|
0:00:13
|
So let’s take a moment to look at Network Infrastructure Basics and IP Phone Registration.
|
|
0:00:22
|
Now we’re going to look at some acronyms and Basic Network Infrastructure Services and I should probably start off by saying, if you’re not used to too many acronyms
|
|
0:00:33
|
which you probably already, or if you’ve gone through CCNA Network Routing and Switch, it's time to get used to them because in the voice world,
|
|
0:00:43
|
we are going to have just myriads of TLAs or three-letter acronyms, well, actually, a lot more than three letters.
|
|
0:00:52
|
So first of all let’s take a look at the necessary pieces of infrastructure and the first thing starting out is Power.
|
|
0:01:00
|
We need to be able to provide, obviously, a switch infrastructure over Ethernet for our IP phones to connect,
|
|
0:01:07
|
but we need either PoE which is the Industry Standard 82.3af or 82.3at for a little bit more Power,
|
|
0:01:19
|
but Power of Ethernet, or else the predecessor before the RFC was ratified, which was Cisco’s Proprietary Inline Power.
|
|
0:01:30
|
We will also take a look at VLANs or Virtual LANs, reason being, we can pass information
|
|
0:01:37
|
via CDP or Cisco Discovery Protocol to Cisco IP phones telling them which voice VLAN they should assign their internal switch port to.
|
|
0:01:49
|
So, if you have a phone with only one Ethernet port on the back going to the switch or you happen to have a phone with two Ethernet ports on the back,
|
|
0:01:58
|
one going to the switch and one going to some sort of a computer, macro PC,
|
|
0:02:04
|
then there is always another hidden switch port
|
|
0:02:10
|
and that is the internal switch port that actually goes to the phone itself and the Voice VLAN is assigned to that if it is present.
|
|
0:02:18
|
If the Voice VLAN is not specified and not present, then it will use the Access VLAN.
|
|
0:02:24
|
And of course, whatever the native or PC or access VLAN that is in untagged.
|
|
0:02:31
|
This is using 82.1q VLAN Trunking Protocol and the Access or Native VLAN will be untag and that will be what all of the native untagged frames,
|
|
0:02:46
|
Ethernet frames coming from the PC or Mac attached to the phone’s PC port will ultimately use.
|
|
0:02:53
|
DHCP or Dynamic Host Configuration Protocol, of course this provides all the parameters for the phones
|
|
0:03:00
|
and possibly, you know, depending on where you assign it for the Macs or PCs attached to those phones to operate on the network
|
|
0:03:12
|
and it should be noted here well briefly that if you’re using a Cisco Unified Communication Manager (CUCM), its DHCP server, first of all, it is only to be used for phones.
|
|
0:03:22
|
It is not intended or supported for any additional computers or any additional nodes in anyway, and it is not to be used for any deployments over 1000 phones.
|
|
0:03:34
|
We’ll take a look a lot more in depth at these but we are just taking a look at the general overview.
|
|
0:03:38
|
TFTP or Trivial File Transfer Protocol. The idea of trivial, there is no username or password.
|
|
0:03:44
|
The host that is requesting the file simply needs to know the IP address of the server who wants to get the file from and it needs to know the proper path and file name.
|
|
0:03:55
|
NTP or Network Time Protocol. Time is crucial in Unified Communications, not only for reporting, for trace synchronization, but also for database replication.
|
|
0:04:09
|
And then we’ll take a brief look, not so much at IP (Unicast) Routing, you should know that from CCNA,
|
|
0:04:15
|
but we will take a brief look at Multicast Routing as part of Network Infrastructure because it is very useful,
|
|
0:04:24
|
the idea being one source of information but many recipients, maybe a thousand or even 5000 phones on a network all go on hold at the same time
|
|
0:04:35
|
and we want to play Music on Hold to them, rather than streaming 5000 separate streams,
|
|
0:04:40
|
we can stream one stream of that Music on Hold file from the particular server.
|
|
0:04:47
|
And while it will go out to 5000 recipients, we only have to stream it once from the server.
|
|
0:04:54
|
It is imperative that you are already familiar with IP (Unicast) Routing.
|
|
0:04:58
|
It is needed to route, obviously, data so our signaling and our voice is now a data as well, or a video.
|
|
0:05:06
|
So if you are not familiar, if you haven’t taken CCNA, the CCNA Route and Switch, if you want to call it that,
|
|
0:05:15
|
then you need that before you can go and take, well, I guess you can go and take the CCNA Voice Exam,
|
|
0:05:22
|
but in order to get a certification, you need to have CCNA before you can take and pass and be certified as a CCNA Voice.
|
|
0:05:34
|
So, moving on to the Phone Registration Process, the phone gets its power from the connected switch.
|
|
0:05:42
|
The phone then boots its current firmware. We are going to take a look at this in an actual animation here in a moment.
|
|
0:05:47
|
The phone boots from the firmware that has already loaded in its stored flash memory,
|
|
0:05:54
|
assuming it has a firmware loaded, assuming it hasn’t been factory defaulted and erased which is certainly possible.
|
|
0:06:01
|
Cisco Discovery Protocol, or else, LLDP-MED. So, the Link Layer Discovery Protocol, which is the industry standard of CDP is obviously Cisco proprietary.
|
|
0:06:12
|
IP phones in their current firmware of support both is by-directionally transmitted to effectively instruct, as we have just mentioned, the IP phone,
|
|
0:06:23
|
what Voice VLAN should be used as a VLAN ID whenever the phone is taking in creating and passing on all of the Ethernet frames
|
|
0:06:36
|
and when it is creating the .1q trunk header with the particular VLAN ID equaling the Voice VLAN ID that was instructed from the switch,
|
|
0:06:46
|
so that all of its Ethernet frames that were generated by the phone itself can be destined for the Voice VLAN
|
|
0:06:53
|
and then all other non-tagged frames that those came from the PC or Mac-attached and those will go into the access VLAN.
|
|
0:07:02
|
So the phone sends a DHCP request once the Voice VLAN has obtained broadcast and does get a response from a server
|
|
0:07:12
|
with the IP address, Subnet mask, Default Router and TFTP Server at a minimum that the phone should use, obviously its IP address to communicate on the network
|
|
0:07:23
|
and Subnet Mask for the same, Default Router for the same and then TFTP Server as a crucial bed of information.
|
|
0:07:29
|
We are going to talk a lot more about all of these in the future slides.
|
|
0:07:34
|
The thing we are going to talk about now is just the basic phone registration process and looking at that.
|
|
0:07:39
|
So then the phone will next contact the TFP Server with the request for SEP, and it’s not really _MAC_Address,
|
|
0:07:49
|
but SEP for Selsius, which is the name of the company that Cisco acquired for call managers, Selsius Ethernet Phone and then the actual Mac address .cnf .xml.
|
|
0:08:00
|
So basically its own file, its own name.
|
|
0:08:03
|
And then if for some reason that file wasn’t available, the phone will follow up that request with a request for a file Default.cnf.xml
|
|
0:08:14
|
So basically if its own SEP_MAC_Address file was not there, maybe there was no record in the database and auto registration is turned on,
|
|
0:08:23
|
It will look for the default record and try to gain information, it will, then, the phone will parse the XML configuration file for the load information
|
|
0:08:34
|
which is effectively the firmware version that it should be running, as well as the , the CPE.
|
|
0:08:40
|
If the load information is different than the load file it currently has, back in step 2, then it will contact the TFTP server and try to download the newer firmware,
|
|
0:08:53
|
the newer load file and upgrade itself and then it will finally contact that .
|
|
0:08:58
|
The Code Processing Engine is what CPE stands for.
|
|
0:09:04
|
And then next the phone will register with CUCM or CUCM(E), Cisco Unified Communication Manager or Express with the Call Processing Engine for the
|
|
0:09:17
|
and then finally the server will hand down any specifics to the phone such as Line, Soft Key Template, etc.
|
|
0:09:23
|
The bad information for Line, Softkey Template that it does not come in the configuration file.
|
|
0:09:28
|
The configuration tells it a lot of things such as whether the Web Server can be run on the phone or things like that, but then it will hand down any specifics later.
|
|
0:09:40
|
So let us just take a look at an animated view of this.
|
|
0:09:43
|
We got our PoE switch on the right, either 82.3af which is the first PoE standard
|
|
0:09:52
|
or .at which is the enhanced standard to allow for more power
|
|
0:09:57
|
and a newer standard is actually even right around the ban of being ratified for even more power
|
|
0:10:04
|
but what happens is, to begin with, first of all,
|
|
0:10:07
|
this is PoE. So this is the industry standard of 82.3af.
|
|
0:10:11
|
There is a resistance sent across the line and power classification
|
|
0:10:18
|
and as long as the phone returns the resistance of between 19 and 26.5k Ohms
|
|
0:10:25
|
then the PoE switch know that this is a device that needs power.
|
|
0:10:32
|
The PoE switch is also referred to as a PSE or a Power Source Equipment
|
|
0:10:35
|
and the phone would be also referred to as a PD-empowered Device.
|
|
0:10:41
|
Now this is the PoE Standard, the 82.3af Standard.
|
|
0:10:46
|
If these were the Cisco Inline Power at proprietary pre-standard, then everything else that we’re going to look at would be the same.
|
|
0:10:56
|
Beyond this, however, the first thing instead of a specific resistance detection is,
|
|
0:11:04
|
a what would be sent from the switch is something called a Fast Link Pulse (FLP).
|
|
0:11:11
|
That is really the main difference basically how it is sending out that very low resistance detection
|
|
0:11:23
|
or actually sending out a Fast Link Pulse for PoE and Inline Power respectively.
|
|
0:11:31
|
So the phone will, depending on the resistance that was returned,
|
|
0:11:37
|
the switch knows the classification of the power supplied
|
|
0:11:40
|
and there are things such as Class II would indicate that the phone needs between 3.8 and 6.4 watts,
|
|
0:11:49
|
6.3 as what a lot of the older phones used, the 7960, 7940, that is considered a low-powered device.
|
|
0:11:56
|
Class IV, would be considered a high-powered device
|
|
0:11:59
|
and the PoE switch will know that it can deliver between 12.9 and 25.5 watts
|
|
0:12:07
|
and there’s actually even a newer standard coming out that will allow up to 55 watts.
|
|
0:12:15
|
The new IP phone transmits at Layer 2 CDP does not need an IP address for this.
|
|
0:12:21
|
This is the Layer 2 Protocol. I am Phone X. Here is my CDP device name
|
|
0:12:30
|
and let’s say, for instance, the 9971.
|
|
0:12:33
|
So maybe it instructs that I need 14.4 watts of Power
|
|
0:12:37
|
and CDP or LLDP-MED will inform, it will adjust the power if necessary but then the PoE switch will also inform the phone of the Voice
|
|
0:12:48
|
sometimes referred to as the Auxiliary VLAN.
|
|
0:12:51
|
So then using that new VLAN ID, with its 82.1q tag.
|
|
0:12:56
|
This is assuming that a trunk was negotiated, it does not have to use a trunk.
|
|
0:13:01
|
As we have mentioned, it could just be a switch port mode access or an access ID
|
|
0:13:06
|
where there is no trunk, no .1q
|
|
0:13:09
|
but over that Voice VLAN, if we are using 82.1q trunking,
|
|
0:13:15
|
dynamically negotiated the IP phone will send its DHCP request.
|
|
0:13:21
|
The DHCP server, obviously if it’s not on the same broadcast Layer 2 domain,
|
|
0:13:27
|
we need to have some sort of a DHCP relay or IP helper address,
|
|
0:13:30
|
things that we will take a look at in the future,
|
|
0:13:33
|
but the DHCP response will hopefully come back with the IP address, Subnet Mask, Gateway, TFTP,
|
|
0:13:39
|
possibly Domain Name and possibly even DNS Server.
|
|
0:13:42
|
The last two aren’t absolutely necessary for the phone to register but are provided in many environments.
|
|
0:13:50
|
The phone will then send the TFTP ‘GET’ for the SEP_MAC_Address.cnf.xml
|
|
0:13:58
|
and hopefully the TFTP server is reachable and will send that data back
|
|
0:14:04
|
and then finally the IP phone with either the skinny, or SEP protocol that it is
|
|
0:14:09
|
configured with its firmware loadwidth will register with the Call Processing Engine
|
|
0:14:15
|
and the Call Processing Engine will send, as we mentioned, additional information such as lines, button, softkey templates, all sorts of other stuff.
|
|
0:14:25
|
I just want to go over this brief overview and we’re going to take a look at everything in a lot more detail in the upcoming videos.
|
