The Dynamic Host Configuration Protocol (DHCP) is an auto configuration protocol used on IP networks and an extension of the Bootstrap Protocol. DHCP allows for computers to be configured automatically to communicate with each other over an IP network without the need for manual setup by a network administrator. The implementation of DHCP relies on a DHCP server to hand out network configuration information to DHCP-capable clients that request an IP address (and other information required or useful in communicating with other devices on an IP network). In addition to an IP address, common configuration information served over DHCP includes a default gateway, subnet mask and DNS sever(s).

Share tech news, updates, or what's on your mind.

Sign up to Post

I need to move DHCP from a Linux server to Server 2016.  This is for a PXE boot environment.  I need some help figuring out the DHCP options.

Here is how the Linux DHCP is configured:
class "pxeclients" {
match if substring (option vendor-class-identifier, 0, 9) = "PXEClient";
next-server x.x.x.x
if option arch = 00:06 {
filename "shimx64.efi";
} else if option arch = 00:07 {
filename "efi64/shimx64.efi";
} else {
filename "bios/pxelinux.0";
Microsoft updates claimed another machine.  this domain workstation this morning gives the 'no computer account for trust relationship' error refusing login to domain network/server.  i can disconnect the Ethernet cable, log in, reconnect the cable and all is perfect.  this is a doc workstation with client server patient management, xray, patient charts,  plus all the usual suspects.  6 mo old Dell 640 single Server 2016 domain controller, dhcp role, dns role, os raid 1, data 8 drive raid 6.  medium size dental practice.

i know to recreate the trust the hard way but will recreate user and loose his desktop, setting etc which i would prefer to avoid as would he.

can i reestablish the computer account trust via a simpler and less destructive procedure?
Morning folks - I want to set up a Dell 2724 Managed switch for my home / home lab to differentiate some of the traffic.

So, a quick hardware breakdown

Dell 2724 Switch
Internet Router (DNS done through this for home etc)
Home network on 192.168.0.x private network
I have introduced 3 VMWare Esx boxes with 4x NICs each, together with 2x NAS boxes (again 4x NICs).
The home network will just go out from the switch to a central DHCP server and should be on VLAN "Home" (Cables 1 - 10) (192.168.0.x) - the first NICs of the ESX hosts and NAS boxes would also go on this subnet.
I want to set up a second VLAN for the vMotion and data migration for the ESX boxes (Cables 11 - 15 for example (3x ESX, 2x NAS) (10.10.10.x ip range), using the 2nd (and possibly third) NICS of the boxes.
The VLANs would need to be separate, so no cross traffic between the two for data, basically I want to flood the ESX network without slowing down the home network.

I've had a look at the setup of the 2724, but am getting a bit confused about the T and U bindings of the VLAN setup.

Any help would be really appreciated.

Many thanks in advance.
Hi there
Win2k16,  DHCP server with scope
installed 3 UniFi Access Points (APs)
want clients to obtain  ip requests from the Windows DHCP
and place under scope ... (Dont have Unifi USG)

please help
we have a failover dhcp server composed of two windows server 2016 standard vms.
lately, we notice frequently that ip addresses become marqued as BAD_ADDRESS and computers that were holding these ip addresses cannot get another valid ip address,
we discovered that some support guys, set manually network configuration for some computers (that become BAD_ADDRESS), and those ip addresses are in dhcp scopes, which makes a part of computers disconnect from network.
when I tried to reproduce this problem in a newly installed lab, I find that the address becomes BAD_ADDRESS but the client gets another IP address,
I would like to know what could prevent dhcp servers from acting correctly and not handling ip address for client after detecting BAD_ADDRESS in network.
thank you

I am trying to boot up PXE on at host that is connected to a Juniper Ex2200 switch running DHCP, but it just hangs everytime it starts up... does anybody have an idea what might be missing on the DHCP from the Switch?

Tks in advance.
Have a Windows 2012 Server running a standard VPN connection to itself (no Direct Access)

Server Name:  CHSERVER

Server has IP of

When connected via VPN, user can't access server when using \\CHSERVER\...

And when try to connect with \\\...  same result

Also, can't ping via VPN


When you try to connect with \\ which is the first IP of the allow DHCP scope for the VPN and which appears to be assigned to the Internal Interface, it will allow you to connect.  Also, you can ping CHSERVER with

I tried deleting the VPN setup and starting over again, but same results.  Was working a couple of weeks ago, but I can't see where anything changed.

Need some help here.
Windows 10 Workstations are not getting the correct DNS server from the Server 2016 DHCP Server.

Any suggestions?
We have several Avaya 9611G that are randomly re-registering.
Not at the same time.  And occasionally they will reboot.
This happens to all phones at periodic times during the day.

I can't find a correlation or cause.  I've removed the PC from the chain
and gone direct connect.  I've set a static IP to rule out the DHCP server.
The firmware has been updated.  Problem persists.

The sip server is hosted.  The host has connected a phone to the system
but did not have the same problem.

Our PoE switch, Cisco 300sg, doesn't show any issues with power consumption to the ports.
Only using a third of its total capacity.  Don't see any event issues that correspond to re-registers time stamp.
Original subnet is 10.010.0/24 and we are out of IP addresses, and are already allocated. Superscope added using and secondary address added to port on router to allow communications.

Now original subnet 10.0.10.x gets identified as unknown network and new subnet shows our domain name

Printers on original subnet stopped working until after PC reboots even for PC’s on same subnet.

DHCP service is on our File/Print server (Windows Server 2012).  Both scopes added to a superscope “folder”.

Unknown reasons why original subnet is showing unknown network or why printers not working.

Are there additional settings that should be configured?  Can we resubnet using 10.0.9.x or 10.0.11.x with different subnet mask and keep the and operational?
I have a 3-node Hyper-V HA cluster running Server 2019. I have everything identical, except I am trying to solve two problems:
I am pinging from a laptop on the same network as the nodes. Node 1 returns pings with higher-than-normal latency compared with the other two nodes- Normally they are <1ms-1ms. Node 1 returns a range from 1-40ms with drops about .5%-1% of the time. All nodes plugged into the same switches. Each node has 8 network adapters, 6 are teamed as "cluster" and two are teamed as "backup". The ping times are weird when the backups are not running, not sure about during backup windows. All four virtual adapters on all three nodes have statically-assigned MAC addresses that are unique and were set up to address other issues that Microsoft has acknowledged are actual issues. The 4 adapters on each server with these static MAC's are the two teams I mentioned and the Hyper-V switch associated with each team- So we have a vEthernet Cluster, a vEthernet Backups, Cluster, and Backups, each team is assigned to a Hyper-V switch. Each of the switch adapters is assigned a static IP address on the two networks- for example, node 1 is on Cluster and on Backups. Both networks are routable to each other.
On node three, I have a weird problem, andit might be related- ipconfig /all shows two IP addresses on the Cluster interface- and Both are statically assigned (No DHCP server on the network). When I go …
I do work for a municipality and have run into an issue.  The building is run on a domain with a domain controller that runs the DNS and DHCP.   The Internet provider owns the firewall so they control it and we cannot even have the admin password.   There is also a DSL line that is used to go directly to the state.   Up until two moths ago the two computers that use the DSL line to the state would also be on the domain for the building.   Then there was an issue where the state said their site was down over the whole state.   Then when the state brought the site back up, those two computers had to be detached from the building network and be directly plugged into the DSL router only.   Again, before this happened the two computers were able to get to the state site via the DSL line, but then the users could also print to the building network printers even is they downloaded from the state site and needed that file printed.   If I do a trace route, the first thing it hits is the ISP network firewall then it hits the DSL router, then it goes out into the cloud to the states server.      Any ideas how this could have been setup or why it might not be working now?  Any ideas?
I have a 3-hypervisor cluster where each server has 8 network adapters, which were originally teamed. Right around the time we started a DHCP server upgrade, virtual machines started acting squirrelly in large groups- we would have 6 or 8 become completely unresponsive and after failing over, the same thing would happen again with different machines. We traced the problem to NIC's in the teams on the hypervisors where some had received IP addresses from the new DHCP server and some would not renew properly. To solve the stability problem, I had to unteam the NIC's. It has not returned, so I want to make sure I rebuild the team properly on each server. The hypervisors are 2019. Should I statically assign IP addresses to the NIC's prior to teaming them, or should I leave them as DHCP? Our network is very odd- static addresses are assigned in one subnet and DHCP in There is a router between both networks. Any ideas?
I just set up and am test a failover DHCP server system using two Windows 2019 Standard servers. I cannot confirm if this is a problem on Windows 7 clients. Both of the DHCP servers are authorized. I built this by creating an empty DHCP server, backing up the production DHCP server, and restoring the backup to the new DHCP server, then set up the second DHCP server as a failover partner. Both servers say they are in sync with each other, and they are handing out IP addresses in the same range the old DHCP server did. The Windows 10 64-bit clients display the network name as "Domain.local 2 (Unauthenticated)" any idea why or how to fix this? All the answers I have seen focus on the local computer.... but in this case, the local computers are not the issue- I can toggle this by changing which servers have the DHCP Server service running. Any ideas what I need to configure on the DHCP servers to fix this?
Hi, I'm looking for some guidance as to how to approach this.

I presently have an ISA550 and an SG300-52p switch. Unfortunately, ISA550 doesn't have a direct replacement and I've decided to purchase the ASA 5525-X to learn and use as my router. I'm new to CLI and the ADSM configuration so I'm trying to get a handle on what is the best approach.

At present, my ISA550 manages the DHCP Route thru multiple VLAN's, assigned to the TRUNK going to my SG300-52p. I have multiple computers on their own VLAN and WIFI AP connected to their own VLAN. Additionally, my ISA550 allows me to do DHCP reservations which helps me ensure specific machines and devices are assigned a single IP.

With the move to ASA552-X, I've lost my way as to what approach to take. From what I'm being told ASA5525-X does not do DHCP Reservations, the closest is ARP tables, but not really the same thing. So I've been told the SG300-52p can be my DHCP server and I'd retain the ability to manage DHCP Reservations. Great Idea I though, only to realize I don't understand how to configure the device so that it works for me.

So it could help me understand what configurations do I need to do to support the following:
1. On SG300 I was able to configure some DHCP Pools. I now need to know how to configure the ASA5525-x so that it passes the multiple TAGs to the Switch.
2. Do I still need to configure a DHCP pool (BVI: Bridge Group Interface) on the ASA5525?
     a) if not, then how does the internet get …

There are 2 DHCP servers in a cluster 50% 50% each one.

When a new scope is created in dhcp 1 is not replicated  to dhcp 2. Same problem with static reservations.

Any idea?

Good day,
I am assisting a school with an extremely strange issue.  The situation is pretty simple.  We have a DHCP server that is broadcasting, but some of the computers simply won't get an IP address.  Most of the computers get a 169 address.  But we cannot find what is wrong.

If we statically assign the IP on the computer, it works totally fine.  Access is good, ping is all good.
We have rebuilt the DHCP server that is on the OS.  Deleted the config, and recreated the subnet.
We have reinstalled the DHCP application.  We are using DHCP for Windows, its worked fine for 2 years.
We even disabled the DHCP server, and enabled the DHCP server on the router.  It has the exact same result.
From the router, it is sent to 4 newish (last month) gigabit unmanaged switches.  Most computers are connected direct to this, but there are a few unmanaged small 5 port switches on the network as well.
We have even disconnected each switch one by one, but the issue remains until no computers are on the LAN.  Seemed to only be reliable when it was just the router with 2 devices connecting wirelessly.
Obviously we cannot run like this.

I don't believe there is a rogue router onsite, as even when we disable our DHCP server, it doesn't get an address.  

Is it possible that the DHCP requests are being blocked or disrupted somehow?  And if so, how do we find it?  THere are about 20 staff computers running windows, and 200 chrome boxes/books, and 20ish other network devices, like …
My private IP addressing is in the 10.10.10.x network, subnet mask, is the default gateway. Now on some computers, I am getting a subnet mask of I couldn’t figure out why?
Can someone help me with this?

I actually replaced the switch, thinking that some ports on the switch were bad, but I keep getting the same issues.


I have tried to setup the DHCP to be from the Windows Server instead of the DHCP on the Juniper but I cant seem to get any of the computers on the subnets I am trying to relay to get leases from the Windows server... does anybody have a good tutorial or any CLI command list to accomplish this?? Juniper is a Ex2200 and Windows Server 2016..
We are moving to a new Windows domain on the same network. The current DC is also running DHCP. We can setup AD and DNS on the new DC without issue, but we need DHCP to join PCs to the new domain. Can two DHCP servers be configured to run on the same network, Do we need to stop the live, production DHCP service to perform tests with a computer on the new domain?

What is best way to move from an existing DHCP server to a new one with the same scope?

Good Afternoon,

Would any one happen to have  powershell script that will interrogate a DHCP server (that I input) and ask for the beginning and end range of the scope and produce a report on the scope name, number of leases, number of leases used, percentage used.

I have tried some scripts that I have found but they check all scopes on the server and even miss some of them out.

Many thanks in advance

We have Meraki Wifi and have experienced outages for users, but this only seems to be affecting a certain AP.
Looking at the event logs, I can see at the time of the outage there's Multiple DHCP servers entries in the log.
AP is connected to a Cisco access switch, port is trunk and allows two vlans.
Access switch connects to a core switch on which there is ip helper-address config for two different DHCP servers.

interface VlanX
 description Data-Vlan
 ip address x.x.8.15 secondary
 ip address x.x.11.20
 ip helper-address x.x.8.20
 ip helper-address x.x.56.128

the above is the vlan for the local LAN and another vlan allows users who connect on another SSID for direct internet access.

Event log extract:

Multiple DHCP servers detected      vap: 1, original_server_ip: x.x.8.20, original_server_mac: 00:50:56:96:EF:9D  « hide
server_ip      x.x.8.15
server_mac      08:CC:A7:77:D9:F7

Any assistance or experience with this is appreciated.
Hi - we have a wired/wireless network at a school, multiple VLANS, IP phone system, using a Sonicwall NSA2600. For the last 4 days, I received calls where the internet was "down" for the computers, yet the phones were working (both coming into the building and going out of the building). I was NOT able to access any computers that I have unattended access to, except for the computers that are on only 1 VLAN. So this 1 VLAN and the phone system, both still DHCP, were unaffected by this issue. A simple reboot of the firewall "fixes" the issue temporarily, so they're at least back up and running. The most puzzling part of troubleshooting this is, why this one VLAN isn't affected by the issue where every other one is. Initially I thought it may have been a router that a student setup (intentionally) creating a rogue DHCP server, or some type of broadcast storm. But still cant figure out why that one VLAN is unaffected. Anyone have a good way of tackling this one?
I have a network which currently has the voice and data all on the same network space ( The DHCP server is a Draytek 3900 Series Firewall and the network switch is Catalyst 2960 L series.

How do I setup the 2nd VLAN for voice and get the firewall to issue a different IP ( for the phones whilst leaving the PC's on the original network.

Sycamore IT
I have been using Netsh to create a back up text file and Database file(db) and automatically copy the results to a server share as follows:
netsh dhcp server dump > \\Server-IP-Address\Sharename$\dhcpcfg.txt
netsh dhcp server export \\Server-IP-Address\Sharename$\dhcpdb all
This worked great with Server 2008 R2 but is not working with Server 2012 R2. Can you please advise me as what the problem and solution might be or another command that might do the same thing?






The Dynamic Host Configuration Protocol (DHCP) is an auto configuration protocol used on IP networks and an extension of the Bootstrap Protocol. DHCP allows for computers to be configured automatically to communicate with each other over an IP network without the need for manual setup by a network administrator. The implementation of DHCP relies on a DHCP server to hand out network configuration information to DHCP-capable clients that request an IP address (and other information required or useful in communicating with other devices on an IP network). In addition to an IP address, common configuration information served over DHCP includes a default gateway, subnet mask and DNS sever(s).