CentOS 7 replies to ARP with wrong MAC address - networking

Question

I've ran into weird issue recently, with my CentOS machine (lets call it server). Server has the IP address 10.150.39.5. The thing is it replies to ARP requests with two MAC addresses, and the last one that comes in breaks all the communication to the server.
Output from tcpdump on mirrored switchport:
18:02:01.388595 00:00:00:e5:64:32 > 00:00:00:67:1d:ea, ethertype ARP (0x0806), length 60: Request who-has 10.150.39.5 (00:00:00:67:1d:ea) tell 10.150.39.116, length 46
18:02:01.388600 00:00:00:67:1d:ea > 00:00:00:e5:64:32, ethertype ARP (0x0806), length 60: Reply 10.150.39.5 is-at 00:00:00:67:1d:ea, length 46
18:02:01.388743 00:00:00:67:1d:ec > 00:00:00:e5:64:32, ethertype ARP (0x0806), length 60: Reply 10.150.39.5 is-at 00:00:00:67:1d:ec, length 46
The most interesting thing is when I make the capture on servers NIC - I see only the first two packets (normal ARP req and ARP resp). So I've made a network SPAN (mirrored all the traffic from the switch interface where server is attached to other machine), on that mirrored port I see the packets I've pasted earlier.
I'm 100% sure that it is not MiTM attack, as it is one physical server attached to the interface and after I've implemented port-security on that port (allowed only correct MAC to come in) the issue is gone. So basically I do not have the issue anymore, but I'm very keen to find what's the reason for that.
Following are the couple of the outputs that might come in handy for you guys trying to help me:
[root#server ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: enp0s25: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:00:00:67:1d:ea brd ff:ff:ff:ff:ff:ff
inet 10.150.39.5/24 brd 10.150.39.255 scope global enp0s25
valid_lft forever preferred_lft forever
inet6 fe80::219:d1ff:fe67:1dea/64 scope link
valid_lft forever preferred_lft forever
3: enp1s0f0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:00:00:e3:ca:82 brd ff:ff:ff:ff:ff:ff
4: enp1s0f1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:00:00:e3:ca:83 brd ff:ff:ff:ff:ff:ff
[root#server ~]# tail -vn +1 /etc/sysconfig/network-scripts/ifcfg-*
==> /etc/sysconfig/network-scripts/ifcfg-enp0s25 <==
TYPE="Ethernet"
BOOTPROTO="static"
UUID="a04fb9bd-0543-4ba6-bd17-72cc3d9f54cc"
DEVICE="enp0s25"
ONBOOT="yes"
IPADDR=10.150.39.5
NETMASK=255.255.255.0
GATEWAY=10.150.39.1
==> /etc/sysconfig/network-scripts/ifcfg-enp1s0f0 <==
TYPE=Ethernet
PROXY_METHOD=none
BROWSER_ONLY=no
BOOTPROTO=dhcp
DEFROUTE=yes
IPV4_FAILURE_FATAL=no
NAME=enp1s0f0
UUID=64e7544d-54b9-40cc-83f0-7e10acbcdeaa
DEVICE=enp1s0f0
ONBOOT=no
==> /etc/sysconfig/network-scripts/ifcfg-enp1s0f1 <==
TYPE=Ethernet
PROXY_METHOD=none
BROWSER_ONLY=no
BOOTPROTO=dhcp
DEFROUTE=yes
IPV4_FAILURE_FATAL=no
NAME=enp1s0f1
UUID=c3085e08-1e14-4098-b77a-b512a2c99e75
DEVICE=enp1s0f1
ONBOOT=no
==> /etc/sysconfig/network-scripts/ifcfg-lo <==
DEVICE=lo
IPADDR=127.0.0.1
NETMASK=255.0.0.0
NETWORK=127.0.0.0
# If you're having problems with gated making 127.0.0.0/8 a martian,
# you can change this to something else (255.255.255.255, for example)
BROADCAST=127.255.255.255
ONBOOT=yes
NAME=loopback

Answer 1

The two MAC addresses 00:00:00:67:1d:ea and 00:00:00:67:1d:ec are almost consecutive (I assume you've obfuscated the first three octets as 00:00:00is no valid OUI) - this indicates a multi-port NIC, multiple logical interfaces or such. Check the switch's MAC table for the port the "wrong" MAC originates from.
If you've bonded/teamed multiple interfaces you have to make sure everything's working as intended. The switch ports need to be set up accordingly with single-MAC bonding.