CCIE Collaboration :: Basic CUBE Setup

morning everybody welcome to another view in the CCI collaboration lab series we're looking at we're going to be looking at cube today just looking at the basic set up of cube once again thanks for joining there's been a little scheduling change and how these are working it's still going to be every Wednesday where these are going to be kicking off but we're going to be running them at 12:00 noon eastern time or 11:00 a.m. central which is where I am so anyway we'll talk about cube today we'll try to get off the ground on some of the topics in cube and just get the basic cube functionality running I do have a nice little PowerPoint for us to watch today it's really quick so we won't do anything crazy there but anyway we'll get right into it with the cube basic set up lecture so we're going to be talking about a few things today first of all we'll give an overview of cube and what it does and why we're going to use it in the lab of course it's on the blueprint so that's one of the major reasons the next thing we'll talk about deployment modes and how you might use cube or how you might be asked to use cube in your lab also we'll talk about the sit bind command and the different ways that can be applied I'll stalk about sip profiles and how powerful those can be in terms of changing things in the actual sip messaging and that'll be something that's that's very useful to you in the lab as well and also we'll talk about dial peers basically now don't let cube scare you it's just just essentially a iOS routing device I mean you're used to using dial peers on you know maybe cm II you know get your PSTN going it's the same thing cube is going to have dial peers that manage its dial plan and that's the exact way that we're going to configure cube so getting into first of all the overview of what it is first of all what does it stand for it's the cisco unified border element and that doesn't really tell you much about it except that is it is a border device meaning that it's going to be on the edge of your network somewhere interfacing with the PSTN provider or something like that it doesn't have to be a PSTN provider it could be you know the central point of your network is actually creating kind of a centralized directory structure so let me just go ahead and pull up a drawing pad so I can illustrate my point here so if we had our nice circle here which is cube in the middle and let's say that we had a call manager device over here and another call manager device over here let's call it HQ and site B these devices these communications manager servers are going to be connected to cube in some fashion this can also be or this can be an H 2 2 3 trunk for example this could be a SIP trunk for example or you know vice versa doesn't really matter so those are the options that we have when connecting cube then cube is going to just accept those connections from each of these 2 devices via a diol peer and then it's going to route from device to device using the dial pair methodology so this is basically all there is to it there's there's nothing crazy about this it's just two clusters being connected together through cube now you can get crazy with it you can throw the things in there like you can go to the PSTN for example maybe your your PSTN connection is a sip connection to a sip provider somewhere and your cube is actually going to foster that communication so it's going to say hey I'm going to connect to the PSTN provider in some fashion and the HQ cluster for example is going to connect to that PSTN provider by way of the cube and the cube acts as a border and between the two networks there so basically if I draw you know to illustrate the point if I draw just a nice little red line across this direction here this is the point of demarcation in your in your network here so for example we have this this PSTN connection going from cube the PSTN and then we have this HQ to cube connection these are completely independent of each other cube is going to handle that communication completely so let's say that HQ connects to 10.10 1.1 which is the IP address of the cube loopback zero so let's say that HQ says you know what I'm going to connect to that loopback address that's the that's the IP that we've configured for that well maybe the PSTN is going to going to interface with a public IP address so let's say that the cube IP address is just you know 888 which it's not that's the Google DNS server but let's just say that that's our cube public address the PSTN obviously at this point has no idea about this address let alone the address that's on an endpoint out here which could be you know 10 10 11 . 100 let's say so that would be the IP address the phone so 10 10 11 100 is going to communicate with 10 10 1 1 which is the loopback address of the cube router and then cube would communicate using its IP address over here to the PSTN so in that way it acts as a border between the two networks so it's kind of almost a protecting device if you want to look at it that way but it's a point of demarcation of the network which means that you're going to have signals that are regenerated by the cube and sent out as a fresh brand-new signal so and we'll talk about that a little bit later but I just wanted to give a quick little overview on that so what do we talk about just the dis then we've basically covered everything in this slide it terminates and originates signals and media streams of course we can go from h.323 to h.323 sip to sip h.323 to sip step to h.323 so if you guys are familiar with with the command that you might put in an iOS device for routing you know this this is looks very similar to this command this allow connections command so you have a voice service VoIP which is your global configuration on the router and then you're going to have a command section that that tells you to allow these type of communications between H 3 to 3 H 2 3 sippin H 3 3 so it you have to physically allow that in the router you have to go ahead and configure them so the next thing it's going to do for you is signal interworking we talked about that a little bit if we go back to the drawing here we talked about h.323 and sip being supported well we have h.323 here and maybe this is a sip trunk over here cube doesn't really care it's going to take each of those signals in and terminate them and then regenerate that signal to the other device so not a problem cube can take care of that so that's signal inter working their media in it working with DTMF fax modem codec transcoding basically DTMF if your DTMF methods are mismatched then cube can resolve that if it can be configured to resolve that also it can can perform transcoding for you if you have a codec mismatch so for for some reason let's say that you have your h.323 trunk on this side set to use g.711 only and then on this side you have your SIP trunk set to use g 729 well obviously at that point you're going to have a codec mismatch so at that point you can invoke what invoke a transcoder directly on cube and at that point it's going to resolve that codec mismatch for you so and you used to have to actually register that to the call manager Express on cube so you have to configure a quick little instance of call manager Express register that transcoder to it but now you can register it directly to cube itself so moving on here to the the next item here that dressed and address important translations obviously if this is going to act as a border element it's going to have to translate between addresses and different ports so that just is kind of what we've been talking about already it's going to act as the border and it's going to act as the demarcation point you have one address one port one address in one port over here and you're good to go so any questions so far basically and the question here basically it is a IP to IP gateway so that's that's all it really is really doing for us I mean it can also be configured you can have a ISDN PRI on if you like you know but it's really not used in that method often it would be mostly used for sip to h.323 sip to sip those kind of things IP to IP gateway a question here is the cube a security risk in any way based on its default configuration so it can be a security risk in some ways there there are there is a command that's actually on by default now in in iOS 15 one I don't remember exactly which I think it's 50 no or it turns it on but it's basically this command IP address trust authenticate and what that is going to do is it's on like I said it's on by default and many of you have already experienced this in your networks when you upgrade to 15 everything bricks for some reason you don't know why well that's the security mechanism kicking in and what that does is actually takes your dial peers and takes all these session targets meeting where it's pointing to the IP address it's pointing to on the dial peer and it's going to add those to a trusted list meaning that those are the only trusted devices that are allowed to communicate well if you don't have dial peers or if it's not used in that way then you you have things that brick so generally what i recommend doing for the specials especially for the CCI collaboration lab is to turn that command off with the no IP address trusted authenticate so that's a security topic there I don't know if that answer the question its I guess it's not necessarily a security risk you can you can firewall it just like you would any other device but the security mechanism that I recommend turning off for the lab would be that no IP address trusted authenticate so won't that make sense there all right all right so cube deployment modes so we'll talk about the different deployment moments that we we have available to us first of all there's media flow through and just logically speaking let's think about what that could possibly mean media flow through hmm well this is one of those things where you what you see is what you get cisco has said media flow through that means the media flows through the cube so there's really not much to it than that I mean and just just if you needed an illustration let me change colors here because this is getting a little messy if you have you know one endpoint over here like I'd already said 10 10 11 100 another end point over here connected to cluster B or site B you're going to have that media that would normally flow from phone to phone just like that you're signaling obviously would have to be set up by the signaling path meaning h cube cube SB they would have to actually set up your signaling portion of that but your media would flow from endpoint to endpoint well now in cube you can go through and say media flow through meaning media flows through the cube itself and there may be situations where that's not so great because you might have this HQ cluster which is in San Jose and this at site B cluster which is in Chicago and maybe your cube is in Florida or something now think about how your media has got to travel your media is going to go from San Jose to Florida to Chicago so it's kind of taking a roundabout path to get to the actual phone itself the signaling is it's okay to do that you know as we might have to use mechanisms such as early offer or age 53 fast start to overcome some of the signaling delay there but with the media stream at that point you're basically forking it through a different area that it really needs to go but that is media flow through that's what we're talking about so the media flows through the cube in that case so by the way that's on by default so that's that's the default mechanism for cube media flow through all right so the next thing we have here is media flow around and that's basically your default behavior of a phone normally if it's not your cube your media is going to flow around whatever the signaling path is so cube is setting up your signaling path and now we're saying media flow arounds that means let the endpoints actually communicate together and don't don't dictate where the media flows at all just set up the signaling path and get out of the way it's basically not let's say so only the signaling is terminated to queue at that point media flows between endpoints and it is recommended for enterprise networks of course because you you are within the secure bubble in this enterprise network right and you're able to communicate with this device or this device in the most efficient way possible in terms of routing how the routing setup you know your routing protocol determines that you know path XYZ is the most efficient way to get there so it's going to take X Y Z of course rather than having to go through intermediary device to take care of that media flow so and the media anti trombone is the other thing too and this is kind of interesting and I think I'm going to have to get a new drawing for this one here folks so the media anti trombone here let's say that this again is cube in the middle and we have a PSTN cloud just a gorgeous cloud there and we have well we'll just say one communications manager so the HQ cluster that's connected to cube let's say via sip doesn't really matter for the illustration purposes so we have a we have a phone that's over here connected to HQ we also have a PSTN phone over here or it doesn't have to be a PSTN phone it could be another network somewhere but just for this example we'll say PSTN so we have a call that goes out to cube it says I want to hit this phone over here whatever extension that is so cube says I know exactly where that's at that's on the PSTN great well now this phone itself has a forwarding rule and I've kind of let me get out of the way here folks sorry it has a forwarding rule set up to go back into the cloud and come back through cube back to this phone or another phone on the network let's say let's say it's forwarding to this phone over here so at that point what that means is that you're basically coming back in the network you know you're you're you're taking that call cube is sending it out and it's coming right back in so what that's doing is kind of looks like a trombone right you're kind of going like this coming back in the network so that's why they call it a trombone now if you say anti trombone cube is actually going to be smart enough to say hey wait a minute this call is forwarding back in and kind of take note of that and rather than having to set up the signaling path all the way through this area it's going to cut it off right here at this key at this point here and just keep it keep it within the network itself rather than having to go out and use another resource of the network so that's the anti trombone mechanism there alright no questions and that looks like we're good okay obviously feel free to ask questions at any time guys so all right so now the SIP bind command now we're getting into some very important stuff here the SIP bind command is going to be where your process is running on the cube itself it's going to pick an interface for this process to run now the SIP bind command is useful in other things for example CME voice register global you know configuration on CM e for sip phones but also in cube it's going to say where your sip packets are going to be accepted so if your sit back 'its are going to be accepted on the loopback interface then you're going to have to bind your traffic for sip or h.323 where whatever you're accepting to the loopback address so in the example of sip because sip is what you're going to be using most often if you have a choice ever in the lab use sip please just save yourself some some headache guess h 2g 3 to sip can cause some issues of course there are ways to resolve that in queue but why have to worry about that in the lab if you don't have to so keep it zipped to sip if you can anyway so it attaches itself to the the SIP process to a specific interface and you can do that a couple different ways you can do that with the global bind command which is under voice service void you can see that down below here on avoid service avoid sip so voice service VoIP is your global voice service configuration then under that we've said we want to configure sip and then from there we say we want to bind our control or media or both or bind all by control the source interface of fastethernet 0/0 in this case so we want to use fastethernet 0/0 IP address as the one that's going to be connected to for the sip process we also can can use multiple different bind interfaces so for example if you did want to want to use a cube to act as a demarcation between your network and the PSTN network maybe you want to use different bind commands for that so now you can do that at the dial peer level itself so if you wanted to go from the HQ cluster to the cube on the internal network you can have that bound to a specific dial Peter incoming and outgoing we'll talk about that and also if you want to go to the PSTN you could have that specifically bound to another interface on the router so it doesn't have to be the same interface for everything globally of course it's you have no choice that's the same interface you can also you can apply a global command but then you can also get more specific and apply a dial peer command of course the dial peer command at that point will take precedence over the global command so otherwise it's going to choose the best interface available meaning the best routed interface so whatever the path is most optimal or wherever the path is most optimal is going to choose that interface to go out and that's not always necessarily the best thing to do because there may be different interfaces that it chooses to go out based on how the routing mechanism has changed or the routing updates have changed or whatever so we want to always try to make that these predictable interface so that way the devices that are trying to communicate with it don't end up having a 404 error because they can't find the other device so ok so now we've got our bind control or bind commands figured out this bind control by media of course it's going to bind each of those different processes whether the control or the media to the interface itself so that's how that's going to work for us and here you can see the configuration committees boy service boy global voice service configuration sip and then we're binding our control in this to the source interface of fasting at 0-0 and then there's the dial peer example it's kind of a mouthful voice - class sip by control source interface fastethernet 0/0 so there's a lot to that of course but feel free to question mark because it makes it a lot easier so bind command very important for connecting your your devices together in the network any questions so far are we looking good okay all right so moving on now to sip profiles sip profiles are super powerful you may end up having to use them in the lab to resolve differences in codec or DTMF relay type or maybe just to do something trivial like change the name of a caller ID or something like that I personally have a lab like that somewhere buried in IP expert stuff but anyway so it allows you to customize sip messaging that's the whole point of a sip profile and and why it's so powerful is that you can put this device cube in the middle of the media path or signaling path and say whatever I got in from this person or this this device that person now we're getting from this device I'm going to change in this method so it's going to search for a string and replace that string is all it really comes down to so just like a voice translation rule you'd be searching for a dialed string and you're going to replace that with another dial string somewhere so with cube you're basically going to be able to do this on a global level or a dial peer level so it's going to allow you to do it very specifically to a certain endpoint or you know maybe if you don't want to get specific with it make it general for every endpoint so keep in mind also that when you look at the actual sip messaging coming from one endpoint to another you see you're going to see the the communication come in from that sip endpoint and then after the SIP profile is applied then you'll see the information go out as you've configured it so for example if we've got another drawing here if we have one device over here HQ cluster again to the egg-shaped cube and to site B and then the other phone over here so the SIP profile let's say that I I wanted to change the calling name actually have this in the demo so that's this is perfect HQ phone to let's say that the calling name is HQ phone to and I want to change that calling name to be SIP profile just to illustrate the purpose of the SIP profile hqp too so cube is going to actually change that for us and then as it's received on this phone over here the display will this will display this name here sit profile hqp too so that's something we can do in a set profile on cube so as its received here we're going to receive the name as HQ phone - it's not going to change what was received it's it can't change that it's you know it can't modify anything that's already gotten so it's got to take in something from cube so at one point you're going to look at the bug and scratch your head like man why didn't it change this this doesn't make any sense well it's because it received that and it can't change what I received so it's going to take what it received then which is going to be a sip invite we'll talk about that messaging in a little bit because that's very important as well but it's going to receive an invite to cube and then it's going to say okay I've got this string here that says HQ phone - I'd really love to change that to sit profile hqp - so depending on where that's been applied whether the dial pier level or global level it'll change that as it goes out to site B so we would we would go through a series of configuration commands - to look at that and apply this at profile accordingly now those configuration commands look like this so we have this one's actually changing the you abandon info here but this is voice classic profiles 20 so that's just an arbitrary number that was chosen for the SIP profile number we've also got the request invite and we're changing the SDP header and everybody remember what the SDP header is right that is going to control your media negotiation control your capabilities between sip endpoints so that's how it's how it's going to control that with with h.323 would be h 245 negotiation would be your media nowis if it's going to be you know STP headers that's going to control that so that we're modifying here the invite the actual invite message that goes out or saying the STP header in the invite the video bandwidth info field we're going to add information we're going to say B equals a s : 4,000 so we're modifying the bandwidth for that specific STP header that goes out depending on where you apply it so you can apply that globally with the voice service port command sip - profiles 20 of course referring to the voice class that profiles 20 also the dial pure voice 20 voice like this would be a dolla pure level command for us and then you can apply that under there with the voice - class command so voice - class sip profiles alright so any questions so far looks like we're looking do it again alright now there's a couple other things I want to talk about of course but this is a great reference material for for kube so if I pull this up real quick I recommend going through this and really understanding this this document really well because this will explain everything that I just kind of went through but there are other things of course that we're going to talk about later that may not necessarily be in here but dial peer matching we haven't talked about dial peers yet but if you're not familiar with that definitely get in to dial peer matching and check it out because that's going to be a very fundamental process of how cube is going to route the call talk about sip profiles media paths that kind of stuff so definitely check this out I'm not going to go through the document right now but I wanted to actually reference it so at least you knew that it was out there so okay so let's get into some demo mode stuff that way you can stop falling asleep on me and all that kind of stuff so we're going to use in this case we're going to use r1 as our cube device so r1 will be the cube between the different clusters that we have and I happen to have three set up which is going to be HQ SB and also the co-manager express device which is going to be the r3 router so r1 the first thing we're going to do on cube is let's say you get in your lab and you say oh there's a cube here what's the first thing I need to do if there's a cube well I've done the command for you here I love you show run section voice service to show you mode border element the command needs to be entered for your cube to function properly so the bummer about this mode border element command is that you once you enter it it's going to ask you to reboot so that's one of the reasons why I've already entered it here for us so I have to wait on a reboot but Mode board element under voice service void is something that should be entered to make sure that your router functions as a cube so mode border element enabled right there also what I mentioned earlier thanks for the question by the way on that the no IP address trusted authenticate so this is going to turn off that authentication for four different devices coming into the router so if there's a device that's not recognized it would drop the communication and you wouldn't know why so I recommend just turning that off by default it's going to save you a lot of time and energy also we talked about these commands allow connections h.323 to h.323 h2 sip sip to h HT h or sip to sip at the end so that's one of those commands also that you should probably just pop in the router by default because why deny something unless they're specifically asking you to deny something why deny something in the lab if you don't have to so I recommend just putting those in right off the bat if you if you have the opportunity to do so then obviously they may ask you to say a make sure that you don't allow h.323 to age 53 connections well at that point you're going to need to remove that command from there because that's what they're asking so anyway moving on here the SIP configuration under voice sir boy busier next thing and notice we have a global bind command done for control and media using the source interface will loop back zero so if I look at the loop back zero interface that happens to be 10 10 1 1 so now we've said that we're binding our sip process to this loop back zero interface for both control and media traffic by the way so we've said that that is is going to be where sip process is running so we can also define that in a diol peer level still we can override that loop back binding with a specific dial Peter if we'd like to do that we don't have to do that though now everything every dollar you create is going to be bound with this source address of the loop back zero now a couple of other commands see I have a sip profile assigned here it will look at what that sip profile does but this command that I did not mention in the PowerPoint presentation is passed through content SDP very important for video communication through the cube now why is that you ask well this might be the perfect time to talk about sip signaling so let's let's examine what what happens in a sip signal or a sip communication between two endpoints so let's say you have n point one and point two and you know what we'll throw in the cube obviously because we're talking about the cube so the first thing that's going to be sent out let's say that n point one is calling endpoint to what is the first thing that we're going to send out and invite right we're going to be sending out an invite Q cube right because we're flowing through Q at this point we we've got our signaling set up to point towards the IP address of Q so we're sending out an invite now next what are we going to do or what's cube going to do cube is actually going to take that invite and say oh thanks for the invite I see that this is pointing towards endpoint two so I need to send this invite to endpoint two so cube then is going to regenerate that signal and say invite ten point two so now endpoint two knows about it so in the meantime while it sent that invite or invite out it's also sent back to endpoint one a message saying that hey I'm working on this right trying 100 trying I'm trying to get this information for you I'm trying to ring that other side and so forth so it's it's a it's a nice message I guess it's an informational message is what it really is informational provisional message but it's in that it's a nice message if you really want to look at it that way so you can remember it it's saying hey I'm working on it and so if you wanted to be even nicer if if you wanted this guy for example to be even nicer you could turn out in Prak which is a proactive acknowledgement meaning that it has to respond to anything with a 1 X X so anything starting with a 1 it would need to respond to so if you're if you're saying 100 trying this has got to respond or endpoint 1 is going to respond with an AK in that case this is okay so that's even more polite but don't need to enable that we don't need to enable that for the cube communication so sorry I don't mean to confuse you guys there hopefully that didn't but more the story we're standing out in invite the cube cube is that sending out that invite to endpoint to meanwhile cube says 100 trying back to the original endpoint that called and now once once endpoint to receives this it's going to actually receive one a 100 trying back cube is from endpoint 2 so that's great so now we know that endpoint two is attempting to communicate with cube and some in some way so that's great so now the next thing it's going to happen is either we're going to have some type of error or we're going to have ringing which means that the phone is actually ringing if endpoint 2 is located the actual phone itself but wants to ring and it's ringing it so at that point we will get a message back from endpoint to saying 180 ringing and by the way these these messages here trying and ringing they don't have to be that they're really all they're looking for is the number itself here so 180 or 100 Communications managers just puts in trying and ringing so that's what we'll always see but it doesn't have to be that to say if it's not a hundred trying it could be a hundred Andy is cool you know but no one would ever put that in there unless they really wanted to but so hundred trying 180 ring comes back cube receives this 180 ringing and now it says ah it's ringing I need to forward that on to the endpoint one in this case so 180 ringing is going to be forwarded there so now one is actually going to get Rin back tone and said say oh it's ringing at that point great so the next thing that happens is obviously we're bringing phone to full one's already sitting there waiting for some somebody to answer on phone - now when phone - answers then that's where we get the actual media negotiation in delayed offer meaning that the the STP or the immediate negotiation happens at the point where the person picks up the phone rather than the point where the person sensed sends the information to ring the phone so basically if I pick up the phone here on on endpoint - in the late offer which is what this is I'm going to now send a 200 okay and that 200 okay is going to contain an SDP an SDP is session description protocol and that's going to describe the media capabilities of the endpoint so when it picks up the phone it says I have these and for these media capabilities to use to speak to you with and it's going to say I have g.711 G 729 I'll be C whatever other codec that it has it may only have one codec so it'll send the only one codec to to negotiate with so but it's going to send its capabilities to cube cube then it's going to say I've got capabilities coming in I've then got to regenerate that and forward it out to endpoint one so then endpoint one gets those capabilities and says I see that you have the ability to run g.711 G 729 but I would actually just want to run g7 29 so they negotiate at that point g7 29 and it sends an ACK back to cube cube will then regenerate that AK 10.2 and that is your sip communication that's that's all you're delayed off or communication I should specify that delayed offer and it sounds bad but but that is you know that's just the way they describe it delayed sounds like I don't want things to be delayed well it's just the way that they describe it the offer is actually coming you know the offer meaning the SDP is actually coming from them from endpoint to when it picks up the phone so that's why it's delayed now an early offer you can have the SDP which comes out first here at the 200 okay you can actually have that come in the invite from the original phone so that way it says hey look I want to call you and I want to use g7 29 it puts an SDP in the original invite so that way when the phone responds with a 180 ringing or a 200 okay that is when it when it actually picks up the phone it says okay I'm okay with those codecs and at that point the media is set up without having to wait for that ACK coming back from the original phone one so Delia offers early offer there but what we're really interested here in is the SDP so when we when we send this SDP information that's going to contain obviously codecs so audio codecs like g.711 G 729 but also with video phones is going to contain codecs like h.264 h.263 whatever so it's going to send those codecs over the cube basically and if you don't have this SIP pass-through SDP command or pass through content SDP it's not going to allow those media capabilities to pass through the endpoint so you would have to negotiate that on the dahle peer level at that point so if you wanted to negotiate h.264 between endpoints it's going to allow that to do so with this pass through content command so you need to have that enabled if you want to pass through the SDP and negotiate meeting capabilities with either endpoint so very important command there because if you don't have it enabled you will not be able to get video up across the queue it just won't work you have to pass through that SDP so very important command there already now let's look at like it like I mentioned here the SIP profiles as well but let's look at first the actual dial peers and how we're going to connect here so if I'm going to connect you know what let me just draw the network too so that way we're kind of good on how this is all laid out so once again HQ device we've got HQ connecting to cube site B and these are all call manager servers except this guy r3 is our CME or C you see em E and that's going to going to be connected to cube these are all connected via sip like I said if you you have a choice use all sip it just makes your life easier rather than having to do this signaling inter working on the cube so we have all SIP trunks basically you know the HQ to cube connection and and cube to SB here this is a SIP trunk from the call manager perspective on the cube obviously it's just a dial peer but we need to create a SIP trunk that communicates with it with this cube itself so I'll show you the configuration for that if you if you've configured a SIP trunk before it's nothing different it's basically just a standard SIP trunk with all the all the default configuration but we'll look at that anyway so we bought we've obviously got this cube running on 10.10 1.1 for every direction we had that enabled on the voice service VoIP command as a global configuration command so that means that the cube is running on 1010 one one the loopback address so all our SIP trunks then are going to point to that address 1010 1 1 including the dial appear over on CM e so the actual dial plant information here we have 2 1 xxx we have 3 2 X xx + 4 3 X X X so those are those are going to be our ranges so if we want to communicate using 5 digits so I would dial this five digit number of CM e for example so 4 3 0 0 2 to get the videophone on CM e and I would dial to 1 0 0 1 to get the videophone at HQ and so forth so we need to create dial peers to accommodate this dial plan so now that we've got the kind of basic layout let's look at the dial pair itself and how that actually works so if you show run section dial alright first of all we have what we go down a little bit let's say that we are communicating to sightsee or the our 3 cm esight we've got a dollop here I've named it now pure voice 43,000 voice so in this case we're just saying for 3,000 this is an arbitrary number but try to name them something that that's familiar to you this is just the way that I need because make sense to me maybe whatever you want that make sense to you there we're saying destination pattern so destination pattern is going to be the actual pattern that you're going to use to route to that destination so if you you've dialed for three thousand one it's going to say oh that pattern matches this for three dot dot dot and I'm going to send that out now to whatever the session target is so the session target would be this 10 10 31 1 so that would be a place where the SIP processes running on a device out there whether that's communications manager or CMA it doesn't matter happens to be seeing me in this case session protocol sip v2 is going to set the actual protocol that's used so you can do a session protocol sip b2 session protocol Cisco for h.323 which is the default so whenever you see a sip dial pair you always see this because it's not the default so that's good to know when you're going through it also we've got this other thing here incoming called number 4 3 dot dot and also I to mention here we have a dollar sign which is going to signify the end of the string in regular expression form basically so it says I dial for three dot dot dot and no more nothing else can come after that that string if you don't put that dollar sign you can have an an overlapping dial and basically if you have an overlapping dial and it will match both of the dial peers then maybe that's not what you want so I'm in the habit of using that dollar sign to just lock it all down make make sure that that is in fact the end of the string so I'm saying for three sign so for three zero zero one that's the end of the string nothing else can be dialed at that point so but back to the incoming call number incoming called number for three dot means if that number comes into the router or the cube in this case with this dialed number string for three dot dot match this dial pair and why would we want to even have this commanded here well if you think about it if we don't have an incoming dial pier then we're going to be matching dial P or zero which has all the default configurations so vad would be turned on in that case voice activity detection of course you want to always turn that off because that is bad that's all what everybody always says so why not continue that that is bad so we we want to make sure that we are matching a dial pair that's not Donald pair zero so we can actually use the settings that we want to use so we could definitely break this out into different dial peers if we wanted to we could say incoming called number four three dot dot and put that in a different dial Pierre at that point we wouldn't we wouldn't need the destination pattern but we would need codec transparent and no bad meaning that we're allowing these endpoints negotiate together and that's what Kotick transparent is if a if a communication passes through an endpoint through a dial Pierre is and it has codec transparent on it that means it's not going to attempt to terminate that set that signaling session it's going to allow the endpoints negotiate the codec together so it's going to pass through whatever capabilities or media negotiation has been sent through that dial periods it's not going to attempt to negotiate it for you so codec transparent will be for that reason and also like we talked about no bad so but we could break this out we could say session protocol sit be - we wouldn't need the session target command we wouldn't need the destination fat of command it would only need these last three incoming call number called a transparent no bad so that would be our incoming dollop here at that point but if you want to save time why break it out and - I mean just put it in the same dial period if you can if it's feasible to do that so in this case it is because I want to use this as my incoming dompier and then when it goes out of the router when it tries to find a destination to go to it's going to use this destination pattern so match those up accordingly and then you don't have to use multiple developers for that so next thing so that means that that's going to allow us to go to this dial here or to this to this destination so 10 10 31 1 so if we go look at that on the CM e side and we look at show run section voice service and we've got our bind source interface VLAN 31 so let's check out VLAN 31 VLAN 30 one is our 10 10 31 one interface so that's what we were talking about if we go back to cube here 10 10 31 1 so we're pointing towards the interface that's bound to the SIP process here under boy service voice on the CME router so we're pointing to that interface and of course we need a dollop here on the CMS side so if I look at that I've got now this dial pure voice to VoIP and I've got incoming call number dot so I'm allowing that then for any single number that comes in this router any any any number that comes in as a VoIP call can be is or will be assigned to this dial peer so it's just a general dial up here here we have a voice class codec command with a couple different options and it also have no VAD again and also we're setting the session protocol to sip so any sip call that comes in here with any number is going to be selected this type he will be selected and from there of course route to the phone itself so the phone being either a skinny or a sip endpoint so anyway back to cube so that that's the doll appear to get there so that means that any call that comes in from any number whether that be HQ or site B we can go out to stop here because the incoming called number will be four three thousand one from HQ or site B so that's the match on the incoming and then we have a destination pattern for three thousand one so we're going to use the same dial peer to go out there now in much the same way we need to do that for both the HQ site and the site B site now if you want to add redundancy in there what I've already got that so when I had redundancy there for the HQ site we have to add to dial peers and you can see those right here I've got the two dial peer selected now destination pattern to one thousand 1 - 1 - 1 . . you know so the same destination pattern in in the dial pair itself but one of them has a preference one one has a preference - so the the dial Peter with the preference one is going to be selected first so the lower preference wins so it's the preferred one in this case a preference one it's going to select the stop here meaning that it's going to make the communicate or make the connection first with this server 10 10 13 . 12 so it's going to say I'm going to send out the SIP communication to this subscriber first if I don't get a response then I'm going to try to use this server which happens to be the publisher on the side so this this is diapir logic that's that's of course been around forever but bit on since see me as well so you've used this in cm e probably or HT two three gateways depending on what you've set up so once again same settings here we've got our session protocol sip because we're communicating via sip we've got codec transparent no bad we're good to go there nothing else we really need to do so the next thing of course as you could imagine is we have to do the site B that appear and then it happens to be right here this dial pure voice 32,000 void so we're it's just a dial up here pointing out to the other cluster so the site P cost for ten ten twenty three eleven so great okay so now we've got everything set up we've got our voice service VoIP our global configuration set up we've got our dial peer set up we've got our destination patterns and everything set up to point to the right places now all that's left to do is test the thing so if I'm calling from let's say I'm looking over my phone's here from two thousand to one zero zero one two three two zero zero one so first of all let's go ahead and turn out a debug so we can see what that looks like and I've already got one on but that would be we just turn it off so you can see what the debug is I'm using this is the best debug ever that you're interested using the lab it used to be debug ist and q.931 was your your best friend but now this guys are going to be your best friend debug CC sip messages that is the best debug that your you're going to come across in the CCI collaboration lab so show debug we've got the debug on now and I'm going to make sure I am logging or I'm not logging to the console because there could be a lot of information all right so I'll do a quick show log check out the buffer size and it's going to be sufficient enough it looks like so let me clear the log and now we're ready to make this call so we're going to call from two one thousand one two three 2001 all right so we're calling now that's ringing of course let me answer that all right so I left the call up there so the call that has now gone through the cube and we we are up and running basically so let's have a look now at the actual debug so show log obviously the first thing we see here we've got an invite an invite coming through so we've got an invite going to three two thousand one and of course you see here 10 10 1.1 and that is the who bag address of the queue so that's it was sent from HQ to the cube so we can see the direction that communication so it's coming from HQ phone one and this is T 1001 at ten ten thirteen dot twelve so it's coming from the subscriber so that must be where the phone is registered so we're saying we're saying that this is the source address now so two 1001 1010 thirteen twelve is where it's coming from two three 2001 at 1010 one one notice that this is the IP address of the cube it doesn't know where this phone actually resides it only knows that cube is the method by which you're going to gain access to that phone that's the only thing it knows now obviously if I look at this communication it's going to be hard going through to tell the direction of this communication the way that you're going to be able to tell is using the IP addresses it's going to help you out because this from and two is not going to change the call is always going to be from HQ phone one to this other phone three two thousand one here side B phone one so it's always going to be that in that direction so the way you turn determine what device is talking to what is via the IP address because site B doesn't know about this IP address at all n HQ doesn't know about site B's IP address so basically it's going to use this cube address all the time so anyway so we've got the invite coming out here we go through and we see up we get this centered here the content length on the invite is zero so that means we don't have anything following the invite we don't have an SDP here at all so this is delayed offer this is called delayed offer so if you saw a content length here and you saw an SDP then this would be an early offer call so now we have we have the late offer in this case so that content length zero so that means we're going to expect media negotiation to happen once the other end picks up the phone when we receive a 200 okay from that other end so now now we see this 100 try and come back and to determine the direction of that obviously we're going to use the IP address to do that we've got see this from that we see on cube first of all it's sent it sent a 100 trying and where did it send it it sent it to 13 dot 12 so that's how you determine where it's actually going what direction so obviously you still got the phone look at the to address in there so it sent a hundred trying back as a way of acknowledging that I got your invite I'm going to try and look into it so next thing it's going to do is send the invite out to the far end so it says all right now it's time for me to do my job as the cube and send out the call to the far end so in this case it's going to say 3 2001 and it's going to be 10 10 20 3 11 and it got that from the session target of the dial pier so that's how it knows what afford that those dial peers have dictated where this call gets forwarded 10 10 23 11 so we've got now the the from here has changed you see it's from still from 2 1001 but it actually has this 10 10 1-1 address which is the loopback address of the cube in this case so it's using that to to basically hide the address of the HQ side so it only knows about the cuba dress so it's so we know now that if we ever see 2 1001 at 10 10 1 1 that means that it's that the cube is talking to site B because site B doesn't know about any other address so the cube now 3 2001 at 10:10 23 11 so we know that that's sending to site B so it's talking to site B directly all right so once again invite goes out we see that there is a content like the zero so there's no early offer and then invite at all you actually have cubed generate an early offer for you if you want it really offer forced command we didn't do that in this case so that's the content length zero and now we receive a hundred trying from the far end meaning let's see from 10 10 23 11 so we know that that's coming from scipy at this point all right so now we received a 180 ringing so it's it was trying now it found that endpoint and it's actually ringing that endpoint so we get the message on that so we receive the invite from site B than it is ringing at that point and now we're going to send that invites of sent a 180 ringing out to the HQ device 10 10 13 12 so it's doing its job perfectly now it's it's taking the invite or the the 180 ring that it got from site B that's going to send that out to HQ so now at this point we know that like the phone is actually ringing we're getting ring back on the phone that dialed the number and we're getting ringing on the phone that we called so now the next step it would be for the phone to actually pick up the call and that's where we receive this 200 okay see if we're 200 okay from 10 10 23 11 so that's coming from the site B cluster towards cube again so we received that on cube and now we see this content like 233 which means that there's actually something attached to this 200 okay so if we look at it we see the media capability negotiation so we see that these these media capabilities are being discussed so if you look at this carefully we see that this is a sip call the connection IP this 10 10 21 252 it's actually going to be the IP address at the endpoint so that's that's going to be the where the media is going to flow at some point we have bandwidth modifiers for 8008 meaning G 729 in this case but you can really tell this M equals audio here that's 3 2 3 1 2 that's actually to refer to the port number and there's RTP a VP 18 that doesn't make a whole lot of sense what is that so this is actually if you go down to the next line AE equal RTP map of 18 which is G 729 so it's basically saying that I have G 729 to offer you in this end is 200 okay and there's no other codecs there so I'm saying G 729 is basically in so you have to accept that we're going to have a problem so anyway so we move on here and we we see that this this 200 okay is now sent to the HQ cluster that - 1001 and we've basically forwarded that that request or that that mediate negotiation here in the media capabilities G 729 we forwarded that right along and now to complete the communication we receive an ACK from the HQ cluster 10 10 13 12 and we're going to say you know what G 729 is completely cool with me I'm okay with that being the codec that we use and then now Qube is going to send that ACK along to the site B cluster and there's the there's the codec that actually sends along to site B and that's it you see that that's the end of the log because that's the entire communication that happens between those two endpoints at that point so meeting negotiations obviously one thing we didn't see here is video so let's actually take a look at a video call and see how that differs from my audio call in the SDP negotiation so let me clear this log first of all on the end the call and then I'll clear the log all right give the long a buffer now I'm going to call site C for 3,000 - that's going to allow me to share some video here so for 3,000 - all right so I've got video going now between those two phones between HQ phone one side see phone too so let's have a look at that and we'll go through this a little bit quicker just to get to the SDP so get invite coming in to cube of course from HQ we're sending an invite out to sightsee so cube is now regenerating that invite we're sending a 100 trying back to HQ because that's that's what I need what needs to happen it needs to let HQ know that it is trying it we've received 100 trying now from site C from the r3 device and now we received 180 ringing from the r3 device we forward that 180 ringing on to the HQ cluster and now we receive a 200 okay from the r3 device with the content length of 366 that means there's an SDP here so we look at that and now we see that there is our video information there as well we had this M equals audio line before on this port number and now we see that we can actually use video with this M equals video line and that's using port 1 7 0 at 0 this other thing here is the payload type 97 this is a dynamic payload type between 97 and 126 it can be but 97 is chosen here at random but it's usually 97 is what it ends up as I've only seen 97 and 126 be used anyway so 97 being the payload type here and the connection information so 10 10 130 1.2 so that's going to be the IP address of the phone there the RTP map now so 97 we see RTP AVP 97 is actually referring to h.264 so h.264 in this case is the video codec so once again this is the 200 okay received from our 3 2 cube now what we've got to do is for that it back out to the HQ device and that says the same thing we've got the RTP map 97 we're saying sending H 260 or we're only offering that codec and then now the HQ side is going to say ACK I completely accept that codec I'm cool with that h.264 we're good and then we get an act back sent out to the r3 device and that's it we've got our video call up now because of this now remember there's a couple things that we did to get that video fall going if we do show run section voice service pass through content SDP definitely need to have that command on there to allow it to pass through cube so that's how the video call will get up between those two devices at that point any questions so far doing okay okay so with that there's really not a whole lot more to talk about it's basically that's that's the basic setup a cube allows you to get the video calls going allows you get the audio calls going we basically just use dial peers we've used H through G or H 2 to 3 or SIP trunks in this case we just use sip one more thing I do want to show you too that I didn't show you yet is the actual configuration of the SIP trunk and this will be the same on both clusters both HQ and site B that is and there's really nothing to it and if you configure the SIP trunk before it's just like any other SIP trunk go device trunk and I've already got it created and this happens to be the cube SIP trunk I can't stress enough how much you should put this in a specific device pool you can control codec negotiation so much better if you have the trunk to assign to a specific device Bowl because your region is assigned to the device pool and this is something that you'll see as a recurring theme you know throughout trunking configurations is that you need to have a separate device pool for that trunk you can control you know which servers are going to be connecting to which devices you can control the codec negotiation which is probably the most important for the CCI collaboration lab you know if you are asked to have a specific codec across the cube trunk then you need to be able to have that in specific device pool so definitely important to have a specific device full here other than that everything else is is straight up we have this calling search space you know like I said recommend calling search bases and partitions all the time it's not a gimme where you can just use the none partition anywhere really just put put get in the habit of using partitions and calling search spaces it's going to help you out in the long run the gateway trunk calling search space here just has access to the internal phones and I've have all four significant digits so of course you saw that we're using five digit dialing between these two sites so that is just like any other gateway we're going to accept the digits coming in and we're going to forward them on to whatever the partition is in this device that it has access to so that's for inbound calls of course as we keep scrolling through we've got our destination of 10 10 1 1 which was the loopback address of cube and we also did the SIP trunk security profile non secure SIP trunk profile and standard sip profile and like I said that is it there was nothing else to it I mean maybe you were expecting oh we have to put in something clever here in the SIP trunk security profile or maybe we have to figure out something for the SIP profile we don't really have to at all those are the options you can add of course but for your basic setup for cube there's really no need to do that so this is the actual trunk configuration here on communications manager on both sides whether it be HQ recipe so any questions there all right all right we're looking good so guys I really appreciate you coming coming to the lecture today obviously if there are any questions reach out to me a vas or at IP expert comm and love to help so if you have any questions or anything you can think of outside of the lecture please let me know otherwise I really appreciate you come today and I will see you guys next week Thanks

Channel: IPexpertInc

Views: 32,043

Keywords: CCIE Certification, CCIE Collaboration, CCIE Training

Id: M7y62dQN3S0

Length: 64min 56sec (3896 seconds)

Published: Wed Aug 20 2014