DHCP Relay - Users unable to get IP address until I delete and re-add the servers

[s0lselcia]

There are plenty of IP addresses and not that many users, and sometimes the users are unable to get an IP address, but when I delete and re-add the servers in the DCHP relay, they are able to get IP addresses again, this happens frequently on and off.

[Mumford]

Hello! 

Apologies if I misunderstood your post, but is your firewall hosting the DHCP server? If it is, is the DHCP set to auto or enabled mode?
We had a problem with our DHCP when it was set to auto where it was detecting another DHCP server leasing IPs which would cause our firewall to stop handing out its own leases. We temporarily changed it to enable until we found out what on our network was broadcasting DHCP services. 

I hope this helps!

[s0lselcia]

Hi @Mumford 

Thanks for your response. The DHCP is hosted on a windows server and the clients are on a different subnet, the DHCP relay is set up on the firewall, to forwards ip addresses to those clients.

[BPry]

@s0lselcia,

Do you see anything when you do a system logs search for (subtype eq dhcp)? 

[kelkins]

Hi,

We face the same issue whenever there is a network event.  We are running version 11.0.2-h2 on models PA-3420, 3410, and 1410.  If an internet circuit blips, and causes a routing change, or even when we do software upgrades on the access switches, this will cause DHCP relay failures.  We host our DHCP servers in AWS.  We have been deleting all the DHCP relay server entries, then adding them back, in order to resolve the issue.  I think next time this occurs, we will try restarting the routing daemon via cli commands, at least until we figure out this bug.

These routing events show up in the system log, after the DHCP relay settings are changed.

[TheJourney]

We experienced this issue today and in the past on 1420's and now 440s running Software Version 11.1.4-h1. Anyway to fix this? Is there a bug ID or anything?