Link to home
Start Free TrialLog in
Avatar of Rob Womble
Rob WombleFlag for United States of America

asked on

how do you find a seemingly rogue dhcp server with no traffic related to the IP address displayed in DHCP Find utility?

I had this question after viewing DHCP Find is picking up a rogue server.

I am experiencing the exact same issues as the gentleman in the posted link. I'm seeing DHCP offers from a 112.112.108.116 IP address (which when looked up originates from China) Here's the caveat...The offers are IP addresses in the range configured in our regular DHCP server. There is no traffic captured in Wireshark from the IP address and there are no ARP entries anywhere with that IP address. I also have that IP blocked at the firewall. The accepted answer to that post was to see if there was a BLU phone connected to the network somewhere. There was no confirmation that this was actually the case. Anyone out there have any thoughts or ideas that they can send my way? I'm at a loss here...
Avatar of Dr. Klahn
Dr. Klahn

There is no traffic captured in Wireshark from the IP address

No, there wouldn't be; DHCP uses UDP.

You might try this:

Configure Wireshark to catch UDP packets of DHCP format.  Temporarily split the network in two at the router.  Connect Wireshark to one half and bring up a system or systems to see if any DHCP responses are caught from the rogue server.  If not, connect to the other half and check again.  Continue splitting at each hub or switch until you find the segment that the offending server is on.  Then go find it.  But note that this will work only if the device is wired.  If it's wireless, about all you can do is capture its MAC and make a guess at what it might be.
ASKER CERTIFIED SOLUTION
Avatar of atlas_shuddered
atlas_shuddered
Flag of United States of America image

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
SOLUTION
Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Avatar of Rob Womble

ASKER

Thank you guys for your responses. I'm looking into your suggestions now. I've already done a lot of what was already suggested. This issue seems to be affecting Mac's more than anything. Windows machines are getting addresses and Mac's are just not getting addresses and ending up with self assigned addresses.
Update: This issue seems to only affect Apple products. iPhones and MacBooks are getting addresses but no apple products. Also I see in the DHCP logs that there are several iphones that are trying to renew their addresses. There will be 4-5 renew requests in a row in the log per device and then cycles through with renew requests from each device over and over. Not sure what to think about this. Windows machines are not affected and receiving IP addresses as they should be.
4 questions:

1.  Is the network population the same as it was prior to posting this question?  In other words, have the Mac products received addresses previously without incident?
2.  You mention iPhones, I have to assume you are seeing those on wireless network.  Are the other affected devices attempting to connect via wireless?  Are the windows machines wireless or fixed port?
3.  Are you using DHCP reservations?
4.  What is the utilization of addresses in your DHCP database?
Thanks Atlas..

1.  Is the network population the same as it was prior to posting this question?  In other words, have the Mac products received addresses previously without incident?
Yes the network population is the same and the Mac products have previously received addresses. They are not receiving addresses either wired or wirelessly
2.  You mention iPhones, I have to assume you are seeing those on wireless network.  Are the other affected devices attempting to connect via wireless? Not that I know of..windows laptops are connecting just fine both wired and wirelessly.  Are the windows machines wireless or fixed port? Most Windows machines are fixed port. Some are laptops that have no trouble wired or wirelessly.
3.  Are you using DHCP reservations? No
4.  What is the utilization of addresses in your DHCP database? Currently at 74%
Now worries Rob.  I'll have to default back to my original recommendation of implementing snooping to attempt to find nullify your rogue and ping/arp to trace him down.  The update your provided gave me some hope that it may be some type of media problem for a minute but you have put that through the proverbial wood chipper.  There may still be some weird protocol issue at work.  You'd be back to a packet capture on that front.  Sorry
 
One other thing that you can do is run a debug ip dhcp after you have implemented snooping.  If he is out there, debug will show his drops in the logging.

Just to be clear on the snooping.  By setting up the trusted port you are telling the switch(es) what ports they are to expect to see dhcp ack/offers from.  Any port that you do not specifically place in trust mode will kill any traffic they see.  This does not impact dhcp request packets.
I've enabled dhcp snooping trusted the ports to the dhcp server only. Still being plagued by DHCP issues. To me...it looks like there's just a flood of DHCP offers coming from an external DHCP server...although I have that address blocked at the firewall and there is no traffic with that address as a source or destination address. here is what is showing up in DHCP find.
User generated image
Rob - I am posting the config notes from cisco on dhcp snooping configuration.  Looking back through my notes, I can see at least on cli comment that I did not include.  (ip dhcp snooping vlan "x" - where x is the vlan you want snooping to be activated).  

The link also includes directions for setting up source-guard but I would advice against this one unless absolutely necessary for resolution.  Implementing can have consequences that take your problem from bad to worse if not properly planned out and understood before input.

Have a look at the document at the link.  If you configure and still seeing the problem, I would suggest spanning your dhcp server port and sniffing the traffic to determine if it is pushing the rogue packets itself before considering source-guard.

https://www.cisco.com/c/en/us/td/docs/switches/lan/catalyst4500/12-2/25ew/configuration/guide/conf/dhcp.html
Thank you Atlas,
 I really do appreciate your wilingness to provide suggestions. I've temporarily shut down the Windows DHCP server and enabled DHCP on another device The traffic has immediately calmed down from 7% of the traffic down to .4% so I'm thinking there's something going on with the Windows server so I'll dig into that next. Thanks again for all your help
No worries Rob.  Good luck with the rest.