Hello Sunil,
OSPF uses IP MTU to peer neighborship and it cannot be set per interface. If the two interfaces have different MTU values configured, the OSPF between these interfaces will not come up (Default behaviour).
The formula for calculating the IP MTU is as follows:
IP MTU = Media MTU - Ethernet Header Length (14 bytes)
For example, if the max-frame-size = 1514, then the IP MTU = 1500.
However, the mismatch could occur if you are using VLANs:
Ethernet header sizes:
Standard = 14 bytes
Vlan tagging = 14+4 = 18 Bytes
Flexible-Vlan tagging = 18 + 4 = 22 Bytes
In the Junos CLI, you can set the interface MTU in the range of below values.
root@re0# set interfaces ge-2/2/2 mtu ?
Possible completions:
<mtu> Maximum transmit packet size (256..16000)
I'm not very sure if there could be any scenario where the DBD MTU value can be set to 0 (as there would be an interface MTU value at peer end), however upto my understanding, if there is an OSPF MTU mismatch, the neighbourship wouldn't come up and it would be stuck in exstart/exchange state.
This is also defined in RFC2328
10.6. Receiving Database Description Packets
.....
If the Interface MTU field in the Database Description packet indicates an IP datagram size that is larger than the router can accept on the receiving interface without fragmentation, the Database Description packet is rejected.
However the Cisco's command brings in proprietary behavior which is not suggested by RFC2328. Juniper does not support this behavior and does not have any special configuration to achieve such proprietry behavior.
Thanks,
Vishaal