Loss of Network connectivity on random Client side computers, DNS/DHCP no longer being detected.

Question

I had to stand up a new DNS/DHCP/AD ((I will refer forward as the DC Server for simplicity)) server because our old one contained all of those plus the SQL Database for our production software and it was running Server 08. I apologize if I've posted this in the wrong area.

I got DC Server stood up and configured properly, least I think I did. I had no issues removing all the networked computers from the old domain and getting them connected to the new domain network, everything was working fine, all computers that needed static adresses were given an IP reservation after DHCP assigned them an address.

About two weeks ago, some of the computers started losing network connectivity with no rhyme or reason.

It's like they didn't even recognize the domain network at all, couldn't ping the server or anything.

Steps:

1. Restarted the affected computers
2. Restarted the switches in between the computers and the DC Server.
3. Checked IPv4 Settings for the corresponding network adapters for DNS: 10.63.1.11, as it should be
4. Checked to make sure DHCP was enabled, yes
5. Checked my network configurations for the DC Server. Static IP Address matches, 10.63.1.11, Subnet: 255.255.255.0, DNS is 127.0.0.1
6. Checked DNS: Forward lookup has all computers IP addresses, DNS Server itself ANAME 10.63.1.11

-All performed on my computer as it doesn't have network either, and wanted to test a temporary fix before pushing it to to the affected computers-

6. Flushdns, ipconfig release and renew
7. Get DHCP assigned address from the server. 10.63.1.107, Reservation Inactive.
8. Compared uniqueID, Computer Name assigned to my mac address of my physical network adapter on my computer, they both match.
9. Changed the settings on my network adapter to use static IP address .107 above, with Subnet 255.255.255.0 and DNS of 10.63.1.1, connection returns. I can pign the Server, RDP into it, and everything.

I at leats can get the Operator terminals back working so they can continue their work.

And this is where Im at in my trying to hunt down the root cause so I can this squared away.

Answer 1

Hello,

Do you have any issues with the client losing network connection?Thank you very much for your selfless sharing of the troubleshooting of this issue. Your sharing can expand the perspective of thinking for me and other users and promote the continuous development of the community.

Best regards

Zunhui

Answer 2

Yes. It started with one computer and its slowly cascaded to others over a two week span. Once it happens, they no longer have network connection to the domain nor the internet unless I assign the IPv4 address on the adapter itself.

Im still learning. Ive done small office networking for but I haven't had to deal with DNS/DHCP before, so its been a wild ride! I've never been able to learn by reading books, I have to do it/figure it out or enact for me to learn. Learn by breaking. My first time in an IT manager role, its been a wild ride. Have learned alot but so much more I need to learn.

There was one other thing I noticed while RDPed into the DC server. Looking at its network connection, it was listed as private network instead of domain network. I got that sorted with a regedit ((With Backup of original registry) under NLAsvc for it to be dependent on DNS, I failed to mention it in my original post. It did show the domain name on the network connection after a restart.

Answer 3

Hi JWReev,

Thank you for your reply. I suggest that you check the event logs of DNS, DHCP server and client to see if there is any relevant error information next time the problem recurs. This will make it easier for you to troubleshoot the problem. I suggest you refer to the following link:

Guidance for troubleshooting DNS - Windows Server | Microsoft Learn

Guidance for troubleshooting DHCP - Windows Server | Microsoft Learn

I hope the above information is helpful to you.

Best regards

Zunhui

Answer 4

So, I ultimately ended up standing up a DHCP server (By way of split scope at first as a test, onto another VM on the same server), which seemed to work, no issues over the weekend, Im going to continue monitoring it. THis morning, I brought the rest of the scope onto the new DHCP server since no issues were found.

There was nothing in the DNS logs when I checked. I looked over Weeks worth of DHCP logs, and the only things i've found was Computers making multiple IP renewal requests within a short period of time. I have a feeling this probably has something to do with my configurations and settings while figuring things out.

If all goes well, Im probably going to Deauthorize the original DHCP server since that split scope has the full Scope range from the Old DHCP server. Least.. I think it should work.. or try and force a failover and bring the old one down. Actually.. I have the old ones scope already deactivated. Since the Other one is bearing the load now, Might be safe for me to deauthorize it and be rid of DHCP on the DC.