WEBVTT

0:00:08.320000 --> 0:00:12.480000
 In this video, I'm going to go into the
 OSPF Designated Router and backup

0:00:12.480000 --> 0:00:16.840000
 Designate Router in theory and in practice
 and talk about why these things

0:00:16.840000 --> 0:00:18.780000
 were developed for OSPF.

0:00:18.780000 --> 0:00:21.620000
 So to begin with, let's go ahead and
 go to the whiteboard first before

0:00:21.620000 --> 0:00:24.760000
 we go into our slides to
 talk about this concept.

0:00:24.760000 --> 0:00:30.000000
 And like I frequently like to do, I
 like to contrast OSPF and EIGRP so

0:00:30.000000 --> 0:00:34.300000
 you can see some of their similarities
 and some of their differences.

0:00:34.300000 --> 0:00:37.520000
 So here we have four routers and all
 four routers are connected to the

0:00:37.520000 --> 0:00:41.160000
 same switch and they're all in the same
 broadcast domain as you can see

0:00:41.160000 --> 0:00:43.320000
 from their IP address.

0:00:43.320000 --> 0:00:51.660000
 So if we were talking about EIGRP as
 an example, Router 1 would list R2,

0:00:51.660000 --> 0:00:56.580000
 R3, and R4 as neighbors.

0:00:56.580000 --> 0:01:02.760000
 Now let's say the R1 had an EIGRP
 update that it needed to send.

0:01:02.760000 --> 0:01:07.480000
 We know the EIGRP updates go out as multicast
 packets and what do switches

0:01:07.480000 --> 0:01:09.120000
 do with multicasts.

0:01:09.120000 --> 0:01:13.100000
 They flood them just like they flood
 broadcast so that multicast packet

0:01:13.100000 --> 0:01:18.980000
 would be flooded as
 R2, R3, and R4.

0:01:18.980000 --> 0:01:22.880000
 And now because R1 is neighbors with
 all three of those routers, he would

0:01:22.880000 --> 0:01:28.120000
 wait and he'd say, well, I need to get
 an EIGRP acknowledgment from R2,

0:01:28.120000 --> 0:01:34.920000
 R3, and R4. So all of a sudden here's
 R2 sends an EIGRP hello packet with

0:01:34.920000 --> 0:01:36.880000
 the sequence number saying
 I acknowledge.

0:01:36.880000 --> 0:01:39.140000
 And he says, okay, R2,
 you're good to go.

0:01:39.140000 --> 0:01:40.440000
 Thank you for acknowledging.

0:01:40.440000 --> 0:01:46.340000
 And here comes our three sending a
 hello packet and we check that off.

0:01:46.340000 --> 0:01:49.900000
 And then hopefully eventually our four
 would send his own hello packet

0:01:49.900000 --> 0:01:52.680000
 with that same sequence
 number in it.

0:01:52.680000 --> 0:01:56.080000
 And so R1 would say, okay, all three
 of my neighbors have acknowledged

0:01:56.080000 --> 0:02:01.700000
 this update. In the event that something
 went wrong and one of these routers

0:02:01.700000 --> 0:02:07.640000
 did not send an acknowledgment, then
 R1 would resend his EIGRP update

0:02:07.640000 --> 0:02:11.420000
 as a unicast packet directly
 to that neighbor.

0:02:11.420000 --> 0:02:14.780000
 Okay, so what does that
 have to do with OSPF?

0:02:14.780000 --> 0:02:18.100000
 Well, I'll remember that with OSPF, we
 saw that when we were talking about

0:02:18.100000 --> 0:02:21.360000
 the process of building a neighbor
 relationship, that it was important

0:02:21.360000 --> 0:02:26.020000
 that all the OSPF routers were neighbors
 had a common database that their

0:02:26.020000 --> 0:02:27.960000
 LSA's were the same.

0:02:27.960000 --> 0:02:32.040000
 And so there were about six different
 steps we had to go through to get

0:02:32.040000 --> 0:02:37.760000
 to that point. We had to go to init
 two way EX-START, exchange, loading,

0:02:37.760000 --> 0:02:41.200000
 and then finally full
 to do all that.

0:02:41.200000 --> 0:02:46.260000
 So in this type of environment, in this
 type of a network where a router

0:02:46.260000 --> 0:02:51.180000
 says, okay, there could be more
 than one router on this network.

0:02:51.180000 --> 0:02:53.080000
 For example, think about Ethernet.


0:02:53.080000 --> 0:02:57.660000
 The nature of Ethernet is that you can
 put one frame on the wire and you

0:02:57.660000 --> 0:02:59.480000
 don't know how many people
 are going to see that.

0:02:59.480000 --> 0:03:03.720000
 You don't know how many people are
 on that same Ethernet cable as you.

0:03:03.720000 --> 0:03:08.700000
 The nature of Ethernet is that it could
 be two, three, four, forty-four.

0:03:08.700000 --> 0:03:13.660000
 You don't know. Ethernet is what
 we call a multi-axis medium.

0:03:13.660000 --> 0:03:19.100000
 So the desires of OSPF, they said, well,
 we could on a multi-axis medium

0:03:19.100000 --> 0:03:23.080000
 like Ethernet, where there's a potential
 of lots of different routers

0:03:23.080000 --> 0:03:27.240000
 out there on this wire, we
 could do like EIGRP does.

0:03:27.240000 --> 0:03:31.020000
 We could have all the routers go through
 all five or six steps with each

0:03:31.020000 --> 0:03:35.060000
 other, all the routers
 become fully adjacent.

0:03:35.060000 --> 0:03:41.960000
 And they said, and then when an OSPF
 router sends out a link state update,

0:03:41.960000 --> 0:03:45.980000
 we could do like EIGRP does, and all
 the other routers have to acknowledge

0:03:45.980000 --> 0:03:51.340000
 it back. But the desires of OSPF
 took a different approach.

0:03:51.340000 --> 0:03:54.100000
 They said, here's what
 we're going to do.

0:03:54.100000 --> 0:04:00.220000
 On a multi-axis network segment like
 Ethernet, we're going to have one

0:04:00.220000 --> 0:04:04.320000
 router elected as a very special position,
 which is going to be called

0:04:04.320000 --> 0:04:06.080000
 the designated router.

0:04:06.080000 --> 0:04:09.340000
 Let's just say it's
 R4 as an example.

0:04:09.340000 --> 0:04:14.040000
 During the exchange of Hello packets
 is when this election takes process.

0:04:14.040000 --> 0:04:17.540000
 So as these four routers are exchanging
 Hello's, remember, their Hello's

0:04:17.540000 --> 0:04:23.300000
 are going out as two?

0:04:23.300000 --> 0:04:32.260000
 224.0.0.5. Hello's always
 go to that address.

0:04:32.260000 --> 0:04:37.880000
 So inside that multi-cast Hello, there
 are some fields that help these

0:04:37.880000 --> 0:04:42.600000
 routers to figure out who this designator
 router is going to be.

0:04:42.600000 --> 0:04:45.600000
 Well, what's the significance
 of the designated router?

0:04:45.600000 --> 0:04:49.840000
 Well, once we know who the designated
 router is, router one will go through

0:04:49.840000 --> 0:04:53.700000
 all six steps with router four.

0:04:53.700000 --> 0:04:57.440000
 In it, two way, start exchange,
 loading full.

0:04:57.440000 --> 0:05:02.020000
 Router two will become fully adjacent
 with router four, and router three

0:05:02.020000 --> 0:05:04.980000
 will become fully adjacent
 with router four.

0:05:04.980000 --> 0:05:11.900000
 But notice R1, R2, and R3, they're not
 becoming fully adjacent with each

0:05:11.900000 --> 0:05:13.940000
 other. They see each other.

0:05:13.940000 --> 0:05:16.300000
 They're still seeing each other as
 Hello's because those Hello packets

0:05:16.300000 --> 0:05:18.940000
 are multi-cast, and so
 they're seeing that.

0:05:18.940000 --> 0:05:23.080000
 But when they recognize when R1 looks
 at R2 and says, oh, neither one

0:05:23.080000 --> 0:05:27.720000
 of us is a designated router, we don't
 need to go beyond the two way state.

0:05:27.720000 --> 0:05:29.620000
 I see you and you see me.

0:05:29.620000 --> 0:05:32.800000
 Okay, good. We're in the two way state,
 but that's as far as we need to

0:05:32.800000 --> 0:05:34.960000
 go. We can stop there.

0:05:34.960000 --> 0:05:39.940000
 But with the DR, everybody
 becomes fully adjacent.

0:05:39.940000 --> 0:05:41.540000
 What's the purpose of that?

0:05:41.540000 --> 0:05:45.180000
 Well, now if R1, let's just
 take R1 as an example.

0:05:45.180000 --> 0:05:50.540000
 Let's say the R1 has a link state update
 that it needs to send containing

0:05:50.540000 --> 0:05:54.540000
 one or more new LSA's
 inside of it.

0:05:54.540000 --> 0:05:59.400000
 R1 is going to say, I can only send my
 link state updates to OSPF neighbors

0:05:59.400000 --> 0:06:03.340000
 that I am fully adjacent with,
 that I am in the full state.

0:06:03.340000 --> 0:06:06.960000
 I went through those five or six steps with,
 which in this case is a designated

0:06:06.960000 --> 0:06:11.060000
 router. He says, hmm, I know there's
 other routers here on this segment

0:06:11.060000 --> 0:06:17.180000
 besides R4, the DR, how do I get this
 link state update just to him?

0:06:17.180000 --> 0:06:18.960000
 Ah, here's what I know I'll do.

0:06:18.960000 --> 0:06:27.900000
 I'll send my link state update to a special
 reserved OSPF address of 224006.

0:06:27.900000 --> 0:06:32.700000
 Only the designated router is
 listening to that address.

0:06:32.700000 --> 0:06:37.260000
 So now in this link state update packet
 goes in, it is a multicast, so

0:06:37.260000 --> 0:06:38.520000
 it will be flooded.

0:06:38.520000 --> 0:06:42.640000
 Our two, our three, and our four will
 see it, but our two and our three

0:06:42.640000 --> 0:06:43.540000
 will discard it.

0:06:43.540000 --> 0:06:47.900000
 They'll say, oh, I'm not a designated
 router, this is not for me.

0:06:47.900000 --> 0:06:53.660000
 And so now that link state update will
 only be seen and processed by R4.

0:06:53.660000 --> 0:06:57.140000
 Now R4 gets it, here comes in.

0:06:57.140000 --> 0:07:02.280000
 Now R4 says, great, okay, my link state
 database is synchronized with

0:07:02.280000 --> 0:07:06.080000
 R1. R1 and I are in sync,
 I got his link state.

0:07:06.080000 --> 0:07:08.220000
 Oh, by the way, let me
 acknowledge that back.

0:07:08.220000 --> 0:07:11.440000
 So R4 will do that, he'll send a link
 state acknowledgement packet back.

0:07:11.440000 --> 0:07:18.720000
 Now R4 will say, but I have two other
 neighbors on this wire, R2 and R3.

0:07:18.720000 --> 0:07:24.680000
 So now what R4 will do is he'll turn
 around and he'll send a link state

0:07:24.680000 --> 0:07:31.120000
 update to R2 and R3, link state update,
 he'll reflect that link state

0:07:31.120000 --> 0:07:39.520000
 update right back going
 to 224.0.0.5.

0:07:39.520000 --> 0:07:44.120000
 Now you might be thinking, well, wait
 a second, if R4 does that, is an

0:07:44.120000 --> 0:07:49.400000
 R1 going to see his very own link state
 update packet coming back to him?

0:07:49.400000 --> 0:07:51.460000
 As a matter of fact, yes, he will.


0:07:51.460000 --> 0:07:55.080000
 But remember, in that link state, through
 that link state update has an

0:07:55.080000 --> 0:07:58.580000
 LSA inside of it, maybe
 more than one LSA.

0:07:58.580000 --> 0:08:02.920000
 Remember, every LSA has a field that
 says, who does an LSA belong to,

0:08:02.920000 --> 0:08:06.260000
 who created it, who is the
 originator of the LSA?

0:08:06.260000 --> 0:08:10.000000
 So in this particular case, when R1
 gets that link state update and he

0:08:10.000000 --> 0:08:13.300000
 looks inside of it, he'll say, oh, here's
 an LSA that was advertised by

0:08:13.300000 --> 0:08:15.760000
 R1. Oh, that's me.

0:08:15.760000 --> 0:08:16.740000
 Okay, I can ignore that.

0:08:16.740000 --> 0:08:18.960000
 Here's another one that
 was advertised by R1.

0:08:18.960000 --> 0:08:21.780000
 Oh, okay, that's me,
 I can ignore that.

0:08:21.780000 --> 0:08:26.860000
 So in this regard, all these riders
 will become fully synchronized.

0:08:26.860000 --> 0:08:32.080000
 So once again, just to repeat the
 process, the DR is elected.

0:08:32.080000 --> 0:08:36.260000
 All these riders exchange hello packets
 and within the hello packets are

0:08:36.260000 --> 0:08:39.280000
 some fields, which I haven't talked
 about yet, which will help them to

0:08:39.280000 --> 0:08:42.480000
 figure out who the designated
 router should be.

0:08:42.480000 --> 0:08:47.940000
 Once that takes place, all of the routers
 become fully adjacent with the

0:08:47.940000 --> 0:08:48.880000
 designated router.

0:08:48.880000 --> 0:08:52.380000
 They go through all five or
 six of those steps with him.

0:08:52.380000 --> 0:08:57.260000
 After that point, if anybody needs to
 send a link state update, they will

0:08:57.260000 --> 0:09:07.420000
 send it to 006. That is an address
 that only the DR is listening to.

0:09:07.420000 --> 0:09:13.120000
 The DR internal reflect that link state
 update right back again so that

0:09:13.120000 --> 0:09:17.460000
 all the other routers can get it, including
 the originator who will say,

0:09:17.460000 --> 0:09:24.440000
 nope, and that will go to 224 005.


0:09:24.440000 --> 0:09:28.140000
 So the idea being here is that if I
 have an Ethernet switch that let's

0:09:28.140000 --> 0:09:33.140000
 say has 20 routers on it, they don't
 all have to become fully adjacent

0:09:33.140000 --> 0:09:34.500000
 with each other.

0:09:34.500000 --> 0:09:37.780000
 They only have to become fully adjacent
 with the designated router.

0:09:37.780000 --> 0:09:41.040000
 They send their updates to him and
 then they don't have to worry about

0:09:41.040000 --> 0:09:43.440000
 it. He's the one who's going to turn
 around and reflect it to everybody

0:09:43.440000 --> 0:09:47.800000
 else. So the designated router's job
 is to make sure that everybody is

0:09:47.800000 --> 0:09:49.620000
 synchronized with everybody else.

0:09:49.620000 --> 0:09:51.140000
 That is his job.

0:09:51.140000 --> 0:09:55.100000
 Oh, and by the way, there's one other
 router here who is also elected

0:09:55.100000 --> 0:10:01.380000
 as a backup designated
 router or a BDR.

0:10:01.380000 --> 0:10:09.960000
 Just like the DR is listening to 224
 006, the BDR is also listening to

0:10:09.960000 --> 0:10:16.480000
 224 006. So when that initial update,
 when R1 sent his initial link state

0:10:16.480000 --> 0:10:21.840000
 update to 224 006, both
 R3 and R4 got it.

0:10:21.840000 --> 0:10:27.320000
 But it was only R4's job as a designated
 router to turn around and reflect

0:10:27.320000 --> 0:10:32.020000
 that back so that in this
 case, R2 could get it.

0:10:32.020000 --> 0:10:34.000000
 R3 is sort of just sitting
 there silently.

0:10:34.000000 --> 0:10:38.240000
 He's taking stuff in because he's listening
 to .6, but it's not his job

0:10:38.240000 --> 0:10:40.320000
 to turn around and
 reflect stuff back.

0:10:40.320000 --> 0:10:46.020000
 Only if the designated router fails
 will R3 take up the job as the new

0:10:46.020000 --> 0:10:48.440000
 designated router.

0:10:48.440000 --> 0:10:53.800000
 So now that we know that, let's
 go into these slides.

0:10:53.800000 --> 0:10:59.820000
 So a DR and a BDR are elected in broadcast
 and non-broadcast multi-access

0:10:59.820000 --> 0:11:02.960000
 networks. Now we haven't really talked
 about those two terms yet that's

0:11:02.960000 --> 0:11:06.240000
 coming up in a little bit in a different
 video, but just think about it

0:11:06.240000 --> 0:11:12.300000
 this way. We know that at layer two
 at the data link layer of the OSI

0:11:12.300000 --> 0:11:18.200000
 model, there are several different
 encapsulation types, Ethernet, PPP,

0:11:18.200000 --> 0:11:20.520000
 HDLC, Frame Relay.

0:11:20.520000 --> 0:11:24.440000
 You could probably think of one or two
 or three more in addition to that.

0:11:24.440000 --> 0:11:27.900000
 What's one thing that makes them
 different from each other?

0:11:27.900000 --> 0:11:32.260000
 Some of those encapsulation types were
 built around the assumption that

0:11:32.260000 --> 0:11:37.560000
 I could put one frame on the wire and
 multiple devices could see that

0:11:37.560000 --> 0:11:39.840000
 one frame, like Ethernet, right?

0:11:39.840000 --> 0:11:40.960000
 That's the nature of Ethernet.

0:11:40.960000 --> 0:11:44.600000
 That's the nature of, well,
 it's a stick with Ethernet.

0:11:44.600000 --> 0:11:45.080000
 That's Ethernet.

0:11:45.080000 --> 0:11:47.960000
 You can put one frame on there
 and everybody will see it.

0:11:47.960000 --> 0:11:52.060000
 Other layer two encapsulation types
 say, well, if I put one frame on the

0:11:52.060000 --> 0:11:56.280000
 wire, it's only going to go one place,
 like the point-to-point protocol

0:11:56.280000 --> 0:12:01.020000
 or HDLC. Those are layer two protocols
 that say this was designed for

0:12:01.020000 --> 0:12:03.160000
 a cable. It's just
 got two devices.

0:12:03.160000 --> 0:12:05.720000
 Put something on the cable,
 it only goes over here.

0:12:05.720000 --> 0:12:08.100000
 This guy put something on the
 cable, it only goes over here.

0:12:08.100000 --> 0:12:11.720000
 So there's really no need for an address
 because there's just the two

0:12:11.720000 --> 0:12:15.560000
 of us. PPP and HDLC are like that.


0:12:15.560000 --> 0:12:19.360000
 Then there's other types of encapsulation
 types like Frame Relay that

0:12:19.360000 --> 0:12:23.440000
 say, well, this particular cable
 could go multiple places.

0:12:23.440000 --> 0:12:25.900000
 For example, think about your
 connection to the internet.

0:12:25.900000 --> 0:12:28.500000
 You're sitting at home, you have
 a connection to the internet.

0:12:28.500000 --> 0:12:33.360000
 That one cable you have leading out
 of your house can reach billions of

0:12:33.360000 --> 0:12:35.080000
 remote destinations.

0:12:35.080000 --> 0:12:37.340000
 So it's sort of like
 Ethan that way.

0:12:37.340000 --> 0:12:41.620000
 You could put one thing on there, one
 cable that gets you multiple places.

0:12:41.620000 --> 0:12:46.580000
 But is it possible for you to put something
 on that cable that all those

0:12:46.580000 --> 0:12:48.680000
 remote destinations
 are going to see?

0:12:48.680000 --> 0:12:54.200000
 No, it's not. In the case of your internet
 connection, you have to address

0:12:54.200000 --> 0:12:57.420000
 everything you put on that cable
 to who you want to send it to.

0:12:57.420000 --> 0:12:58.820000
 It doesn't support broadcast.

0:12:58.820000 --> 0:13:03.460000
 You can't put a broadcast on your cable
 modem or your DSL connection to

0:13:03.460000 --> 0:13:07.820000
 Time Warner cable or Comcast and expect
 that all the billions of devices

0:13:07.820000 --> 0:13:09.580000
 on the internet are
 going to see it.

0:13:09.580000 --> 0:13:11.880000
 It doesn't work that way.

0:13:11.880000 --> 0:13:16.440000
 So that's what we consider a non
-broadcast multi-access network.

0:13:16.440000 --> 0:13:19.240000
 But in all these networks where I can
 put something on a cable and I can

0:13:19.240000 --> 0:13:23.900000
 say to myself, well, there could be
 more than one device reachable on

0:13:23.900000 --> 0:13:27.660000
 this cable. In that particular case,
 a designated router and a backup

0:13:27.660000 --> 0:13:31.440000
 designated router will be elected.


0:13:31.440000 --> 0:13:35.260000
 And we've talked about these last
 two bullet points right here.

0:13:35.260000 --> 0:13:38.700000
 Okay, so how do we determine
 which router that will be?

0:13:38.700000 --> 0:13:45.220000
 Well, by default, the very first router
 that comes online on a cable will

0:13:45.220000 --> 0:13:46.480000
 be the designated router.

0:13:46.480000 --> 0:13:47.940000
 That's pretty much it.

0:13:47.940000 --> 0:13:51.360000
 Now, if two routers come online at the
 same time, if they start exchanging

0:13:51.360000 --> 0:13:58.240000
 hellos at the highest interface priority
 will become the designated router.

0:13:58.240000 --> 0:14:00.320000
 And this is the first time
 I've talked about this.

0:14:00.320000 --> 0:14:05.320000
 So an OSPF, when you use the network
 command under OSPF and OSPF says,

0:14:05.320000 --> 0:14:09.040000
 oh, here's an interface right here that
 falls in line with that network.

0:14:09.040000 --> 0:14:10.100000
 I'll start speaking.

0:14:10.100000 --> 0:14:12.120000
 I'll start sending
 out hello packets.

0:14:12.120000 --> 0:14:16.000000
 Well, on this interface, he gives
 it an OSPF priority value.

0:14:16.000000 --> 0:14:18.300000
 It's actually the value
 of one by default.

0:14:18.300000 --> 0:14:21.580000
 And so in my hello pack as I send out,
 that's one of the things I put

0:14:21.580000 --> 0:14:24.340000
 in there is what my interface
 priority is.

0:14:24.340000 --> 0:14:26.920000
 So if I'm sending hello packets to you
 and you're sending hello packets

0:14:26.920000 --> 0:14:31.720000
 to me, if one of us has a higher interface
 priority than the other, we're

0:14:31.720000 --> 0:14:35.460000
 done. The highest interface priority
 will win and that will become the

0:14:35.460000 --> 0:14:37.800000
 designated router.

0:14:37.800000 --> 0:14:42.880000
 And when you do the command show IP OSPF
 neighbor to view your neighbors,

0:14:42.880000 --> 0:14:46.320000
 you will see on like an Ethernet segment,
 you'll see at least one neighbor

0:14:46.320000 --> 0:14:51.120000
 shown as DR for a designated router,
 possibly another one that says BDR

0:14:51.120000 --> 0:14:54.580000
 and then others that
 will say DR others.

0:14:54.580000 --> 0:15:01.400000
 So DR others mean routers that currently
 are not the DR, not the BDR,

0:15:01.400000 --> 0:15:03.580000
 but they could be in the future.

0:15:03.580000 --> 0:15:08.360000
 If the DR and BDR fail, these other routers
 could potentially become designated

0:15:08.360000 --> 0:15:15.120000
 routers. Now let's say that you have
 not manually configured the interface

0:15:15.120000 --> 0:15:18.600000
 priority. And remember it's interface
 priority, it's not a global priority.

0:15:18.600000 --> 0:15:22.740000
 So on a router, if he's got five interfaces,
 if you wanted to, you could

0:15:22.740000 --> 0:15:25.600000
 have each interface have a
 different priority value.

0:15:25.600000 --> 0:15:27.700000
 But by default, they'll
 all be the same.

0:15:27.700000 --> 0:15:29.320000
 They'll be the value of one.

0:15:29.320000 --> 0:15:32.300000
 Well, if you and I exchange OSPF hello
 packets over an Ethernet segment

0:15:32.300000 --> 0:15:36.240000
 and we both have the exact same interface
 priority, then we're going to

0:15:36.240000 --> 0:15:40.660000
 compare each other's names, our
 router IDs, our OSPF router IDs.

0:15:40.660000 --> 0:15:45.000000
 And whoever has the highest router
 ID will be the winner.

0:15:45.000000 --> 0:15:49.660000
 That will become the designated router
 and the second highest will become

0:15:49.660000 --> 0:15:53.500000
 the backup designated router.

0:15:53.500000 --> 0:15:58.360000
 So a lot of times people don't really
 necessarily care which router will

0:15:58.360000 --> 0:16:00.120000
 be the DR or the BDR.

0:16:00.120000 --> 0:16:02.840000
 They just let OSPF take
 care of that by itself.

0:16:02.840000 --> 0:16:04.940000
 But let's say you want to have
 some control over that.

0:16:04.940000 --> 0:16:08.660000
 As the network administrator, you want
 to be in control over which router

0:16:08.660000 --> 0:16:11.400000
 was going to become
 the DR and the BDR.

0:16:11.400000 --> 0:16:15.240000
 Well, the easiest way to do that would
 be to go to the interface and change

0:16:15.240000 --> 0:16:17.460000
 the OSPF interface priority.

0:16:17.460000 --> 0:16:21.080000
 And in a video coming up where we actually
 go over the configuration commands,

0:16:21.080000 --> 0:16:22.580000
 I'll show you how to do that.

0:16:22.580000 --> 0:16:26.800000
 The second way you could do it is to
 by manipulating the OSPF router ID

0:16:26.800000 --> 0:16:33.960000
 itself. So as soon as you type router
 OSPF and then some process ID and

0:16:33.960000 --> 0:16:36.960000
 hit enter, it elects a
 router ID for itself.

0:16:36.960000 --> 0:16:40.400000
 It looks at all of its interfaces and
 says, I need to pull an IP address

0:16:40.400000 --> 0:16:45.120000
 here and use that as my
 name, as my router ID.

0:16:45.120000 --> 0:16:46.780000
 So what's the order?

0:16:46.780000 --> 0:16:47.580000
 How's it going to do that?

0:16:47.580000 --> 0:16:50.340000
 Well, number one, first thing OSPF is
 going to say is, all right, under

0:16:50.340000 --> 0:16:55.240000
 my configuration, under my router OSPF
 configuration, did some human being

0:16:55.240000 --> 0:16:59.580000
 actually type the
 router-id command?

0:16:59.580000 --> 0:17:04.140000
 If they did, that's what
 I'll use as my router ID.

0:17:04.140000 --> 0:17:07.700000
 What if you did not type
 the router-id command?

0:17:07.700000 --> 0:17:11.440000
 Well, then the router will look to see
 if you have any loopback interfaces.

0:17:11.440000 --> 0:17:16.100000
 If you do, whatever the highest IP address
 is of all of your loopbacks,

0:17:16.100000 --> 0:17:18.540000
 that will be your router ID.

0:17:18.540000 --> 0:17:22.880000
 And then lastly, if there are no loopbacks
 and there's no OSPF router

0:17:22.880000 --> 0:17:26.460000
 ID that was manually configured, it'll
 just look around at your interfaces

0:17:26.460000 --> 0:17:35.440000
 and select an IP address from
 that as its router ID.

0:17:35.440000 --> 0:17:37.860000
 Now interesting little
 factoid here.

0:17:37.860000 --> 0:17:41.480000
 A lot of times people think, oh, well,
 only interfaces that are actually

0:17:41.480000 --> 0:17:46.020000
 up up are candidates
 for the router ID.

0:17:46.020000 --> 0:17:47.860000
 That's not actually true.

0:17:47.860000 --> 0:17:52.560000
 Even an interface that is up down or
 down down is a candidate for router

0:17:52.560000 --> 0:17:56.800000
 ID. The only time there's an IP address
 on an interface, and OSPF says,

0:17:56.800000 --> 0:17:57.640000
 I can't use that.

0:17:57.640000 --> 0:17:58.820000
 I can't even consider that.

0:17:58.820000 --> 0:18:05.300000
 If the interface is administratively
 disabled if it is shut down.

0:18:05.300000 --> 0:18:09.920000
 Other than that, the IP address could
 potentially be used as the router

0:18:09.920000 --> 0:18:14.800000
 ID depending on if it's not a loopback
 and if it's the highest IP address

0:18:14.800000 --> 0:18:21.460000
 on the box. So that ends this video
 on the OSPF designated router and

0:18:21.460000 --> 0:18:23.720000
 backup designated router process.

