Slow DHCP Responses

[Gruher, Joseph R]

I have CentOS 6.4 and all available updates installed to my DHCP server.  Hardware should be way more than adequate, E3 Xeon with minimal utilization, 32GB RAM, 4x HDD in RAID10, server is pretty much idle so nothing should be competing for hardware resources.  My dhcpd.conf is simple, just one scope and 40-ish host entries, and I'm also using some simple "if" logic to hand out PXE and iSCSI information.  The dhcpd.conf file is attached if list allows attachments.  I'm on a closed network, just a lab environment with a single switch and the server and about 40 clients.

My problem is DHCP requests take a very long time, generally around 30-40 seconds.  This is consistent across different clients and across different sources of the DHCP request (Linux, Windows, BIOS PXE, iPXE).  They do always complete eventually but it seems like they should only take far, far less time.  I'm at a bit of a loss how to proceed on debugging the problem.  Any suggestions to point me in the right direction?

Thanks!

-Joe
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.isc.org/pipermail/dhcp-users/attachments/20130807/b46f6eaa/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: dhcpd.conf
Type: application/octet-stream
Size: 12500 bytes
Desc: dhcpd.conf
URL: <https://lists.isc.org/pipermail/dhcp-users/attachments/20130807/b46f6eaa/attachment-0001.obj>