I have a GLBP config consisting of two Cat4506e switches, connecting over 10G fiber to Cat3560E switches. Testing, I have a laptop connected to one of the 3560 switches with a constant ping to the Virtual IP addresses of the GLBP on two different VLANs, and the switches themselves. Both 4500s are configured the same (as seen below). When I reload the .252 switch, I do not lose a packet, but when I reload the .253 switch, I lose about 6 seconds of ICMP. It looks like when the 10G interface comes up, it puts all switchports in a blocking mode. I checked the int. when testing, but it shows it as up/up. Any ideas?

Switch 1 (.252)

interface TenGigabitEthernet2/1
switchport mode trunk
!
interface TenGigabitEthernet2/2
switchport mode trunk
!

interface Vlan1
ip address 172.31.7.252 255.255.248.0
glbp 1 ip 172.31.7.250
glbp 1 priority 110
glbp 1 preempt
glbp 1 authentication md5 key-string th1sk3y
!
interface Vlan20
ip address 192.168.20.2 255.255.255.0
glbp 20 ip 192.168.20.1
glbp 20 priority 110
glbp 20 preempt
glbp 20 authentication md5 key-string th1sk3y

Swtich 2 (.253)

interface TenGigabitEthernet2/1
switchport mode trunk
!
interface TenGigabitEthernet2/2
switchport mode trunk
!

interface Vlan1
ip address 172.31.7.253 255.255.248.0
glbp 1 ip 172.31.7.250
glbp 1 priority 110
glbp 1 preempt
glbp 1 authentication md5 key-string th1sk3y
!
interface Vlan20
ip address 192.168.20.3 255.255.255.0
glbp 20 ip 192.168.20.1
glbp 20 priority 110
glbp 20 preempt
glbp 20 authentication md5 key-string th1sk3y

The 3650 switches - both conf. like this

interface TenGigabitEthernet0/1
switchport trunk encapsulation dot1q
switchport mode trunk
!
interface TenGigabitEthernet0/2
switchport trunk encapsulation dot1q
switchport mode trunk