3750 DHCP Pool Exhaustion Issue

[jason.henderson]

Hi all,

I've ran into an issue with a dhcp pool on a WS-C3750-48P running  c3750-ipservicesk9-mz.122-50.SE2.  The UserVoice2 pool (see below) has  run out of addresses but should still have 10 left - debug ip dhcp  server events shows clients trying to get a lease and being knocked back  because the pool is exhausted.  I've included the config from another  pool called UserVoice1, which is set up identically and is working ok.

Has  anyone seen this issue before?  I can't see anything relevant on the  bug toolkit.

Thanks in advance,

Jason.

ip dhcp excluded-address  10.153.28.126
ip dhcp excluded-address 10.153.28.254
ip dhcp  excluded-address 10.153.28.200
!
ip dhcp pool UserVoice1
    network 10.153.28.64 255.255.255.192
   default-router 10.153.28.126
!
ip  dhcp pool UserVoice2
   network 10.153.28.192 255.255.255.192
    default-router 10.153.28.254
!


#sh ip dhcp pool UserVoice1

Pool  UserVoice1 :
Utilization mark (high/low)    : 100 / 0
Subnet  size (first/next)       : 0 / 0
Total addresses                : 62
Leased  addresses               : 54
Excluded addresses             : 1
Pending  event                  : none
1 subnet is currently in the pool :
Current  index        IP address range                    Leased/Excluded/Total
10.153.28.114         10.153.28.65     - 10.153.28.126     54    / 1     / 62


#sh  ip dhcp pool UserVoice2

Pool UserVoice2 :
Utilization mark  (high/low)    : 100 / 0
Subnet size (first/next)       : 0 / 0
Total  addresses                : 62
Leased addresses               : 50
Excluded  addresses             : 12
Pending event                  : none
1  subnet is currently in the pool :
Current index        IP address  range                    Leased/Excluded/Total
10.153.28.193         10.153.28.193    - 10.153.28.254     50    / 12    / 62

[Reza Sharifi]

Hi Jason,

Is there any other command you my have in your config that excluded more addresses?  UserVoice2 shows you excluded 12 but your config shows only 2

Reza

[jason.henderson]

Hi Reza,

No, there are only 2 addresses excluded from the pool, which is why i've been thinking there may be a bug in the IOS.  I'm not in a position to delete and recreate the pool unfortunately, so i don't know if that would have an effect.

Thanks,

Jason.

[Reza Sharifi]

Hi Jason,

If you can't delete and recreate the pool, may be an after hour reboot can cause the switch to release those 10 IPs.

The other long term solution could be looking at upgradeing the IOS.  I think the latest on CCO is 12.2(53).

HTH

Reza

[engineer_msu]

Dear,

Try using the command "no ip dhcp conflict logging" this may resolve the issue. You can take the reference of the document http://www.cisco.com/en/US/tech/tk648/tk361/technologies_tech_note09186a00800f0804.shtml

[Roger De Couto]

Issue resolved with 'no ip dhcp conflict logging' command - http://www.cisco.com/c/en/us/support/docs/ip/dynamic-address-allocation-resolution/27470-100.html