Current Events > That weird network issue isn't turning out to be so simple...

(+) Yeah! (+)

We have about 400 branch locations and one of them can't communicate with our cluster of DHCP servers in our data centers.

We've made some progress in isolating where the problem is, at least. The problem started after we upgraded the software on our head-end sdwan firewalls. Our Splunk logs show a spike in DHCP requests from that office right after that happened, and the DHCP requests dropped off after I made that 1 branch location firewall a local DHCP server.

This particular type of branch location only has a connection into our data centers through 1 of our 2 head-end firewalls, so I'm assuming it would affect any site of this type that we set up right now. This is the only one that is online at the moment, though.

The DHCP requests are coming from the device, hitting the branch firewall, and then hitting the head-end firewall and I believe they are dying there. We did some captures on the DHCP server and on the firewall. Pings go all the way back and forth as expected. DHCP relay requests do not.

So now I'm working with vendor support. I also ran into a BGP bug while troubleshooting this bug... Sigh...