<Jeremy's IT Lab> Day 26: OSPF (part 1)

OSPF (part 1)

 

OSPF (Open Shortest Path First)

-> Routers independently use map calculated as the best routes

-> Dijkstra's algorithm (다익스트라 알고리즘)

-> v1 : old / v2: IPv4 (common used) / v3: IPv6

-> Store info about NW in LSAs (Link State Advertisements) organized in structure LSDB (Link State database)

-> Router flood LSAs until develop the same map of NW(LSDB)

 

LSA flooding :

-> tell its negibors about new NW

-> flood sharing the same LSDB

 

Areas (divide up NW):

-> small NW be single-area

-> larger NW be single-area would be negetive effect

  <prob>

   - SPF takes more time to cal / processing power 

   - larger LSDB needs more mem

   - small change occurs every router's flooding it

  <sol>

   - divide into smaller area

Terms

1. Area: share same LSDB

2. backbone area: must connect to all area

3. internal routers: routers inside the area

4. ABRs(Area border routers): covered multi areas

5. backbone router: all routers in the backbone area

6. intra-area route: transmission in the same area

7. interarea route: transmission out of the area

8. Autonomous system boundary router (ASBR): connect to the internet

 

(config) router ospf 1[process ID]

* process ID (!= RIP, should match ID in the same RIP)

logically significant (diff ID can become ISPF neigbor)

 

Router ID priority (same as EIGRP)

1) manual

2) loopback int

3) physical int