Bridge/router and method for avoiding packet replication during a failure in a network
A bridge/router and method are described herein that are capable of avoiding packet replication at layer 2 where there is a failure in a network. In one embodiment, the bridge/router is informed about a fault which occurred within the network. Then, the bridge/router makes a determination about whether or not multiple LSPs would be used to flood/carry the same IP traffic out off a single port. If yes, the first bridge/router chooses one of the LSPs to carry the IP traffic out the single port and then prevents the IP traffic from being carried on the remaining LSP(s) out the single port which avoids the packet replication on a link which is associated with the single port. If no, then the first bridge/router allows the IP traffic to be carried on the LSPs out off more than one port.
1. A method implemented within a bridge/router for avoiding packet replication when there is a failure in a network, said method comprising the steps of:
determining by the bridge/router if multiple label switch paths would be used to carry Internet Protocol (IP) traffic out of a single port of the bridge/router when there is a failure in the network; and
if yes, choosing one of the label switch paths to carry the IP traffic out the single port and then preventing the IP traffic from being carried on the remaining label switch path(s) out the single port which avoids the packet replication on a link which is associated with the single port;
if no, allowing the IP traffic to be carried on the label switch paths.
2. The method of claim 1 , wherein said determining step further includes accessing at least one forwarding database to determine if multiple label switch paths are to be used to carry the IP traffic out of the single port.
3. The method of claim 1 , wherein if a topology of said network is known then said choosing step further includes applying hard-coded rules to choose the label switch path which is going be used to carry the IP traffic out of the single port.
4. The method of claim 1 , wherein if a topology of said network is not known then said choosing step further includes:
snooping messages transmitted from a plurality of routers;
determining which router from the plurality of routers is a designated router; and
choosing the label switch path which leads to the designated router to be the one to carry the IP traffic out of the single port.
5. A bridge/router, comprising:
a processor;
a memory; and
instructions which are accessible from said memory and processable by said processor to avoid packet replication when there is a failure in a network by:
determining if multiple label switch paths would be used to carry Internet Protocol (IP) traffic out of a single port when there is a failure in the network; and
if yes, choosing one of the label switch paths to carry the IP traffic out the single port and then preventing the IP traffic from being carried on the remaining label switch path(s) out the single port which avoids the packet replication on a link which is associated with the single port;
if no, allowing the IP traffic to be carried on the label switch paths.
6. The bridge/router of claim 5 , wherein said processor facilitates said determining operation by accessing at least one forwarding database to determine if multiple label switch paths are to be used to carry the IP traffic out of the single port.
7. The bridge/router of claim 5 , wherein if a topology of said network is known then said processor facilitates said choosing operation by applying hard-coded rules to choose the label switch path which is going to be used to carry the IP traffic out of the single port.
8. The bridge/router of claim 5 , wherein if a topology of said network is not known then said processor facilitates said choosing operation by:
snooping messages transmitted from a plurality of routers;
determining which router from the plurality of routers is a designated router; and
choosing the label switch path which leads to the designated router to be the one to carry the IP traffic out of the single port.
9. A video hub office, comprising:
a first bridge/router;
a second bridge/router;
a first router; and
a second router;
wherein each of said first bridge/router and said second bridge/router avoid packet replication during a failure in a network by determining if multiple label switch paths (LSPs) would be used to carry Internet Protocol (IP) traffic out of a single port when there is a failure in the network; and
if yes, choosing one of the label paths to carry the IP traffic out the single port and then preventing the IP traffic from being carried on the remaining label switch path(s) out the single port which avoids the packet replication on a link which is associated with the single port; and
if no, allowing the IP traffic to be carried on the label switch paths.
10. The video hub office of claim 9 , wherein each of said first bridge/router and said second bridge/router facilitates said determining step by accessing at least one forwarding database to determine if multiple label switch paths are to be used to carry the IP traffic out of the single port.
11. The video hub office of claim 9 , wherein each of said first bridge/router and said second bridge/router facilitates said choosing step by applying hard-coded rules to choose the label switch path which will be used to carry the IP traffic out of the single port.
12. The video hub office of claim 9 , wherein each of said first bridge/router and said second bridge/router facilitates said choosing operation by:
snooping messages transmitted from a plurality of routers;
determining which router from the plurality of routers is a designated router; and
choosing the label switch path which leads to the designated router to be the one to carry the IP traffic out of the single port.