CCNA Training - EIGRP Configuration

[Music] [Music] all right what's going on everyone happy Tuesday night getting ready to do some CCNA eigrp configuration let's get ready to go here so let's just go ahead and pull up hey Jake thanks for joining in Nate good to see you let's let's take a look at the agenda die so we're going to be talking a little bit when we're walking all the way three idrp configurations so last week we did you at OSP F this week we're having some fun with the IG RP you probably go a little bit beyond CCNA tonight we're gonna be drilling into the topology table hope nobody minds going a little too deep hopefully it'll be a good good conversation so definitely chime in with your questions as we dive into EIGRP in a little more detail than maybe what the CCNA expects us to know so like CCNA nice I like that we are gonna be talking about the metrics to comparing it to OSPF cost a GRPs metric is way more complicated than OSPF son fortunately so we need to make sure we understand that as well and then summarization is something that we can't do in single area OSPF but we can do with the edge erp so we're gonna of course take advantage of that tonight let's go ahead and pull up our topology here how do we actually you know what tell you what we're just gonna go straight here so this should be taking a look at the router configuration you've got the topology right there in the corner I tried to pretty up the document a little bit so hopefully you can see it a little bit better when you full-screen if for those who have blogs or full screen I suppose hey Gus good to see you thanks for joining excellent we're just getting started here so ideally what we're going to do tonight is just start to bring up our convergence and we'll just kind of go from there so hold up here what's going on nothing there there we go okay I think we're almost good what is happening here there we go okay now we're good alright so based on this topology which I've got a printout right in front of me because that way I don't have to keep trying to manage Licking over at other screens and such let's go ahead and just focus on what yeah ARP configuration looks like / router basis so well go to router one here in the top left of this diagram we've got two pulled up here already and we're going to do a config T and do a router a IG RP question mark now right away we're greeted with the first difference between EIGRP and OSPF obviously under the hood there's a ton of differences but it's from a configuration perspective we need to make sure that we understand the autonomous systems matter with the edge erp OSPF the process ID is locally significant only so if I get on to a router you know we've last week of for those who are here last week we actually saw that because we got on to two different routers and we had like OSPF one here and OSPF I don't even remember like 300 or something that whatever we used and so they were mismatched but no problem they came right up because it's locally significant to those routers where as an autonomous system it's every router in the autonomous system and it allows us to make sure that we're in sync with well don't think of how to say this we could in theory deploy multiple autonomous systems with the add GRP have multiple routing domains no problem with OSPF we have a big problem with that if I have two routers talking to OSPF and i Nabal OSPF on that link they're gonna form neighbor ships if i have two OSPF domains i try to connect them it's gonna become one OSPF domain with the risk of segmenting area is zero if the two area zeros don't touch each other so oh s PF doesn't allow us to do that yet Europe he does I've never really seen it a good idea to have to ya GRP domains I have I have seen it in a network it did accomplish a purpose but there are other ways to accomplish what they were setting out to do and so generally speaking yeah I don't know I mean I'm sure there is a use case out there that the benefits from having two autonomous system numbers in your network big poppa good to see you by the way out of curiosity who is going to start streaming study sessions for encore yeah well you know I might have to do that myself here because all of you all who have been taking the CCNA have been passing the CCNA and so maybe it's just time to step this study group up to core level I'm I don't know if you're here right at the start but yeah GRP we are going into the encore level e IG RP a little bit here tonight delving a little bit more into the topology table so you can I hang around hopefully you'll be entertained so okay router EIGRP so let's go and get this fired up I know this is the basics but we got to get spun up right so we'll just choose autonomous system number 100 tonight and one thing that we talked about last week that we did not demonstrate is passive interface and let's be a little I want to be a little real tonight so we're talking like real world scenarios we should probably not probably best practice is certainly to enable passive interface default so we're going to do that on our router here passive interface default and remember what passive interface does in order to remember a passive interface does we need to back up and remember what the network statement does because I've tried to drill that home I've said it a hundred times by now so you what most of you probably understand it a network statement activates interfaces and yes one statement can activate many interfaces so anything I whatever I put in for my network statement if I put in a slash 24 for example every Network address that matches that 24 that slash 24 is going that those interfaces will be activated once that interface has been activated two things happen one the routing process will look at that subnet and say I'm going to advertise that into the routing protocol but the other thing it does is start sending hellos out saying I'm gonna find a neighbor out this interface and number one you might find a neighbor you didn't want to find a number two at a minimum you're sending multicast out out towards the network and nobody actually wants to listen to it probably not the worst thing in the world but at the same time let's just make our networks clean best practice this so passive interface to say it acted you know we're still dealing with an active interface we're still advertising that subnet into the network or into the routing process but we're no longer sending hellos out that interface so yeah so absolutely using passive interface default hopefully is something that you'll see out on network so but then of course we have to disable passive internet specimen face to fault maybe a little clearer what it's doing but it actually enables passive interface on every interface and of course we do want some Ito GRP neighbors so let's go ahead and get those turned out and we'll do a no passive interface on gig one zero and on gig to zero how's the drawing working out I tried to redo it today does it do the numbers need to be bigger or are they are they pretty good at this point I'm kind of curious give me it give me some feedback on that guy's all right in fact we're going to actually no let's go into our network statements next so a network statement we'll go ahead and do here's another best practice concept so we we usually want to use all zeros for our wild card masks if we remember that's because it will target one IP address which means by a very nature of things it's only going to target one interface we want to control which interfaces we're activating a routing process on so by choosing the IP the actual IP address of the interface using a wildcard mask of all zeros that will activate one and only one interface and leaves me in control of things so given this is the first router I'm not going to see a whole lot of activity in fact I should see zero activity as I bring activations up and I'm actually going to activate by the way this back-end Network 10.1 does 0.0 actually that one so that that big green cloud in the in the top left so for now this is it for router 1 I've done passive interface default I've disabled passive interface on the interfaces that will have EIGRP neighbors and I've activated the 3 interfaces that we see connecting into router 1 all right so I see it looks good looks good looks good ok good good I'm glad I'm glad it's I'm glad it's coming through all right it's hard to it's hard to make a drawing that everybody can see and while were configuring and make the text as you know that big - all right so anyways on to router 2 so let's go to router 2 and same thing we're gonna do a config T router yeah ARP one remember that has to match a passive interface default we can start to abbreviate this to be a little faster no passive int and we got what gig 1 0 we've got gig to 0 and then network statements so let's do Network let's do the network towards router want to make sure that everything is working like we think it will be so they reminder that the IP address belonging to the router will always be its router ID so every link on router 2 will end with that to try to keep it intuitive so that 10 dot 12.0.0 that we see at the top dot 2 for router 2 we'll hit enter we should see any edge erp neighbor come up and we do very good so we've established a neighbor good stuff if we do a show IP EIGRP neighbor will see it there in the list and life is looking good all right very good so let's go and activate the other interfaces let's do a network statement on 10.24.0.2 that should get our connection to router for with all zeros and the back end network which again would end with that too so we should be good there oh the serial interface I almost forgot that so no passive interface serial 3/2 and network on that is 10.12 0 dot i think i put a typo in there that should be 23 oh I I forgot this subnet originally on the drawing and I scrambled like literally five minutes before this thing started and I pulled it up and I copied and I pasted it down and I changed the color I forgot to change the subnet so yeah that should definitely be 10.20 3.0.2 oh well best laid plans all right one other thing I wanna have some fun with we touched on authentication last week with OSPF this is definitely not a CCNA concept I mean maybe the concept of authentication but configuring authentication eigrp is a little more cumbersome than it was in OSPF unfortunately so we have this concept of a keychain a keychain is effectively a list of authentication keys and I can use you know once I establish a keychain and maybe I have multiple keys so think passwords a key is a password so I've got multiple passwords in this keychain and then I can have keys come and go I could actually configure them to expire eventually so I could I could put keychain keep you know very similar key changes they need to be the same keychains on all my routers that actually progressively expire keys over time so it's a little more secure it's a little more fun that way but it's also a little more cumbersome so we've got to think through what that looks like and we'll just go ahead and walk through this example here let's make all that let's add authentication between routers 2 and 4 ok so before we even touch yet GRP configuration we have to configure that keychain so do keychain we'll just call it yeah ARP key yes that's all right all right so then we need to specify what key number we're going to create and you know again we can whatever the identifier is here we've got I don't know really quickly what is that how many bits is that I didn't I don't even know so it's a lot of bits so I'll just say key 1 keep it simple and then key - string this would be the identifier so are not identifiers are passwords so we're going to activate by the way md5 authentication on EA GRP remember OSPF has like the they have the three types at the type zero which is no authentication type one which is simple ie plain text and type two which is md5 leo GRP also has no authentication that's the default that's what we've been doing so far and it skips right up to type two not that they call it type two an eigrp but it skips right up to an md5 so there is no plaintext authentication eigrp which is nice so let's just call this I don't know yeah GRP I already called the keychain AIG RP I don't feel like I should I will say GRP pass how about that all right we're not going to configure expiration or anything like that okay so we've configured our keychain now and by the way keychains are used by more than EIGRP we can use keychains or well I don't know yeah I mean II trying to think of what else uses keychains there's there's a lot that uses it and I'm blanking so no chime into the chat if you know something that that uses keychains alright so in the meantime we have to get onto the interface that we're going to apply authentication to and one thing to keep in mind with this inspector protocols distance-vector there's a lot that can be done on a per interface basis summarization when I look at summarization later and that goes on the interface and that's because again I've stressed a few times VI GRP and distance-vector is routing by rumor it's literally I'm trusting what you're telling me if I'm your neighbor and you tell me something I believe you 100% and I have no way of validating it so I can lie to you and you can lie to me which is what summarization is right I mean summarization is saying I've got five routes behind me I'm not gonna tell you I have five routes behind me I'm gonna tell you I have one route behind me and you're going to believe me and that's why we can't do that with you with OSPF because OSPF here's so to speak it's probably not the best analogy but it's like I'm hearing things from everybody and everybody has to agree if it nobody if two people are telling me two different things the database breaks and OSPF breaks is relying on everybody having the exact same view of the network well with the edge ERP and distance-vector there is no view of the network there are simply I'm connected to you or neighbors and whatever you tell me is that is what I'm going to believe so again that's why we can summarize so a lot of times there's a lot of configuration on interfaces and what I want to show you here let's get on to gig let's do an ERP authentication between what I say routers to and for such would be gig one zero and this is interesting watch you'd think the command for anything as your P will be IP EIGRP like remember IP ospf was on the interface command you know for enabling OSPF on an interface so you think yeah IP eigrp except there is no IP EIGRP command and cisco had to probably bad choices or bad options available to one was to put everything behind eigrp and then duplicate that with rip because rip is a distance vector protocol and you've got all of these interface you know summarizations and split horizon and authentication all kinds of things that are interface level and they could have put all of that behind the edge ERP and duplicated it for rip and put it all behind like IP rip and instead what they chose was to say well we're gonna say IP authentication and then you do the question mark well actually sorry let's specify the keychain first keychain would I say EIGRP key keychain what we're doing here oh there it is all right so EIGRP now there isn't even a rip option for md5 authentication which is why it doesn't show up there but well I don't know maybe there it's that's right I can't control W with this lab environment all right let's do ten ocation mode yeah GRP all right there was I thought no I thought you could activate that viewer I guess there isn't Rob mp5 on rip which sounds familiar I just there was something I did earlier that showed that rip you know IP let's just do split horizon for example if I wanted to activate split horizon on this interface why isn't rip showing up oh it's because I don't have rod or if enabled hmm I'm confused by that I swear earlier I had rip on there so maybe maybe I'm just crazy but either way the the I've totally lost control of what I'm trying to say I the commands free edge ERP don't start with IP eigrp that's what I was trying to get at and my example fell apart so check it out we got to do on an IP authentication and then we saw two different things there we saw a keychain and then he had GRP and then we specify the autonomous system number and then we specify the name of the keychain which was the edge ERP e key okay not the password the keychain name and then I activate it by saying IP authentication mode and then we do EIGRP 100 md5 because that's and by the way just to show the question mark is our only option we can only enable md5 authentication so md5 now we have enabled authentication for EIGRP on this interface alright we'll do it again because we have to do it again on router for in order to bring up that connection so let's man let's get prior to three let's just go to router four and we're gonna say what we got here alright so we've got to do everything so config T router EIGRP 100 passive interface default because we're gonna do it the right way no passive interface on gig to zero and fast ethernet for one and we don't have a serial interface now we just gotta activate the networks 10.24.0.2 that that should not actually create a neighbor ship because we know router 2 is waiting for an authentication enabled and an authentication enabled hello and we don't we don't actually see that it's coming up so it's exactly what we'd expect we'll get to the authentication in a moment we'll activate 10 that 34 to 0 to 4 and will activate 10.40 does it for back in network ok so let's get that keychain created we'll do key chain yeah ARP whoops key this name could be different let's just call this key key - because it's a locally significant value I'll just go ahead and show that key one key - string now this has this it'd be the same right we add GRP passes what we had said now we'll get on to that interface gig to zero and do an IP authentication and then keychain the edge erp 100 and now we specify eh ERP key and we enter it should come up it should come up oh forgot one step I P authentication mode we got that we got to actually enable it so the keychain is on the interface but that we haven't actually enabled authentication so yeah RP 100 md5 there that's why I get for uh sticking my neck out there all right now it should come up interface too big to zero I activated it I did a no passive interface did I get the right IP address all right let's take a look here okay it's a GRP pass RP pass I want to make sure I got the same password Oh somebody's probably know okay that's my problem right there I threw myself off with the two so let's do this command again but add the actual - yeah now all right there we go real life all right so authentication is there so I wasn't smooth but hey again we got the job done again two commands we have to well I didn't even say that the first step is to create the keychain then we get onto the interface then there's two commands we have to enable the keychain itself and hopefully type in the right name for the keychain and then second of all we have to enable the mode okay all right so next router three config T almost done with the boring stuff although the authentication was actually pretty interesting no way passive int default no passive int gig 1/0 no passive int zero well we're gonna do the serial as well and now let's get those Network commands in there so Network 10.13 dot zero dot three we should see all of these neighbor ships come up as we type these in ten dot whoops that's my typo again 23.3 all zeroes is everybody clear and why we're doing all zeros if you're not clear then please raise your hand in some form in the chat I want to make sure that we're all good with that the wild card mass that's targeting a specific IP address and then we're gonna enable that back-end Network 10.30 at four oh no I did it for I wasn't paying attention three and now network 10.30 dot three all right and we'll go ahead and get rid of the bad Network statement that I put in there all right do show run by the way if you know aren't aware of using the pipe section command and routers it's really powerful you do section or SEC or whatever you want to breathe you section router it'll give you not just the line router which is good not really because what I really want is the whole section and again it really does give you that whole section which is nice so we've got the four Network statements we've got the three no passive interfaces and we should see that we have three neighbors and we do okay well I think our CI GRP topology is established at this point so let's go ahead and have some fun taking a look around to things by the way um I had my notes but is bright not much point demoing it so last week we demoed router IDs so just keep in mind router ID concept is the same within EIGRP it's it's not like it's interesting because righto SPF the OSPF database relies heavily on that router ID I mean eigrp does as well but generally speaking it follows the same path for configuring that router ID and you can always whoops where are we oh that's funny all right I don't know what's going on here I I'm just gonna leave it alone [Laughter] all right duh what was gonna show do oh yeah yeah router ID right ID this isn't it not yeah I don't know maybe it's not alright well we won't worry about that's what I get for going off-script big papa I'm liking the authentication stuff I have a lot of issues with buffering I'm good with the zeros okay good you're having issues with the video buffering like like the actual stream is is buffering that's no good I need to get a hardwired port and pulled in my office I've got a wireless access point right there but you know how those mesh access points work it's still doing a hop to get back to the main base so alright so hopefully hopefully the video is settling out here alright so yeah I think that's oh I know I think I know what's going on here yeah I'm trying to I'm just trying to get to my notes and my notes are all messed up for some reason but I think we're good here ok so let's go ahead and dive in you just keep hitting play okay good well as if anybody else is having issues let me know because I'd like to know if that's on my end obviously I want to make sure that the stream is good quality alright so let's talk about the topology table so I've mentioned OSPF database already quite a few times for being an erp stream the the you always have the you know the information that the routing protocol the routing process has under the hood so OSPF has a dos database and the the route the I'm sorry idrp has the topology table we're gonna pull that up here and we'll just pick router 3 why not it doesn't really matter at this point show IP yeah grpe topology okay so the topology is going to show us what's happening underneath the hood before we even drill into this if I were just do a show IP route Vig Rp just to clear out all the non eigrp routes we CEO grps referenced represented by d we do need to know that if you're studying for your CCNA administrative distant same thing we know is 90 here's this metric and again we can see that the metric is already not nearly as clean as I was PF most OSPF costs they call it a cost was in like maybe 10 to 100 range I mean they were they were realistic values and these are pretty values because we're using gig and gig Ethernet is pretty fast and so it keeps the metric ID down but it's not uncommon to see six or seven figures in this metric calculation so what we'll get to the metric how that metric is calculated after we talk about the topology table so yeah arp has liz let's go and pull that topology table back up nate things are good on your and thanks for confirming alright let's look at this topology table so we're looking at this at this what what this is really doing is collecting all of the route information and storing what i need to know even if i don't install routes into the routing table for example if i were to go to the routing table call it and enter a static route so i say ok well the route to this network while i'm non router 3 that's a bad example right here the route to this network here which is the network between routers 2 & 4 I'm just gonna enter a static route what could go in or a static route and that would be fine and and that would override ei GRP because the default administrative distance first attic route is 1 so assuming I left the default it's less than 90 it's going to override what EIGRP has but that doesn't pull the route out of the EIGRP table ya GRP needs this because what happens if that static route goes away maybe I have that static route tied to a condition that it you know it goes away or what have you and or hey maybe I just go and I delete the static route and so now we actually could use that route in the routing table and so the router is constantly looking to make sure that the routing table is built like it should be built based on all the information this is why by the way we have route like my routers could be running GRP and OSPF at the same time in fact cisco on some level especially if you go towards the CCIE level we talked about route redistribution we need to understand what that concept is even at the even at the CCNA level honestly taking routes from OSPF and advertising them into EIGRP and vice versa so that's right redistribution but if I have a router that's simply running EIGRP and OSPF how do I choose which routes to use I mean we we understand the administrative distance concept if I learn OSPF routes and EIGRP routes by default 90 is less than 110 and so I'm gonna install a GRP routes but I've still got all that information all SPF because what happens if my eigrp link goes down well then I'm gonna need to Oh SPF information to populate the routing table a big pop I'm sorry did I miss something regarding the reference bandwidth there Yeah right with the with OSPF yeah that was the trap that we laid last week right it's not even a trap it's it's just the default that OSPF defaults to 100 Meg being the maximum link that's considered which interestingly enough by the way yeah GRP is 10 gig and i need to look up and see if there's a way to change that yeah I don't know but yeah yeah GRP basically uses the same concept as a reference bandwidth except it starts with a tenth gig instead of a hundred Meg which is good that's better but I mean we're already over a hundred gigs so you'd think that we'd we'd want to fix that whoo ok so assuming we leave aside the metric calculation again we're gonna come back to this let's just take take these metrics these values at face value at this point so this router right here is learning these the this route right here 10.24.0.2 our diagram down below they're the ones in the corners that have three connections so they're actually connected to all three routers this is why we can have three connections on this so we actually see router one is advertising that network router for advertising that network router two is advertising that network and they're all advertising it with very different metrics so this is what we call we would take a look at the second number right here you'll notice that there's always a higher number followed by a lower number this is where it starts to get a little bit complicated with the ID European this is ccnp stuff okay so when we're talking understanding what's going on the topology table we need to understand this concept of reported distance or Rd and feasible distance FD okay so we see FD right here is listed for the entire route and so one time it says the feasible distance is 33 28 and we should see that that matches one of the FDS listed and we do it simple right here it's the route via router 1 so if I were to do a show IP route let's just do a show IP route eeeh GRP again 10.24.0.2 is the rider one okay and now we know why because if we scroll back to this topology table right here oops I scroll a little too far and let's just pull it back up this way okay so this route right here the lowest way for me to get there is via router one right here 33 28 is my ability to get there that's what the feasible distance is if I'm looking at the feasible distance it's me saying okay if I were to send it to router one the metric would be 33 28 if our descent it's a router for my metric would be 28 for 16 and if I were to send it to router 2 holy smokes it would be again remember I mentioned 7 digit numbers that's a seven digit number that's 2.1 million it would be 2.1 million for me to send it to router 3 so what's going on here well two things first of all this reported distance concept happens first what happens is all of these routers at first tell router this is router 3 so they tell router 3 this is how far it is how far away I am from this network okay that's the reported distance so router 1 says that I am 3072 away from this network ok router 4 says that I am 2816 away from this network and by the way router 2 says the same thing which makes sense because router 2 and router routers 2 and 4 they're actually connected to that Network we know that if we look at the diagram right routers 2 & 4 are on the edges of 10.24 so you think that they would be the best way to get there kind of odd that we're not choosing either one of those for our connection but there's a good reason for that so yeah I mean they're they're actually saying that they can get there better than router 1 and that's what we'd expect router one should be further away from that network than routers 2 & 4 so why are we sending it to router 1 well that's where the feasible distance comes into play the feasible distance is me saying okay I've got 3072 is what writer 1 is telling me but my wit my best path to router 1 is somewhere in the vicinity of 256 or so okay so I can get to rider 1 in 256 router 1 can get there in 3072 so if I'm 5 miles away from router 1 and router 1 is 10 miles away how far away am I well I'm 5 plus 10 15 miles away exact same thing here if router one's reported distance is 3000 and my ability to get to you is 256 then my feasible distance the distance for me to get to this network via router 1 is well in this case 33 28 okay BiPAP I've heard the reported distance also called advertised distance yes it could be the same thing I took moment to see they were the same number sorry about that or maybe you're saying that um in your studies earlier you were seeing that as well yeah I actually originally learned it as advertised distance I think one reason why potentially we like to go with reported distance is because administrative distance is also a thing and so we especially within the world of routing protocols and so if we had administrative distance and advertised distance and they're both 80 and they both end with distance then you know it's just one more thing to get confused about so but either what you you should be aware that they could be either one either way it's the same thing right there advertising it to me they're reporting to me they're distance okay so in our network this is where things get interesting because routers to again in 2 & 4 they're actually telling me that they're closer to the network and yet when I add my distance to them check out what happens this it goes from basically 2,800 to 28,000 so I don't know - 8 16 that would be 27 627 600 - 2,800 I don't know you know I mean it's a it's it's roughly probably 25,000 away at this point so if I'm roughly 25,000 away what's going on why am I 5,000 away well look at the topology that is a fast ethernet link and a fast ethernet link is one-tenth the bandwidth of a gig and so my distance - you know from router 3 to robber 4 is like 25,000 so I'm gonna add my 25,000 I'm gonna add it to the to the what what's being advertised the report and height the 20 20 and 16 so I add those two numbers together and I get this 28 for 16 so that is my distance to get to that network via router 4 and we're gonna do the exact same thing why in the world is this such a bad number because it's across a serial link and that's probably a 1.5 Meg serial link and the cost of the metric on that is going to be atrocious the bandwidth is low the delay is actually really high on that link and so you know 2.1 million that's where we get that from so at this point this is this is pittance this is that even 1% I mean I add the 21 16 to whatever you know two point one seven million and I basically get two point one seven million I notice maybe two point one six to two point one seven baby toe can you explain the ad again okay sorry about that I did gloss over that so and I'm not sure which ad you mean we have to a DS we have administrative distance the administrative distance is this concept of I might run five different routing protocols and all five of those writing protocols across different leaks links usually all five of those writing protocols might tell me that they can get to a network and I need to know which routing protocol I trust more and so administrative distance is a ranking system and it's one of these things where the lower the administrative distance is actually the better and so we trust the edge ERP more than OSPF and the primary reason we trust ya GRP more than all SPF on an cisco routers because cisco invented EIGRP and that's all there is to it so EIGRP has administrative distance of 90 OSPF has an administrative distance of 110 and so that's one ad the other ad would be this right here I call the reporting a reported distance or Rd advertised distance or ADA is another way of saying it but it's what my router neighbor ships my neighbors this is what they are telling me is their distance to the network in question in this case this network 10.24.0.2 is the advertised distance reported distance okay so let's be let's be really clear on this actually I might let me just let me just do a quick drawing I think that'll probably be best if I switch over to you'd think I'd have this down by now let's see here oh it's called is it this nope huh what happened to my I lost my button for switching over to there we go hmm my stream duck is out of date I don't know what happened there all right so let's just draw this out real quick okay so I have give me a brush oh it's doing its thing again hang on I think it actually glitches out and I don't know how to fix this or you all can't even see what I'm drawing because I don't even know why all right so if I let me delete this little eat this can I draw on this one okay can draw on this one hmm I don't know why all right there we go I'm not on my chalk so it looks weird to draw this on a chalkboard but all right so I let's say this is my router this happens to be router 3 this is the one we've been talking about right Oh real quick big pop of different books called different names of the same number reported distance makes more sense to avoid confusion with the mystery of distance yeah thank you that just in summary that's what I was trying to explain earlier but I want to be clear on this concept okay so if I've got a router up here with some connection and maybe it's through other routers it doesn't really matter but some connection to a network let's say this is network a over here okay and what's gonna happen is these routers are talking and let's just assign metrics to these that are easy for us to see and okay let's say that this is a metric of 100 let's say this is a metric of 200 let's say this is a metric of 100 here okay so this routers gonna advertise to that router that hadn't get there in 100 this router here is going to add 100 and 200 and say I can get here in 300 okay again think of this like cities and miles you know if I have three cities and you know sequentially in order and I've got to drive 100 miles to get to the next city and then 200 miles and then 100 miles to get to my destination it's gonna be 400 right I mean we would do this all the time but this is the reported distance we'll call it and so the reported distance is going to be still stored in the topology table and there's a reason for this we've got to keep track of both of these numbers independently so the router 3 is gonna say ok for Network a we're going to write this down as saying that I can get there in 400 this is the feasible distance again this is again it's weird that they pick the word feasible but hey it is what it is the feasible distance is 400 and the RV or ad whatever we want to call it the reported distance is 300 okay the reported distance is important because let's let's do another quick example here let me erase this right here I wanted to cover this and now's a good time to cover it yeah I GRP and loop prevention okay it's doing it again hmm why would it be doing that man all right well I guess I get one layer tonight every now and again just bugs out and doesn't give me multiple layers okay so we're gonna have let's let's say we have another router here and another router over here and this is a weird Network I get that totally agree it's a weird network I don't know what's with this alright router 3 there we go this is still router 3 here ok so what am I gonna do I've got this route 2 a now it's in my routing table have installed in my writing table I'm gonna advertise a out to these neighbors and they're going to eventually cycle this back around and tell me that they have a path to network a this is the nature of distance vector protocols again routing by rumor right I mean does I don't know what this is let's say router 7 does router 7 have a path to network a yes does it go through router 3 yes that's a loop that's not good so why is router 7 telling router 3 that it has a path to a because Roger 7 doesn't know that it goes through router 3 again dis inspector I have my corner of the world and that is it now yes sir I know three would be advertising to router 7 that it also has a path to a but it doesn't that doesn't mean that I don't get an advertisement back from router 7 because right ourselves will eventually in all likelihood have an opportunity to tell me that has a path to router a or to network a big papa fast convergence with the feasible successor already in the routing table our topology table yes and we'll get to that here in a moment so loop prevention first we'll get to we'll start with we're gonna start with loop prevention which is a big deal within the geography topology table and then we're going to get to feasible successors okay which does enable fast convergence big Papa's right about that so how do we prevent this from happening well the way we prevent this from or from looping I should say like what if router 3 trusted router 7 what if router 3 happened to lose this connection right here just at the very moment that router 7 advertises that it's got a route to a you know now all of a sudden we have this network loop problem or what have you so so whatever the scenario is okay router 7 has advertised a router 3 that's it's it's got a path to a let's just make all of these 100 to keep it simple 101 100 so what is router 7 telling router 3 that it's distances well router 3 told this router here that it's able to get there in 400 right it's reporting it's feasible distance so router will just call this router 6 so router 6 advertises to router 7 that I can get to a and I can get to it in well we add the hundred so it's wagging it to it and 500 so router 7 is going to advertise it to router 3 and say that I can get there in 600 well there is a potential we look at we here's the good news like we look at this and we know that there's a loop the network doesn't know that there's a loop because it doesn't have a picture of this again OSPF effectively gives the routers a picture of the topology and EIGRP can't tell what the topology is it's just trusting what people are telling it so what it does is it compares this number 600 to its own feasible distance okay if an advertised route comes in and it's greater than my feasible distance to get to this route to the same network I'm just going to ignore it I'm going to drop it now we know that there's a loop in this case there could not be a loop we could have a route down here that comes up this way and it's like 700 to get there maybe it's a backup serial link and so he's legitimately telling router 3 that hey I have a path to a and it costs I guess with technically 800 and router 3 has no way of knowing whether there's a loop there or whether there's not you know the this you know scenario up in the upper left that is a loop we understand that that's somebody coming back to router 3 and telling them that I've got a path to a even though it goes through router 3 but the bottom right situation is different that just has a slower link but we have to treat them you know both scenarios the exact same and so if our reported distance coming in is greater than my feasible distance then I'm simply I'm not going to store it in my topology table I'm going to ignore it okay that is the result of we call it duel you know the defusing algorithm defusing update defusing update algorithm I think dua L is the this algorithm that it's using to effectively make sure that there's a loop free topology with Ethernet we by and large we can't control whether there's loops that's why we have spanning tree you know like Ethernet doesn't do anything to recognize that there's been a loop routing protocols though we don't want to we don't block links like we do a spanning tree we rely on the networking protocols to to figure out what a loop free topology looks like and make it loop free OSPF and EIGRP both guarantee loop free topology and this is how a jarppi does that all right the good news is with loops and a layer 3 by the way is that a layer 3 loop we rely on that TTL that time to live and so at least within the IP world if a packet goes around and round and round it around forever then eventually it's going to die just because we're decrementing that time to live and what's the time to live reaches zero the the packet dies but with layer two we don't get that which is why we're pretty desperate for spanning tree unfortunately ok any questions on this I'm not seeing anything come in so let me know if if you've got any questions about the loop free part of this topology ok let's go back to let's go back to the let's go back here ok oh there's my button ok I need to get the I guess it was there the whole time go figure ok you all got that back up that looks good ok so let's take a look at ok now I need to pull up there go no lost my session never yeah all right so now we need to talk about this concept of feasible successor okay given the previous example where that bottom in the bottom right where we have this really bad cost come in and we thought there might be a loop alright I mean there could be a loop that way so we don't install that into our topology table we don't have a fast backup path if if we can prove that yes you have a worse you have you have a bad path a bad cost but but if it's impossible for there be a loop there then in the event of an outage going to this network then I could send my traffic that way instantly because I can trust that there wasn't a loop there I'm not going with you because you're a worse path but you're still a valid path does that make sense so maybe we should just flip back and take a look at that let's go ahead and do that real quick so if I pull that real quick back up let's say for example that let's choose a different color here let's say that this path right here isn't 700 well that's kind of cool I like the blurring thing that's going on there whatever that is all about oops can't be that low 300 150 yeah let's call this 150 that should work 250 300 200 300 400 yeah okay so call this 150 so now this router down here is telling our big router right here oh router 3 yeah I forgot we didn't name it so it's telling router 3 that my distance to get there is 100 plus 150 it's 250 darn it that didn't work we need this to be 250 does that work 350 yes that works ok sorry bad math so 100 and 250 is 350 so I'm going to that's still 250 I'm going to say that I can get there now in 350 all right so now I'm getting to advertisements I'm getting 300 and I'm getting 350 so we know which one's going to pick it's going to still pick the same path say okay I'm gonna go this way to get to network a because 300 is better than 350 makes sense by the way in both cases I'm assuming that this is 100 so really it's looking at my feasible distance and my feasible distance is 400 right so let's just write this down it knows game a little busy feasible distance is 400 to get to path a but interestingly enough your advertise distance is 350 350 happens to be less than 400 so what's remember what our loop prevention mechanism is our loop prevention mechanism is saying that if this number right here 400 I'm sorry if this number right here coming in is larger than my path to get there then you might be using me to get there does that make sense so if you're telling me 450 for all I know I'm telling you 400 and you're able to get you know your did your your link to me is 50 so you're thinking oh I can get there I can get there in 450 so there's a there's somehow a loop in this process it's probably not a direct loop back and forth there's usually another router in the path but you get the point right like if if I'm getting there in 400 and you're getting there in more than 400 you might be using me to get there and that's a risk of a loop but but if I'm getting there in 400 and you're getting there in 350 then you're guaranteed not using me you're guaranteed not to be using me so there is no risk of a loop I can trust that you're valid you're not my best path my best path is through a or through whatever this router is at the top here this isn't a this is I don't know router 10 there's throwing random numbers out there my best path is through router 10 but I know that you're good and so I'm going to call you feasible successor that show up no close enough feasible successor there we go alright good so a feasible successor in the event of an outage in the event that this link goes down right here and I you desperate for a route to get to back get to a I'm gonna check my topology table and find out were there any other offers okay it's it's kind of like this I don't know analogy jumped into my head for whatever reason of like losing out on the thing you really wanted to see go and you check to see if your backup option is still there right i mean that's the fact of what we're doing here so the this link goes down you go to your backup oh you go to your topology table say oh i have a feasible successor and instantly i swing over to my feasible successor okay yeah GRP has a lengthy process of querying if you know if that link goes down I'm lost my route I'm going to query my neighbors to try to figure out anybody else has this route it puts the route into active mode you'll study this as part of your I think it's part of CCNA if not as part of your own Coryell you'll cover this that's a very slow process okay the fastest way to converge is if I swing over to a feasible successor eight times go to small group a thanks for joining us before you got a run off on Tuesday night's so today you take care of thanks for joining us okay so um hopefully that made sense again what we're really what we're really focusing here on is the fact that or is it failover scenario all right so now let's go back to the lab on demand and we're gonna take a look at the fact that we have feasible successors so take a look at this for this route we actually have all of these routes stored which is kind of cool okay this is my best path okay my feasible distance is this 3328 and if I'm going to go out towards router one this is how I get there I'm sorry this is my path my cost to get here but look at this look at these reported distances these reported distances of sixteen are less than my feasible distance because they're less than my feasible distance I know that they're not using me to get there again we can look at this oops I didn't put the map up for me get that put back up there we go all right so we can look at this map and say okay yeah I know you're not getting there they're literally directly attached but keep in mind the edge ERP were trusting what people are saying we can't make the assumption of any of that so yeah 2860 now if this if this happened to be higher than 33 28 for whatever reason these entries would not be here okay we would not have them so when you look at some of these other paths let's say this right here so this is that lets a slash 3200 what's up with that all right so 10.30 at 0 that's that back-end Network behind wait I am router 3 why am i ok yeah let's look at here ok finally found one so this is the back end route behind router 1 or the back end that green cloud in the top left the backend network so we've got this 10.1 0.0 it's via router 1 now I have connections I have neighbor ships with routers - and for writers 2 and 4 are telling me guaranteed they're telling me they have a path to 10.1 0.24 they do have a path we look at the topology right router for can get there via router to router 2 can get there via router 1 and so they both have a path to this network and yet I don't see them in the list why don't I see them in the list like I do down here I've got all three routers in this list I've got all three writers in this list well the reason for that is because when they advertise to router 3 they're their path like how fast they can get there it's too long this 28:16 we've pretty well established is about the lowest number on here if my distance if my ability to get there is I did the wrong thing there there it is thirty thirty seventy two so still pretty low so if yeah if either of those guys could get to this network in 28:16 we'd have we'd have that in there right the only one you can get there in twenty eight sixteen is router one because rod R one is directly attached and 2816 is on some level the the gig metric right so the fact that I can get there in thirty seventy two is me adding my my links metric to it 256 again it looks like give or take I don't think that's give or take I think that's I think that's exactly 256 so so I'm dining my 256 metric to it and hitting thirty seventy two so this is my feasible distance they're coming in and saying that they can get there and greater than thirty seventy two if we wanted to see let's go take a look so router two will do a show IP Vig rp2 apology table and here's our networking question ten 1.0.0 so we can indeed get there from router 2 through router one except my ability to get there is 30 72 okay we can both get there if we look at the topology that makes sense router to and router three are connected via get connections to router one and router one has this network directly attached so it's the exact same distance but it's not less so router two is telling router three hey I can get there in thirty seventy two incidentally is happening the other way around to right like rudder three is telling rod or two I can get there in thirty seventy two but there's a chance not a good one but there's a chance that the link that you view the you know the link between us is zero and so if you think you're getting through here to this network through me well that's that's bad news that's a loop right so I can't trust it so that's why it doesn't show up in this list so this is why feasible successors are are better within the world of EIGRP is because I can very quickly flip over to them if I lose this connection right here - you know - router one if I lose my neighbor connection this my ability to get to this route is drastically impeded it's going to be a slow process for me to get back I'm gonna have to send queries out two routers two and four and say hey can you get to this network still and they will say yes they'll both respond and I'll get to pick you know between the two which one's the better path but that's a much slower process than me simply redirecting my traffic I mean yeah GRP can redirect that traffic in microseconds well milliseconds really usually in that hundred millisecond range is what I've seen in real life too flowing a route over from one link to another when you have a feasible successor so if it takes you know if I have to go out and query it might take seconds just just for how long it takes to the routers to process the and the other thing to that you'll learn in eigrp is when you send all of your queries out you can't do anything to get all your queries back so if one of those queries get stuck for whatever reason then it causes problems so who okay so um let's see what's next up on the agenda again any questions be sure to ask this whole concept of reported distance and feasible distance is very important especially if you're going for the Encore if you're at CMP level and you're moving forward make sure that you are studying this concept but if your CCNA level then this is this is beyond CCNA don't freak out about this mostly the configuration that we've done so far is is primarily where you're going to need to to get to okay so let's talk about metric how in the world is metric calculated all right um you know I'm just going to I'm just gonna paste this in copy this let's go back to our whoops [Music] pull this up all right so whoo look at that beautiful thing okay so we are a little bit we're almost actually we are a little bit too big so let me get cleaning this up for you all oops I was gonna write this out but then just figure it's probably easier to see it if it's just in text format because this is kind of a nasty algorithm in case you haven't noticed yet alright make this clean so we can talk about it okay so this is yeah I did it's it's it's not great okay I'm not gonna lie you don't usually have to memorize this for any exam that I've ever seen I've never been asked you know which is the right metric calculation so that's the good news we don't have to memorize this algorithm there are parts of this we need to understand okay so let's walk through this first of all sisqó values for concepts within AI GRP metric calculation bandwidth load delay and reliability okay these are the four parameters that cisco is going to use to evaluate a link and figure out what that symmetric value is by the way one thing you don't see on here is MTU so there I've seen things around where like MTU is calculated as part of this it's bizarre because we actually have five constants these K values which are going to talk about a moment are constants that we relate back to these concepts for whatever reason we have two different constants that refer to reliability it there's there's kind of a vague reason for it but either way just the fact that we have five values sometimes we're like okay bandwidth load delay reliability and something else no it's just four we have five K values and four parameters okay so what we're looking at here is bandwidth which is okay so yes it's the bandwidth of the link but when you configure this bandwidth parameter does not change the bandwidth of the link so it's a kind of a one-way street right the the bandwidth of the link can set the bandwidth value of an interface but I've seen people go on to interfaces and be like bandwidth forty Meg or whatever on a hundred like maybe you have a hundred Meg link and you're trying to shrink it to like match what your service provider has or whatever so you do have bandwidth forty that's actually good practice because yeah GRP needs to see that link as a hundred Meg or as forty Megan said of one hundred but it doesn't actually change the bandwidth of the link the link can only run at 100 Meg well short of traffic shaping QoS mechanisms and such which we've covered in the past but so so bandwidth it's gonna look at the bandwidth value not strictly speaking the bits per second on the link but whatever we have configured on that link the load is a I'll call it a dynamic value it depends on how busy the interface is but we're gonna come back to that we're dynamic here in a little bit then we have delay that would be how long it takes to you know sent traffic across this link so the faster the interface the lower the delay typically serial links have notoriously bad delay and reliability so this would be like as packets get dropped on this interface or as discards or a bad frame check sequence is whatever you know as the link looks unstable I'm going to drop this reliability okay who then we take these K values and we decide how much weight effectively we want to give to them now here's the good news in all of this these values are all by default either one or zero and the only ones that are even one are k 1 and k 3 by default we set k 2 to zero which negates this whole part of us taking load into account and K 4 and K 5 or 0 so that gets rid of this entire part right here okay so K 5 is 0 now so there you go although what's interesting about this is that and this is true even though it multiplies this by if true leaf k5 is set to zero the way Cisco made this like the entire thing would be equals 0 because 256 times a bunch of whatever times 0 well anything times anything times 0 0 so for whatever reason they actually have it as an exception that if k5 is equal to 0 which by default it is that this entire thing equates to a 1 I don't you think they could have come up with a better workaround than that but yeah it is what it is so what what that means is since k1 is is I'm sorry K 1 and K 3 are both 1 that means cisco cares about bandwidth and delay because K 2 K 4 K 5 or 0 that means they don't care about load and reliability why don't they care about load and reliability I mean I wasn't there when Cisco designed EIGRP but there are a couple of reasons that are circulated for this one of which is that all of these values are statically determined once the link comes online so the edge ERP doesn't send like you know how rip I don't know if you've studied rip yet but rip sends updates every 30 seconds and so every 30 seconds just sending an update yeah juror B doesn't work that way the edge ERP is only gonna send an update if it needs to send an update otherwise life is good carry on what I told you hasn't changed there's no reason for me to send an update so if if these values load and reliability are constantly changing based on the actual load based on the actual reliability it doesn't actually help because I can't really recalculate I would be going into my topology table and constantly changing values we don't want to do that so anytime I get an update on that link I'm going to look at the load and liability and figure out what it is at that moment in time and it's sort of a snapshot I learned this route when it came in when the load was high and the reliability was low and I inserted into my topology table and it's going to be stuck that way until I get another update when maybe the load and reliability are different so that's one reason I that's probably the largest reason and so the fact that k2 k4 and k5 start at zero makes sense this is something you'll want to memorize you'll want to memorize this I mean the way to memorize this honestly is this it's one zero one zero zero just remember that binary sequence it's binary for 20 we get 16 plus no 8 plus 4 plus no twos Plus no one's so 16 plus 4 it's 20 so if you can remember that in some fashion that's what I did when I was taking CCNP exams way back in the days I memorize that one zero one zero zero again you don't have to memorize this at least I've never experienced where we have to memorize the whole thing so let's look at this algorithm by default so if we're taking the default version we would say 256 times and then we have let's do the brackets here for now we might be able simplify that k1 times bandwidth plus k2 is to 0 so this whole thing is 0 k3 s is 0 so that whole thing is 0 oh I'm sorry cute okay there's not zero don't don't don't ignore what I just said plus k3 times delay okay and I missed a parenthesis there we go big papa default values that's important yeah you got a you got to know him you got a known okay so there we go so so this algorithm just got way simpler like the big complicated parts of it were low and reliability portion so fortunately it's simply a matter of hey k1 and and again let's let's think about the defaults again what are K 1 and K 3 by default one so since 1 times bandwidth is bandwidth and 1 times delay is delay by default we can even get rid of some of these parentheses in it oh that's right this program is silly it doesn't use delete or Alisa treats delete as if it's a backspace instead of delete all right so there we go there's our algorithm 256 times bandwidth plus delay okay the with is in kilobits per second and technically it's like an inverse it's this is okay let me just write this out bandwidth is remember that Barrett reference bandwidth conversation it's ten to the seventh which is ten gig divided by the actual bandwidth of the link because you know if you think about it the higher your bandwidth with this out if this algorithm is just straight up what we're use then the higher the bandwidth the higher your metric well we don't want the metric to go up as your bandwidth goes up you want as bandwidth goes up your metric to go down so we have to somehow inverse that value and the way we inverse that value is we divide it into ten to the seventh so what's interesting about that is it does still technically work well yeah I don't know we have to look that I have to think that that would if your bandwidth is like 40 gig I have to think at that point that it would it would it would go to two one it would just it would just calculate I was a one just what 10 gig does anyways I haven't seen changing the Eid RP default bandwidth but guess what back when I was studying yeah GRP in depth 10 gig was a pipe dream okay so oh yeah alright so there are two things I want to talk about here two bandwidth is the lowest bandwidth in the path okay what that means is if I've got 15 gig links through a bunch of routers and then I've got a Fast Ethernet link to connect to the final destination 'maybe lowest bandwidth in that path is a hundred Meg and so my bandwidth calculation is based on 100 Meg it's always based on the lowest bandwidth so the bandwidth is actually propagated throughout the advertising path as part of the as part of the PRP update otherwise we wouldn't have a clue what's downstream right so the lowest bandwidth in the path is the bandwidth that's used for calculation the delay however is the sum of delays throughout the entire network so when it comes to these two which one do you want to tweak the bandwidth or the delay big pop that's like OSPF looking at the difference 100 Meg and a gig yeah exactly it's the exact same situation it's just off by a factor of two is that right yeah ten gig gig hundred Meg OSPF starts off with like ten to the five and the edge ARP toes ten to the seven for whatever reason so yeah more but can they've just gone like ten to the ten hour ten to the twenty I don't know I guess that would have made for a larger larger metric calculation so they're trying to find that balance which is interesting all right so the question of which one you want to tweak the one you want to tweak is the delay the reason for that is I actually just hit this in this lab I was messing around with this lab earlier and I tweaked to the bandwidth even though I know that you should tweak the delay because I just wasn't thinking I went in there and like I'm just gonna change the bandwidth here and test something and it didn't change nothing changed why didn't anything change because I tweaked the bandwidth except the bandwidth was the minimum in the path and so the bandwidth I tweaked happened to not be the minimum of the path so I took like you know this link bandwidth and I lowered it to here even though this link out here was the lowest one in the path so tweaking the bandwidth can do nothing literally nothing if I tweak the bandwidth drinking the delay will always change the metric in the path because we add up all of the delay and that is that is that is what we use for our calculation so when it comes to manipulation metric manipulation if you get equal path routes and you want to choose one over the other or maybe you've got a I don't know a 20 gig link somewhere but you know that that's I don't know whatever the situation is and you want to change the yeah ARP metrics well this is where you can start to change the delay and that's usually the right way to to make changes in the network okay so I think I think that's about it for the metrics okay if there's any questions about any of that let me know we're gonna do one more thing and we're about right on schedule here we're gonna switch back to the lab and we're gonna take a look at summarization all right so summarization is what I have said a few times this is oh yeah I've got to get the hold up there we go nope there we go all right that looks good so summarization is lying it really is truly saying hey I've got a bunch of routes but I'm not telling you about those routes I'm only telling you about this one so why do we summarize well we summarized because I might truly have 100 routes behind me and I don't need to tell you about all 100 if I can summarize them into one block of addresses you know I mean what why it would be the effectiveness of you know if you're you've got a presentation to give me in 200 pages and we meet up somewhere and you're like alright here's the presentation page 1 page 2 page 3 page 4 page 5 page 6 page 7 and I'm like can't you just give me the binder you know just give me the binder I gotta go you know or whatever the situation is so you give me the binder that's that's the rap summarization right it's like okay really I've got all of these but why bother giving you all of the details when I can summarize this route and so routing tables are not of infinite size we don't usually brush up against the largest or the maxes of our routing table size but we want to be efficient want to do things well and so we can summarize to help reduce routing table loads it increases the speed of Route table lookups so for large networks this makes this make sense it's a big deal I will say this though summarization is not something you typically go into a network and willy-nilly just start summarizing you need to have architected your network to be summarized about for lack of better words if I have 15 sites all hanging off of one wayne router let's say and their networks are all over place well then I can't really summarize that not easily however what if I made them sequential what if it was like okay these 15 or you're gonna be 172 dot 16.0 you're gonna be dot 1 you're gonna be dot 2 your army dot 3 you're gonna be dot for and I sequentially line up their address blocks now I can summarize ok so it takes network planning as part of the network design process it's it's part of you know you you might have to re I P or your network in order to do this I've I've participated in that with companies who have said you know what we we want we need to start summarizing we're large enough but we didn't really build this out right ok well let's go out and you have to reappear or some of your branch sites or what have you so that you can start to summarize those to the rest of the network alright so without further ado let's go and configure this first of all we need some networks to summarize we don't really have anything to summarize right now so let's go ahead and make some loopback addresses and we'll summarize those so router 3 yes I'm gonna do it like this well ok so looking at the diagram what I'm gonna do is I'm gonna make like for loopback addresses on router 3 and I'm going to get under router 1 I'm gonna summarize those routes towards router - ok so let's just for the sake of this let's go ahead and shut down those other two links so s3 3 I'm gonna shut this down and the link to router for I'm going to shut this down ok so now my only path to router 2 is through router 1 ok ideally the other you know the other problem a summers a summarization usually happens out at the edge not really in the core and we have all these this full mesh near full mesh of connections among these 4 routers it's practically like your network core I mean if I really wanted to effectively summarize I'd have to summarize from router 3 out so for this example let's just simplify our network now it's just router 3 connecting to router 1 router 1 talking to router 2 ok so let's make those loopback interfaces interface loopback with 0 IP address let's make this I don't know three dot 3.0 dot zero yeah let's do that two five five two zero whoops Oh Pat right I was thinking about the network dot three to thirty five to two five two to five dot 0 and then we have interphase look back one IP address three dot 3.1.3 255 two to five no there we go that 0 and then interphase loop back to IP address 3.3 dot 2.3 you were lining up that third octet with the loopback number so interphase look back three IP address finally we get to 3.3 to 3.3 that seems like a good ID so let's go ahead and get into our router AIG RP 100 configuration and we're going to add the network statements for these so network 3.3.5 some fun here we we could advertise all four of those links with one network statement and I could just take the easy way out we'll see ya watch watch me blow this but let's let's try to hit let's try to make this right so three dot 3.00 what be the wildcard mask for this we care about the first three we care about the second three now this third octet is is the problem because we know it's gonna be dot X dot 255 right I mean we we know that there's there's gonna be some level of this so how many addresses do we have and we've got four so zero through three would be the answer there so so actually it should be three yay fun yeah I know wildcard masks right good times so let's see if this works we're gonna try to use one Network statement to activate all four links at once but also contain it so we're only activating those three or those four links and so let's go ahead and take a look at router one and we'll say show IP route via GRP let's see if we've got them all up we do look at that well as best I can tell we did it right with the wildcard mask so that's good so we have four routes now coming to router one and router two should see the same thing let's see do show IP route the edge ARP and we do ok good we've got four different routes and that's too many let's let's get this down to one so we could summarize actually on router 3 but let's summarize in router 1 so router one's going to know that there are 4 real routes and it's going to lie to router 2 and convince router 2 that there's only one again this configuration is just like some of them other ones I mentioned earlier it goes on to the interface so we get on to this interface interface gig 1 slash 0 hit enter and we're going to do a summer address so IP again you think of the IP EIGRP but it's IP summary - address then we can see ah this is the one where it was ripping EIGRP all right so I'm not crazy at least so II I GRP we're gonna pick and then 100 for the autonomous system number and here's where we entered the summary IP address so I can there there's nothing stopping me from summarizing this big right like I could say 3 dot I could I could do this 3.0 at 0 at 0 slash 8 does this contain those 4 networks yes is it the most effective way to summarize no it's not if for example a I don't know I mean I'm telling router 2 that I can get to this entire slash a the reality is I can get to a very teeny tiny part of it so let's let's try to be more more specific those I'll start with 3.3 so I could do this 323.00 slash 16 is that more accurate yes it's more accurate is it the most accurate no no it's really not so how many bits are really changing well only 2 bits are changing right because I'm only changing 4 I've got 4 networks it starts with 0 and I'm making that 0 1 2 & 3 1 binary that's 0 0 0 1 0 1 1 so no matter what it's just 2 bits so I could actually summarize this with a 24 minus 2 is a / 22 okay this would be the best way to summarize this network alright so let's go ahead and try this 3.3 - 0 - 0 / 22 and the edge erp is gonna do a little bit of a pseudo hiccup it's kind of funny it says it's gonna resync because there was a summary configured configuring a summary if it didn't update if it didn't send if it didn't trigger an update to its neighbor it would not affect the neighbor right remember when we said he had GRP only sends updates when it needs to so this is a situation where it needs to send an update and just sent one so let's go and take a look at router 2 up here and we're going to run the exact same command and find that it's a little different so 3.0 - 0 - 0 / 22 that would be the summary address that I configured and it's able to get to it via router 1 so let's make sure I've got configure connectivity let's ping what was it three dot 3.0 dot three and I can get there dot one dot three I can get there 2.3 that one looks good and last but not least 3.3 so I can get to all four of those subnets so I so I correctly sub netted and I narrowed it down to just those four subnets so technically I mean summary addresses this is something that they could ask on the encore exam especially would be this exact scenario where they want you to identify the most efficient summary address possible okay so that's where yeah anyways so so if they ask us that question we need to be prepared to answer that and to say okay yeah it would you know going to scroll back up here right well actually looking at router 1 you know is it is it this is it this / twenty - is it a slash 20 is it a slash 18 is the slash 16 and you're gonna have to be able to summarize that so so be aware of that understand how how summery dresses work you do notice by the way automatically rescinded the back end routes which we'd expect which is something to pay attention to is I'm not advertising the summary and those four routes for didn't become five for truly became one so its suppressing all of those routes that fall into this / 22 okay shaving a haircut see fun I that must be a reference to something that I don't get but unless you're dummy I need a shave and a haircut but I feel like I think that's probably not the case alright very good well I think that about covers it for tonight I don't see any other questions airbag facial hello you're joining just in time for us to say goodbye but hey welcome and welcome to our Tuesday night CCNA training we cover CCNA topics every Tuesday at 8 p.m. Eastern that said note oh hey thank you very much for the follow they're a bag that said next week I am taking the night off it is my wife's birthday so I you know I did this I subnet it on my birthday because that was fun but I didn't feel like I should subnet on her birthday somehow so I'm taking the night off next week it'll be a nice break from from the routine so everybody study up on you know just make sure you keep progressing with your studies as needed when we come back on the should be the 25th of February the two weeks from tonight the new CCNA will have launched the new CCNA launches on February 24th we're gonna meets February 25th we're gonna pull up that CCNA blueprint we're going to explore it look at all of the new concepts any questions and answers or you know QA time so bring your questions bring your concerns about the new CCNA and we'll have a lot of fun diving into that all right airbags sewing for CCNA 2020 I will be back for sure awesome very good yeah please in two weeks it'll be really relevant because we'll be looking right at the CCNA blueprint and seeing all of the fun things that they're adding so alright everyone well thank you very much appreciate your time and we'll hopefully see you in a couple weeks have a good night [Music] [Music] [Music] you [Music]

Channel: KishSquared

Views: 583

Rating: 4.7647057 out of 5

Keywords: ccna, eigrp, routing, cisco

Id: ushEHVNjjSw

Length: 91min 14sec (5474 seconds)

Published: Thu Feb 13 2020