The default gateway would be configured as a vlan interface on the access-layer switches. Then the uplinks would be configured as layer-3 routed interfaces. You could also use multiple routing-instances to keep traffic segmented back to the core from the edge switches. A simple config would look like this:
#Access Switch config
interfaces {
ge-0/0/0 {
unit 0 {
family ethernet-switching;
}
}
ge-0/0/1 {
unit 0 {
family ethernet-switching;
}
}
ge-0/1/0 {
vlan-tagging;
unit 500 {
vlan-id 500;
family inet {
address 10.1.1.1/31;
}
}
}
lo0 {
unit 500 {
family inet {
address 10.1.2.1/32;
}
}
}
vlan {
unit 100 {
family inet {
address 192.168.100.1/24;
}
}
}
}
protocols {
ospf {
area 0.0.0.0 {
interface lo0.500;
interface ge-0/1/0.500 {
bfd-liveness-detection {
minimum-interval 100;
multiplier 3;
full-neighbors-only;
}
}
interface vlan.100 {
passive;
}
}
}
}
vlans {
corp_vlan {
vlan-id 100;
interface {
ge-0/0/0.0;
ge-0/0/1.0;
}
l3-interface vlan.100;
}
}
#Core Switch Config
interfaces {
ge-0/1/0 {
vlan-tagging;
unit 500 {
vlan-id 500;
family inet {
address 10.1.1.0/31;
}
}
}
lo0 {
unit 500 {
family inet {
address 10.1.2.0/32;
}
}
}
}
protocols {
ospf {
area 0.0.0.0 {
interface lo0.500;
interface ge-0/1/0.500 {
bfd-liveness-detection {
minimum-interval 100;
multiplier 3;
full-neighbors-only;
}
}
}
}
}
Of course you would need to inject default route into the core switch from somewhere (your firewall perhaps?)
Ron