Have a Cisco 4507R+E. This is an L3 switch and serves as DHCP server for one VLAN and a DHCP Relay for all other VLANs. There have been no issues until today when the DHCP Relay stopped working. No changes made to DHCP settings in the configuration. Did a bit of debugging and get this:
Feb 4 14:24:10.609 EST: IP: s=0.0.0.0 (Vlan311), d=255.255.255.255, len 328, input feature
Feb 4 14:24:10.609 EST: UDP src=68, dst=67, MCI Check(63), rtype 0, forus FALSE, sendself FALSE, mtu 0, fwdchk FALSE
Feb 4 14:24:10.609 EST: FIBipv4-packet-proc: route packet from Vlan311 src 0.0.0.0 dst 255.255.255.255
Feb 4 14:24:10.609 EST: FIBfwd-proc: Default:255.255.255.255/32 receive entry
Feb 4 14:24:10.609 EST: FIBipv4-packet-proc: packet routing failed
the ip helper-address <dhcp server IP address> command is present on all VLANs it is required on. DHCP server is enabled with properly configured pools. As, said, it was all working just fine, now it isn't.
Any ideas? I really appreciate any help.
What was the last thing that was done? An acl on vlan 311 possibly?
*** Please rate all useful posts ***
No changes to the ACL. This switch is our Core, and other than changing a monitor session every now and again nothing is touched. The ACLs haven't been modified in months prior to this event. I have since added a small ACL for debugging purposes.
From an address in vlan 311 can you ping the IP address that is specified in the helper address command for vlan 311? This would verify that it is not an issue in forwarding to the server address.
You mention that there are several VLANs and give us specifics of failure on vlan 311. Are other VLANs working or does the failure impact all VLANs?
Based on DHCP lease expiration dates, it appears that this trouble actually started last week after a series of power outages in quick succession. And thoughts as to the possibility power could cause this even when the switch is on UPS and generator power, considering multiple shifts between AC->UPS->Generator?
Have you checked the DHCP server itself to see if it is reporting issues with the scopes ?
I'm assuming th DHCP server is in a vlan connected to the 4500 and that you can ping it ?
Based on the possibility that it is related to a series of instability associated with power fluctuation then I would advise trying a power cycle of the switch. Something is hung or is out of sync and power cycle will frequently resolve this kind of issue.
Jon: Yes, the DHCP server is up and fine handing out addresses to those on the same network as it is.
Richard: We have scheduled a full power down with a 5 minute wait to clear and capacitors. Was holding out hope for a different solution, but I myself am fairly convinced that this is the only way. Hope does spring eternal though.
I am glad when hope springs eternal. And there is no harm in holding hope for a different solution. But I am glad that you are also moving toward a power cycle.