Dynamic host-configuration protocol (DHCP) has lots of advantages, together with saving time by assigning IP addresses and different attributes to networked units relatively than IT professionals having to do it manually.
Generally, although, issues come up that eat up time otherwise. That is one such case affecting Cisco Catalyst 6500 and 9600 Layer 3 chassis switches used as distribution switches for our community, with completely different teams of buildings linked to them.
DHCP is often deployed on a server to ship IP addresses, subnet masks, default gateways, and DNS server data, however DHCP will also be deployed on switches and routers together with Cisco’s, and that’s one technique utilized in our community.
Particularly, the DHCP server capabilities of our Cisco switches are used to distribute IP addresses to units on our wi-fi community as a way of segmenting wi-fi site visitors from the wired infrastructure, which makes use of a separate DHCP server. An added good thing about utilizing the DHCP assist on switches, particularly in smaller networks, is the price financial savings that may be realized by having the change carry out double obligation versus buying a separate server for DHCP and having to handle it.
The Cisco 6500 had served the community in these capacities for a few years, however as our group moved towards adopting software-defined networking (SDN), it was time to improve to the 6500’s successor, the Cisco 9600, which helps automation and better ports speeds.
A difficulty with each switches was that the DHCP tackle swimming pools would freeze. Gadgets making an attempt to affix the community had been unable to as a result of they didn’t obtain IP addresses, which led to finish customers submitting bother tickets stating that the wi-fi community was down.
The switches can ship details about the accessible DHCP IP addresses utilizing the command “present ip dhcp pool”, which returns a show that appears like this:
Router# present ip dhcp pool 1
Pool 1:
Utilization mark (excessive/low) : 85 / 15
Subnet dimension (first/subsequent) : 24 / 24 (autogrow)
VRF title : abc
Complete addresses : 28
Leased addresses : 11
Pending occasion : none
Present index IP tackle vary Leased addresses
10.1.1.12 10.1.1.1 - 10.1.1.14 11
10.1.1.17 10.1.1.17 - 10.1.1.30 0
Interface Ethernet0/0 tackle task
10.1.1.1 255.255.255.248
10.1.1.17 255.255.255.248 secondary
Over time, our community engineers seen that when customers had bother getting IP addresses, the readout for the “present ip dhcp pool” command revealed that the present index was 0.0.0.0 but additionally confirmed that there have been nonetheless addresses accessible within the pool. That may look one thing like this:
Present index IP tackle vary Leased/Excluded/Complete
0.0.0.0 172.30.52.97 - 172.30.53.128 0 / 7 / 30
The swimming pools had labored nicely once they had been arrange about right here years in the past, then the issue cropped up periodically with no obvious triggering occasion. It wouldn’t have an effect on all of the switches without delay, however relatively sporadically all through the community. Throughout this era extra buildings and areas had been added to the wi-fi community, and extra units had been connecting.
The issue was frequent sufficient that each time the assistance desk reached out to us about bother connecting to the wi-fi community, our go-to troubleshooting steps had been to confirm that the Wi-Fi entry factors had been up and to verify that the DHCP pool wasn’t frozen. If it was, we’d observe the recommendation of the Cisco Technical Help Middle (TAC) and delete the DHCP pool and re-add it. This may reset the pool, and DHCP would start handing out addresses once more. That may appropriate the issue for some time.
This workaround would take a matter of seconds as we had copied and pasted the delete and readd instructions into the change with a view to save time.
Many occasions we needed to take away the swimming pools and re-add them on the 6500s, and we had been wanting ahead to the substitute 9600s to see if they’d resolve the difficulty. They didn’t.
The difficulty continued to happen, not day-after-day, however from time to time, normally reported as a community outage. Different organizations have posted related points that they solved utilizing the identical workaround, so coping with it’s simply a part of utilizing the switches
Regardless of this bug, DHCP on the switches does serve the purpose of segmenting the wi-fi community and, even with the periodic must re-add the DHCP swimming pools, is rather more environment friendly than the choice of constructing DHCP a guide course of.
Copyright © 2022 IDG Communications, Inc.
