Cisco 1130AG not getting dhcp

luthierone · ‎11-08-2006

Hello,

I have 6 1130's running LWAPP that are supposed to connect to a Cisco WLC 2000 series. However they aren't finding the controller. I assume this is because they don't get an ip address from our dhcp servers. LWAPP transport mode is layer 3. Our two dhcp servers are Win 2003. We have added the option 43 for the Cisco AP's. I consoled in to one of the AP's and I get this when I turn on dhcp debugging:

DHCP: SDiscover 301 bytes

*Mar 1 00:06:18.340: B'cast on FastEthernet0 interface from 0.0.0.0

*Mar 1 00:06:22.341: DHCP: QScan: Timed out Selecting state%Unknown DHCP problem.. No allocation possible

*Mar 1 00:06:30.344: DHCP: Waiting for 45 seconds on interface FastEthernet0

It will do this for a while and then reload and start over. Any ideas?

luthierone · ‎11-08-2006

OK,

I cleaned out our DHCP server which was full of BAD_ADDRESS entries and then I got this on the access point:

DHCP Ack Message

*Mar 1 00:20:36.087: DHCP: Lease Seconds: 691200 Renewal secs: 345600 Rebind secs: 604800

*Mar 1 00:20:36.088: DHCP: Server ID Option: 192.168.200.50

*Mar 1 00:20:39.087: DHCP: rcv ack: Address in use..send DECLINE

*Mar 1 00:20:39.088: DHCP: SDecline attempt # 1 for entry:

*Mar 1 00:20:39.088: Temp IP addr: 192.168.201.190 for peer on Interface: FastEthernet0

*Mar 1 00:20:39.088: Temp sub net mask: 255.255.254.0

*Mar 1 00:20:39.088: DHCP Lease server: 192.168.200.50, state: 2 Requesting

*Mar 1 00:20:39.088: DHCP transaction id: CB4

*Mar 1 00:20:39.088: Lease: 691200 secs, Renewal: 0 secs, Rebind: 0 secs

*Mar 1 00:20:39.088: Next timer fires after: 00:00:03

*Mar 1 00:20:39.088: Retry count: 1 Client-ID: 0018.babf.52ac

*Mar 1 00:20:39.088: Client-ID hex dump: 0018BABF52AC

*Mar 1 00:20:39.088: Hostname: AP0018.babf.52ac

*Mar 1 00:20:39.089: DHCP: SDecline placed Server ID option: 192.168.200.50

*Mar 1 00:20:39.089: DHCP: SDecline: 265 bytes

*Mar 1 00:20:39.089: B'cast on FastEthernet0 interface from 0.0.0.0

*Mar 1 00:20:39.089: DHCP: SDiscover attempt # 1 for entry

It just goes through all the available addresses, rejects each one and then the address gets used up on the DHCP server. Have you seen this before?

thanks

luthierone · ‎11-08-2006

Here is a offer message from the DHCP:

DHCP Offer Message Offered Address: 192.168.201.188

*Mar 1 00:27:07.081: DHCP: Lease Seconds: 691200 Renewal secs: 345600 Rebind secs: 604800

*Mar 1 00:27:07.081: DHCP: Server ID Option: 192.168.200.50

*Mar 1 00:27:07.081: DHCP: offer received from 192.168.200.50

*Mar 1 00:27:07.081: DHCP: offer received in bad state: Requesting punt

*Mar 1 00:27:07.081: DHCP: Received a BOOTREP pkt

*Mar 1 00:27:07.081: DHCP: Scan: Message type: DHCP Offer

*Mar 1 00:27:07.081: DHCP: Scan: Subnet Address Option: 255.255.254.0

*Mar 1 00:27:07.081: DHCP: Scan: Renewal time: 345600

*Mar 1 00:27:07.081: DHCP: Scan: Rebind time: 604800

*Mar 1 00:27:07.082: DHCP: Scan: Lease Time: 691200

*Mar 1 00:27:07.082: DHCP: Scan: Server ID Option: 192.168.200.52 = C0A8C834

*Mar 1 00:27:07.082: DHCP: Scan: DNS Name Server Option: 192.168.200.50, 192.168.200.52, &192.168.200.15&

*Mar 1 00:27:07.088: DHCP server identifier: 192.168.200.50

*Mar 1 00:27:07.088: xid: CB9, secs: 0, flags: 0

*Mar 1 00:27:07.088: client: 0.0.0.0, your: 192.168.201.188

*Mar 1 00:27:07.088: srvr: 0.0.0.0, gw: 0.0.0.0

*Mar 1 00:27:07.088: options block length: 82

*Mar 1 00:27:07.088: DHCP Ack Message

*Mar 1 00:27:07.088: DHCP: Lease Seconds: 691200 Renewal secs: 345600 Rebind secs: 604800

*Mar 1 00:27:07.088: DHCP: Server ID Option: 192.168.200.50

*Mar 1 00:27:10.088: DHCP: rcv ack: Address in use..send DECLINE

*Mar 1 00:27:10.089: DHCP: SDecline attempt # 1 for entry:

*Mar 1 00:27:10.089: Temp IP addr: 192.168.201.188 for peer on Interface: FastEthernet0

*Mar 1 00:27:10.089: Temp sub net mask: 255.255.254.0

*Mar 1 00:27:10.089: DHCP Lease server: 192.168.200.50, state: 2 Requesting

*Mar 1 00:27:10.089: DHCP transaction id: CB9

*Mar 1 00:27:10.089: Lease: 691200 secs, Renewal: 0 secs, Rebind: 0 secs

*Mar 1 00:27:10.089: Next timer fires after: 00:00:03

*Mar 1 00:27:10.089: Retry count: 1 Client-ID: 0018.babf.52ac

*Mar 1 00:27:10.089: Client-ID hex dump: 0018BABF52AC

*Mar 1 00:27:10.109: DHCP server identifier: 192.168.200.50

*Mar 1 00:27:10.109: xid: CB9, secs: 0, flags: 0

*Mar 1 00:27:10.109: client: 0.0.0.0, your: 192.168.201.189

*Mar 1 00:27:10.109: srvr: 0.0.0.0, gw: 0.0.0.0

*Mar 1 00:27:10.109: options block length: 82

luthierone · ‎11-08-2006

We plugged the AP into a different DHCP device and it got an address. Here is a successful dhcp transaction:

look up prim DNS for Fa0 from lease good ret: 192.168.1.1

*Mar 1 00:01:56.781: DHCP: look up domain name for Fa0 from lease any got domain name ret: succeed

*Mar 1 00:01:56.788: LWAPP_CLIENT_ERROR: lwapp_name_lookup - Could Not resolve CISCO-LWAPP-CONTROLLER.SME

*Mar 1 00:01:57.439: DHCP: Attempting to speed up renewal timer

*Mar 1 00:01:59.346: DHCP: QScan: Bound..T1 fired..Renewing

*Mar 1 00:01:59.346: DHCP: SRequest attempt # 1 for entry:

*Mar 1 00:01:59.346: Temp IP addr: 192.168.1.100 for peer on Interface: FastEthernet0

*Mar 1 00:01:59.346: Temp sub net mask: 255.255.255.0

*Mar 1 00:01:59.346: DHCP Lease server: 192.168.1.1, state: 5 Renewing

*Mar 1 00:01:59.346: DHCP transaction id: CA1

*Mar 1 00:01:59.346: Lease: 86400 secs, Renewal: 43200 secs, Rebind: 75600 secs

*Mar 1 00:01:59.346: Temp default-gateway addr: 192.168.1.1

*Mar 1 00:01:59.346: Next timer fires after: 09:00:01

*Mar 1 00:01:59.346: Retry count: 1 Client-ID: 0018.babf.52ac

*Mar 1 00:01:59.346: Client-ID hex dump: 0018BABF52AC

*Mar 1 00:01:59.347: Hostname: AP0018.babf.52ac

*Mar 1 00:01:59.347: DHCP: SRequest - ciaddr: 192.168.1.100

*Mar 1 00:01:59.347: DHCP: SRequest placed lease len option: 86400

*Mar 1 00:01:59.347: DHCP: SRequest placed class-id option: 436973636F204150206331313330

*Mar 1 00:01:59.347: DHCP: SRequest: 307 bytes

*Mar 1 00:01:59.367: DHCP: Received a BOOTREP pkt

*Mar 1 00:01:59.367: DHCP: Scan: Message type: DHCP Ack

*Mar 1 00:01:59.367: DHCP: Scan: Server ID Option: 192.168.1.1 = C0A80101

*Mar 1 00:01:59.367: DHCP: Scan: Lease Time: 86400

*Mar 1 00:01:59.367: DHCP: Scan: Subnet Address Option: 255.255.255.0

*Mar 1 00:01:59.367: DHCP: Scan: DNS Name Server Option: 192.168.1.1

*Mar 1 00:01:59.367: DHCP: Scan: Domain Name: SME

*Mar 1 00:01:59.367: DHCP: Scan: Router Option: 192.168.1.1

*Mar 1 00:01:59.367: DHCP: rcvd pkt source: 192.168.1.1, destination: 192.168.1.100

*Mar 1 00:01:59.367: UDP sport: 43, dport: 44, length: 308

*Mar 1 00:01:59.368: DHCP op: 2, htype: 1, hlen: 6, hops: 0

*Mar 1 00:01:59.368: DHCP server identifier: 192.168.1.1

*Mar 1 00:01:59.368: xid: CA1, secs: 0, flags: 0

*Mar 1 00:01:59.368: client: 192.168.1.100, your: 192.168.1.100

*Mar 1 00:01:59.368: srvr: 192.168.1.1, gw: 0.0.0.0

*Mar 1 00:01:59.368: options block length: 60

*Mar 1 00:01:59.368: DHCP Ack Message

*Mar 1 00:01:59.368: DHCP: Lease Seconds: 86400

*Mar 1 00:01:59.368: DHCP: Server ID Option: 192.168.1.1

*Mar 1 00:01:59.368: DHCP: Releasing ipl options:

*Mar 1 00:01:59.368: DHCP: Applying DHCP options:

*Mar 1 00:01:59.368: Setting default_gateway to 192.168.1.1

*Mar 1 00:01:59.369: Adding default route 192.168.1.1

*Mar 1 00:01:59.369: Adding DNS server address 192.168.1.1

*Mar 1 00:01:59.369: Setting default domain to SME

*Mar 1 00:01:59.369: DHCP Client Pooling: ***Allocated IP address: 192.168.1.100

Translating "CISCO-LWAPP-CONTROLLER.SME"...domain server (192.168.1.1)

*Mar 1 00:02:06.788: DHCP: look up prim DNS for Fa0 from lease good ret: 192.168.1.1

*Mar 1 00:02:06.788: DHCP: look up domain name for Fa0 from lease any got domain name ret: succeed

*Mar 1 00:02:06.794: LWAPP_CLIENT_ERROR: lwapp_name_lookup - Could Not resolve CISCO-LWAPP-CONTROLLER.SME

AP0018.babf.52ac#

Translating "CISCO-LWAPP-CONTROLLER.SME"...domain server (192.168.1.1)

*Mar 1 00:02:16.794: DHCP: look up prim DNS for Fa0 from lease good ret: 192.168.1.1

*Mar 1 00:02:16.794: DHCP: look up domain name for Fa0 from lease any got domain name ret: succeed

However, this is not a feasible option. I need to use the dhcp servers that are not working with the AP's. I should mention that these servers do work with other devices...just not the cisco AP's.

thanks

Jason Williamson · ‎12-15-2010

"BAD_ADDRESS" is filling up my DHCP scope as well... And I can't find ANYTHING about it on the web that's of any use. Help?

maldehne · ‎12-16-2010

Well , In one of the logs you have captured from the AP , it seems that the AP got an IP and then discovered that this IP is already in use

by some other device. Accordingly it sent a DHCP decline message.

Per RFC, if client rejects an IP address (it found it is in use), server must mark it as reserved.

4.3.3 DHCPDECLINE message

If the server receives a DHCPDECLINE message, the client has

discovered through some other means that the suggested network

address is already in use. The server MUST mark the network address

as not available and SHOULD notify the local system administrator of

a possible configuration problem.

http://tools.ietf.org/html/rfc2131

You should contact or check this with your DHCP server provider, as the RFC is clear on the action that server should take, and it is failing to do so. (I guess server wants to prevent a DoS situation for all address allocation, but this should be configurable)

One more thing, there could be some clients that are getting IP addresses from the server and keep using the same IP address even after it expires from the server without sending a renew request to the server , so when a new client jumps in the server leases that address but when the client wants to confirm that it is not in use by another client it sends arp request with that IP and the lovely malicious clients says "hi, i am still using that one ",........BOM. The poor client sends DHCP decline and it keeps going like that.

There is a well known issue that has been reported with iPad and iPhone (OS 3.2) stopping renewing DHCP lease:

Under certain circumstances, iPhone OS 3.2 on the iPad stops renewing its DHCP lease, yet continues using the IP address after the lease expires. Although the owner of the iPad may not realize there is a problem, this interferes with service to others on the network.

http://www.net.princeton.edu/announcements/ipad-iphoneos32-stops-renewing-lease-keeps-using-IP-address.html

Hope this would help

Mohammad Aldehnee

--------------------------------

Don't forget to rate answers that you find useful

Jason Williamson · ‎02-10-2011

Just a quick note to take care of this loose end. In our case, we found a problem in the core IST trunk. Erroneous MAC entries in the cores.

jeremyarcher · ‎04-26-2011

We've experienced this as well. The AP requests the address (and accepts the address offered) but the Windows DHCP server believes it wasn't accepted and reallocates it into the pool if available addresses. Then, when another client requests the address the Win. DHCP server ping the address first, gets a reply from the AP and marks it as a BAD_ADDRESS.

We have moved to static addresses on the APs to resolve the issue.

Sent from Cisco Technical Support iPhone App