So I've got an issue with a PC that I haven't seen before. If you've worked with Windows 10 at all you know how garbage the updates can be when it comes to breaking the TCP/IP stack and preventing network access. The typical situation involves it not pulling an IP from the DHCP service. Usually this is fixed by resetting the network or running netsh int ipv4 reset at a prompt. In this scenario I appear to be pulling an IP, but upon restarting the adapter it loses the IP and can't get another. I also am unable to ping the gateway and get a "General Failure".

Things I have tried:
netsh
network reset
installing a NIC
removing and reinstalling the driver
happened after an update so I restored to a previous point
pinging the gateway

I'm looking for any information that may help to figure this one out. I had assumed it to be a hardware problem but the new NIC doesn't work either so it would seem to be something in the software.

Edit: worth noting that this isn't a router issue. It's an enterprise environment and the router is working just fine for other devices on the same port.

This post was edited by Lethardus on Jul 19 2017 07:58am

Which nic? I would still think it's hardware related. Seems too funky to not be he related.

This post was edited by Penguins0690 on Jul 19 2017 09:40am

I installed a completely new PCI NIC and it's having the exact same issue.

When you reverted did it work?
Cause I mean this is why deferred updates exists.

Negative. This is a customer machine, I have no control of whether or not they setup deferred updates before it comes to my bench. Believe me, if I could I would.

Guess go the format route and set it up for them.
If reverting didn't fix it, One would assume debuging it would take far too much time.
(since time = money, I personally wouldn't invest any more time into it)

Pretty much the route I was leaning towards, but I wanted some secondary opinions since my company would charge them $145 to do the wipe and reload. It's pretty absurd to be honest but he I don't make the rules.

I feel ya...

I mean sometimes if a client doesn't mind waiting a week and I'm really
determined I would totally just hold it and fuck with it here and there.
But the reality is you need to have a fast turn around working for others.

I've cloned drives in the past to and dumped em in a VM but these are extremes and only to satisfy my own needs to figure soemthing out.

There was an update last year that introduced this feature, and a later one that patched it. Could it be that? Or is it fully up to date now?

Can't say that I heven't done this myself. Some things just pique the intellectual curiosity too much to "let it go". This isn't one of those times.



Not sure what "feature" you are referring to but yes it was fully up to date when it came in. It's obviously now as I did the restoration and the internet connection still doesn't work, thus no updates.