laggy before I turned off ipv6

[radiant surge]

oddly, I installed uptime kuma and was going to build an automation where if my dns was unable to resolve it would reboot the machine -- as a workaround. Ever since then, my server has not become unresponsive. Unknown if it's a coincidence or if the keepalive has kept it up lol

[buoyant ocean]

Weird, anyway I had super slow UI recently despite it's VM with 4 cores assigned, 8 gb ram and it's intel 125h so overkill, idling on 15 %, turned off ipv6 then had to turn it on as matter uses it, so I assigned cloudflare ipv6 dns via HA and it's super snappy again

So well I don't understand this either but it works