You should check the bindings of the dhcp relay to confirm if the core is receiving the requests from hosts. If so, you can move to the server to confirm if it is receiving the dhcp packets from the relay.
If the dhcp relay is the default gateway for the subnet with issues and you have connectivity between the core and access switches, it is required to confirm where the dhcp packets are getting dropped.
Is there any known differences in equipment between users? => for switch A if operating a pure L2, none
One known issue is that Apple devices and Windows devices use DHCP => this is handled by relay agent and/or DHCP Server
BTW, have you checked that the DHCP Scope on your DHCP Server is large enough to provide enough IP address for all of your devices? I assume if you hard code an IP address on any clients/devices in proper subnet, that the device then works fine?
I believe your situation has nothing to do with L2 switch unless DHCP Snooping is enabled. I would suggest you turn off all port level security options (DHCP Snooping, IPSG. etc.) and see if this helps. This would at minimum narrow down where to look.
I would suggest you to run the following command and check whether the packets are sent/received via that interface connected to Cisco Switch - user@host> monitor traffic interface <int-name.unit-num> no-resolve size 1500
Check for any errors on that particular interface - show interface <int-name.unit-num> extensive | match errors <<< Run 3-4 times
Thanks, π00bm@$t€®. Please, Mark My Solution Accepted if it Helped, Kudos are Appreciated too!!!