A friend of mine enlisted my help to upgrade his network form a workgroup to a domain, so I did. I'm give you the technical setup and explain the issue at the bottom.

1 Sonicwall tz100 with static IP and cable connection (runs DHCP and DNS, servers IP is listed ad a DNS source among those provided by the ISP, I did this so the internal PC's can resolve computer names {is that correct?})
1 Cisco Small Business 48 port Switch
15 workstations
1 server with 08 Standard installed (file server only)
All PC's have DHCP enabled

My buddy didnt want DHCP or DNS running on the server once it was upgraded to a domain, but as you may already know, DNS is a required role, so I installed it. The setup was simple and it worked fine up until now (90 days). I gave the firewall the role of DHCP and DNS, the server also has DNS role too. Since when setting up the firewall you can give a DNS address, I entered the servers IP. Since the server runs DNS also, I gave it (forwarders) the external DNS IP's.

I did that so that PC's in the internal network could communicate on a computer name level without just using IP addresses. Anyway, the whole thing worked like a charm and I realize that this setup is unusual, I know. But under the circumstances, I didnt know how else to set it up (besides putting both DHCP and DNS on the DC)

THE ISSUE:
For the past two days, when everyone came into work, (all win7 PC's) they display unidentified network with no internet or network access (most of them, some actually worked fine, all setup with DHCP). We did some troubleshooting over the phone, restarted switch, then firewall (not the server). Once we rebooted the firewall, everything worked. My question is, is there some other way I should have set this up or is there some sort of loopback issue thats causing this issue after such a long time. Or could it just be a bad firewall (9 months old)

Thanks in advance!

Dani AI

Generated

Short expert summary: the pattern — many Win7 clients arriving, showing "Unidentified network" and restoration after rebooting the SonicWall — strongly points to an intermittent DHCP/gateway or DNS-relay failure on the firewall rather than a subtle AD-only problem. Reboot fixing everything is a classic sign of a single-device service failure or resource/firmware bug. Both and touched on the core tradeoffs: remove extra DNS hops where possible and avoid placing DHCP behind a single point of failure.

Immediate diagnostics (when the fault reappears) — collect these outputs and timestamps for correlation:

ipconfig /all
ipconfig /release
ipconfig /renew
nslookup <dc-hostname> <dc-ip>
nslookup google.com <dc-ip>

Key fields to check: DHCP Server, DNS Servers, Connection-specific DNS Suffix, Default Gateway, Lease Obtained/Expires, and any 169.254.x.x address. Check the SonicWall DHCP logs for errors and the Windows Server event logs (DNS, DHCP, System) and run dcdiag on the DC to verify DNS/SRV health.

Practical remediation (prioritized):

  • Ensure the DC has a static IP outside any DHCP scope and hosts an AD‑integrated DNS zone with proper SRV records and forwarders.
  • Configure DHCP to hand out the DC(s) as the primary DNS (option 006) and the AD domain as the DNS suffix (option 015). This can be done on the SonicWall but moving DHCP to the Windows server (authorized in AD) is the more robust long‑term fix because it supports secure dynamic updates.
  • Shorten lease time temporarily for faster recovery while troubleshooting.
  • Update SonicWall firmware, monitor logs for DHCP failures, and consider hardware replacement if instability persists.

Note: using a non‑AD DNS (or a firewall as the primary resolver) for domain-joined clients often causes intermittent authentication/name‑resolution issues and should be avoided for production AD clients.

Recommended Answers

All 2 Replies

Hard to say without more details, but you definately have an unecessary flow of DNS on that network. Rather than having your clients point to hte firewall for DNS, then foward them to server, then forward out, just configure the DHCP scope and set the DNS clients to point directly to the server running DNS. This will remove one level out of this DNS flow and improve the speed of name resolution.

you run DHCP and DNS on Cisco Switch. if the firewall goes down or any problem occurs, the entire LAN will affected, as the machines wont get IP Address.
You internal LAN works well - if the DHCP and DNS works on CISCO SWITCH.

Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.