TrainSignal Chapter 5, Lecture 3 - Total Stub Areas and NSSA

Just alittle quick extra carry over thang, from the last lab, notice the default route we have for Area34 has a cost of 11, 10 being the cost of the ethernet segment and 1 the default for the default route, well you can change that my friends:

So we should now see a cost of 30, 10 for the ethernet segment and the default cost of 20 that we specified above:

yeah boi! that works a treat, right ... lets get to work:

Lets jump on R3 and make area 34 a totally stub area (we dont need it on R4)

The no-summary option added to the ABR of the stub area will make this a total stub area, which prevents LSA types 3,4 and 5 flooding into the area.

The ABR is the only router that needs the no-summary option enabled. no-summary doesnt have to be added to the other routers in the area but they still have to be configured as a stub.

awesome, check out that routing table now, lets test connectivity to them loopbacks etc:

all good still :0)

REMEMBER

The stub bit flag ahas to match for the adjacency to form/stay up, so it is fine to have one end configured as a stub and one a total stub as the bit is set, HOWEVER if the routers in the area all did not agree on this then we would have problems.

 

HINT: you need to fully remove the stub configuration not just the stub area no summary!

 pretty kool huh, but krazy name "not-so-stubby total stub" - not as common as the other stubs though, but they are out there!

TrainSignal Chapter 5, Lecture 2 - Redistribution and Stub Area Config

The Lab Setup:

Router5's config:

Lets Check R1 and see what the big picture is here:

R1's routing table:

Lets check we have connectivity with them RIP routes (loopbacks on R5)

Lets see what R3 see's

 Naturally he does not see any of the RIP routes as he is not running RIP and we have not configured any re-distribution ......yet

R5 see's ... well nothing lol, so lets gets some RIP on!

A SIDE NOTE ABOUT REMEMBERING TO REDISTRIBUTE THE DIRECTLY CONNECTED NETWORKS!!!

Lets double check we can actually ping these RIP routes (as remember if we had not redistributed the CONNECTED networks the next hop would not be valid and the pings would fail)

Sorted, and we can even ping the ethernet segement on R4 (the furthest route from ourselfs) :0)

Lets check what R4 now see's:

Ah so notice the "OSPF External Routes" which are the re-distributed RIP routes and the DEFAULT SEED METRIC OF 20! - for RIP we had to specify a SEED METRIC

and yup, we can ping em:

 We lost the adjacency, if we run a debug we can see the stub bit set is not agreeing with all OSPF routers...

Now the adjacency comes back up :0)

Area 34 is now a stub Area:

With that simple config, the size of the routing table has been halfed, the E2 routes have been replaced with a single default route...

Next lets try and sort out them IA routes as they all still have the same next hop address ;0)

REBOOK

Rebooked exam for 31st July (was booked for 28th June) but iv fallen off track with bla bla bla excuses here .... so, time to get back on the coffee fuelled cisco train and full speed ahead with ROUTE

Trainsignal - Chapter 5 - Advanced OSPF lesson 1 - Multi-Area OSPF & Summarisation

So looking at this lab, we can already tell 3 things:

The hub is going to need neighbour statements

The hub HAS TO COME THE Designated Router (DR)

The spokes must not take part in the DR and BDR election

*another note, remember the BROADCAST option on the frame setup ;0)

and start config on the spokes first as this is where we set the OSPF priority (to take them out of the election process)

Some handy troubleshooting/commands

Lets do that next

Trainsignal, Chapter 4 part 5 Broadcast with Multiple Routers

Notice from R2 it shows a 2-WAY relationship with R1;

More to come in the advanced OSPF section which i will blast friday night