|
0:00:13
|
So for our next module, we'll take a look at an overview of the Unified Communications System
|
|
0:00:20
|
So to begin with, for Cisco Unified Communications Solution,
|
|
0:00:26
|
we're gonna take a look at the various components involve,
|
|
0:00:29
|
and these begin with Cisco Unified Communications Manager
|
|
0:00:34
|
This is also sometimes referred to as Call Manager
|
|
0:00:37
|
because that was its old name.
|
|
0:00:40
|
It will also be referred to by the acronym, CUCM or sometimes UCM.
|
|
0:00:47
|
We'll also have the Cisco Unified Communications Manager Express,
|
|
0:00:51
|
also referred to as CME or sometimes CUCME
|
|
0:00:57
|
We have Cisco Unity Connection, sometimes referred to as CUCx or just CUC.
|
|
0:01:06
|
We also have Cisco Unity Express,
|
|
0:01:11
|
also referred to as CUE or sometimes just called CUE.
|
|
0:01:15
|
And then we have the Cisco Unified Presence Server known as CUPS
|
|
0:01:19
|
or sometimes people say the CUPS server which is a bit redundant
|
|
0:01:27
|
So the general roles that these play,
|
|
0:01:29
|
CUCM or the Cisco Unified Communications Manager,
|
|
0:01:32
|
this deals with the Call Processiong for Phones and Trunks
|
|
0:01:36
|
and is typically use in larger deployments.
|
|
0:01:42
|
For CUCME or CME, the Unified Communications Manager Express,
|
|
0:01:47
|
this is Call Processing for Phones and Trunks typically for smaller deployments.
|
|
0:01:51
|
Now there are other various components such as
|
|
0:01:55
|
CUCM Business Edition and UC 500 or UC 520,
|
|
0:02:01
|
these are variations of, sort of middle grounds for these, between CUCM and CME
|
|
0:02:10
|
but the two main ones that we'll take a look at here are CUCM and CME
|
|
0:02:16
|
Then we deal with Voice Messaging, so we have Cisco Unity Connection,
|
|
0:02:21
|
this is Voice Messaging for large deployments
|
|
0:02:24
|
and then we have CUE and this is messaging for smaller deployments.
|
|
0:02:31
|
Now there is another product that Cisco offers called just Unity
|
|
0:02:35
|
we won't really be taking a look at that anyway
|
|
0:02:38
|
because while that is for really really large deployments,
|
|
0:02:42
|
Unity Connection has been growing in capabilities and actually continues to grow
|
|
0:02:46
|
even with releases that are coming out shortly
|
|
0:02:49
|
and also in its functionality and has really gone to the place where it has surpassed
|
|
0:02:51
|
the capabilities and in some ways and certainly will and every way
|
|
0:02:59
|
of the larger Cisco Unity
|
|
0:03:02
|
the older and more traditional Cisco Unity
|
|
0:03:06
|
and in fact Cisco Unity is most likely just going to go way altogether.
|
|
0:03:13
|
It's based on a Microsoft exchange-type mail platform Unity Connection,
|
|
0:03:19
|
the one that we're using today and we're looking at here, CUCx
|
|
0:03:23
|
Unity Connection is the one that's here just stay
|
|
0:03:27
|
and grow into the next largest messaging deployment.
|
|
0:03:34
|
Dealing with CUPS, the Cisco Unified Presence Server,
|
|
0:03:35
|
this is an additional presence-based platform
|
|
0:03:42
|
that allows for additional software plans
|
|
0:03:45
|
So, most of the presence functionality as we'll talk about
|
|
0:03:49
|
as time goes on is dealt with by CUCM
|
|
0:03:48
|
but if we want to add additional components
|
|
0:03:54
|
such as instant messaging or soft clients on the desktop,
|
|
0:03:58
|
this is where the CUPS server comes in.
|
|
0:04:02
|
So looking first at CUCM at a Hi-Level Overview.
|
|
0:04:08
|
Capabilities of CUCM, first of all it's a fully-featured voice and video solution.
|
|
0:04:14
|
We have the ability to cluster most services to provide five 9's resiliency.
|
|
0:04:20
|
This is typically for larger deployments.
|
|
0:04:23
|
We actually theoretically have the ability to have up to 30,000 phones per cluster
|
|
0:04:28
|
Now, this depends on how many of what device type you have and licensing,
|
|
0:04:35
|
not even just licensing but design sizing
|
|
0:04:38
|
and depending on what types of devices you have rarely
|
|
0:04:42
|
will you see a cluster of actually 30,000 actual IP phones,
|
|
0:04:47
|
although theoretically that is what it could have
|
|
0:04:49
|
if it didn't have any other components to it.
|
|
0:04:52
|
But, it is not rare that you'll see, well depending on
|
|
0:04:57
|
I guess you are in house or consulting decision
|
|
0:05:01
|
I should say it's not rare that you'll see clusters at least talked about by Cisco
|
|
0:05:08
|
but real-world deployment clusters of 15 to sometimes even 20,000 phones
|
|
0:05:13
|
but 30,000 is the number and one thing you keep in mind is that,
|
|
0:05:17
|
as it relates to the certifications,
|
|
0:05:22
|
unless there's some sort of update or adendum to this in the future,
|
|
0:05:25
|
we are dealing with CUCM 8.0 or possibly 8.5 for this version of testing
|
|
0:05:34
|
and not 8.6 or 9.0 or beyond,
|
|
0:05:39
|
where 8.6 actually ups the ability for cluster-size to 40,000 phones per cluster
|
|
0:05:48
|
so it's important when you're thinking about the certification that you're taking
|
|
0:05:54
|
that you try to remember or at least get some sort of information such as from us,
|
|
0:06:00
|
as to what version of software was being tested
|
|
0:06:07
|
or being generally available at the time when the test was written
|
|
0:06:13
|
and therefore what is Cisco asking about in terms of,
|
|
0:06:17
|
you know, on the exam that you're taking
|
|
0:06:19
|
because they don't say typically on the exam,
|
|
0:06:22
|
you know, what's the maximum amount of phones per cluster
|
|
0:06:27
|
as it relates to CUCM 8.0 or 8.5 or 8.6.
|
|
0:06:32
|
So it's important for you, the test taker,
|
|
0:06:34
|
to actually do some research or get that information from us
|
|
0:06:38
|
and find out what exam version or what platform
|
|
0:06:43
|
or server version is being tested on any given exam
|
|
0:06:49
|
So this CUCM runs on a hardened Red Hat Enterprise Linux platform as an appliance
|
|
0:06:56
|
there is no direct shell access.
|
|
0:07:00
|
It runs as an individual 1U rackmount Cisco UCS server,
|
|
0:07:06
|
or it can also run on Cisco UCS blade servers running in VMWare ESXi
|
|
0:07:13
|
and it stores all the data in an IBM Informix database.
|
|
0:07:19
|
Some of the hi-level features of CUCM are the fact that
|
|
0:07:25
|
it is a Call Processing Engine and you'll hear me
|
|
0:07:28
|
refer to this in many future modules as CPE, Call Processing Engine.
|
|
0:07:35
|
Configuration can be done via the web user interface.
|
|
0:07:39
|
Really only in terms of general user configuration,
|
|
0:07:43
|
that is to say there's no true command line configuration.
|
|
0:07:47
|
There's a command line that we can use for operating system
|
|
0:07:51
|
and even doing a lot of show commands to see various things about the platform
|
|
0:07:56
|
but in terms of the configuration, this is all web-based, UI-based.
|
|
0:08:02
|
Now, I will say that configuration can be done
|
|
0:08:07
|
through command line in terms of SQL specific commans
|
|
0:08:13
|
but it's, unless you are specifically dealing with a developer or you are a developer
|
|
0:08:19
|
then this is not something that you ever want to get into and manipulate and
|
|
0:08:26
|
or there is also the capability to configure through SOAP and AXL.
|
|
0:08:31
|
SOAP being a programming, sort of language, sort of scripting framework
|
|
0:08:38
|
and structure that can be used by AXL and other XML type sources
|
|
0:08:45
|
in order to transmit and receive information.
|
|
0:08:48
|
So through SOAP and AXL which is,
|
|
0:08:52
|
AXL is Cisco's administrative XML implementation.
|
|
0:08:56
|
The system can also be programmed or provisioned
|
|
0:09:02
|
but again this is mainly for developers.
|
|
0:09:05
|
This also can be a Music on Hold server, a Conferencing server,
|
|
0:09:11
|
a Automated Recording Annunciator server
|
|
0:09:16
|
has the ability to provide directory services.
|
|
0:09:18
|
Either integrated, within itself so self-contained
|
|
0:09:23
|
and/or it can do LDAP synchronization
|
|
0:09:27
|
and it also has Presence Services.
|
|
0:09:30
|
It also has an integrated Disaster Recovery System for Backup and Restore.
|
|
0:09:35
|
and there is a Unified Serviceability component to it
|
|
0:09:41
|
to serve as various features and functions of the server.
|
|
0:09:46
|
and also Cisco RTMT which is the Real Time Monitoring Tool
|
|
0:09:52
|
to provide the ability for management and troubleshooting.
|
|
0:09:57
|
Looking at Per Hardware Server Specs, if you're running on Cisco UCS servers,
|
|
0:10:06
|
we can run on a Cisco UCS C210 Rack-mount Server
|
|
0:10:13
|
and this can give us approximately 7,500 endpoints/server.
|
|
0:10:19
|
Obviously we have to cluster mulitple servers together
|
|
0:10:22
|
to ever get to the 30,000 endpoint per cluster maximum.
|
|
0:10:29
|
We can use the Cisco UCS B200 Blade Server in order,
|
|
0:10:35
|
so does the C and then the number is typically a standalone Rack-mount server
|
|
0:10:40
|
a B and then a number is typically a Blade server
|
|
0:10:42
|
if you're unfamiliar with the UCS, Cisco UCS or Unified Computing System platform.
|
|
0:10:49
|
So the B200 can handle 7,500 endpoints/server.
|
|
0:10:52
|
The C200 Rack-mount can handle 1000 endpoints/server.
|
|
0:10:57
|
Depending on the version of CUCM you're trying to install,
|
|
0:11:02
|
it can install on older MCS servers as well.
|
|
0:11:05
|
These were really, the Cisco UCS servers
|
|
0:11:09
|
are servers that Cisco of course designed themselves.
|
|
0:11:13
|
Cisco MCS servers are really re-branded HP/Compaq servers
|
|
0:11:18
|
But, the Cisco MCS 7845-I3 can handle 10,000 endpoints/server.
|
|
0:11:28
|
Just the 7845 can handle 7,500 endpoints/server.
|
|
0:11:33
|
The Cisco MC 7835 can handle 2500 endpoints/server.
|
|
0:11:39
|
The Cisco MCS 7825 can handle a thousand per server.
|
|
0:11:45
|
The 7816 can handle 500 and the 7815 can handle 300 per server.
|
|
0:11:52
|
On any server, you have to have a minimum of 2GHz of CPU,
|
|
0:11:59
|
2GB of RAM and 72GB HDD.
|
|
0:12:04
|
Obviously, more is better for a really two cores each at 2GHz
|
|
0:12:12
|
and 4GB of RAM is recommended but those are the minimums.
|
|
0:12:20
|
Looking at UCS Server Specifics, CUCM 8 is the first version
|
|
0:12:25
|
that is actually supported in VMWare.
|
|
0:12:28
|
UCS Hardware must be used and the installation DVD's will actually force you,
|
|
0:12:36
|
It will do a hardware check and not allow you to install it.
|
|
0:12:40
|
Actually, that is not true, it will install on VMWare not in a UCS Blade
|
|
0:12:45
|
but it will be very specifically in red, in bold,
|
|
0:12:51
|
anytime you log in into the web user interface or the command line interface,
|
|
0:12:56
|
it will be very poignantly told to you that
|
|
0:13:01
|
this can only be used as a lab or demo system
|
|
0:13:04
|
Because, if it's not running on ESXi specifically, not just the SX
|
|
0:13:11
|
and with a SAN fiber channel hardware
|
|
0:13:14
|
and it's not on a Cisco UCS hardware Blade server,
|
|
0:13:22
|
Then it's not going to be supported for a production environment.
|
|
0:13:26
|
But for the exam or looking at the production environment
|
|
0:13:29
|
as what we mainly care about.
|
|
0:13:33
|
So, again, VMWare ESXi 4.0 with SAN
|
|
0:13:38
|
and in order to install this you must use the .ova file to install
|
|
0:13:43
|
So the .ova file is the open virtualization architecture
|
|
0:13:48
|
basically a text-based file that tells the VMWare ESXi server how to provision
|
|
0:13:56
|
that VM guest on the system before you do the installation.
|
|
0:14:03
|
You can perform a migration from a hardware based,
|
|
0:14:07
|
hardware based standalone rack-mount server or MCS server
|
|
0:14:12
|
using the Disaster Recovery Tool in order to migrate over to a virtualize environment
|
|
0:14:20
|
and you do have the ability to use mixed hardware in a cluster
|
|
0:14:24
|
so that is to say and especially useful when we are doing a migration
|
|
0:14:28
|
that is to say I could have some of my servers be in VMWare ESXi
|
|
0:14:33
|
and other servers, the actual hardware rack-mount server.
|
|
0:14:39
|
and the license key for CUCM uses VMWare virtual MAC address
|
|
0:14:44
|
if you're using VMWare.
|
|
0:14:47
|
So VMWare Specs, if you are using VMWare,
|
|
0:14:52
|
you do have to, again this is the requirement for the exam
|
|
0:14:56
|
this is the requirement for a production network
|
|
0:14:59
|
and a test network, a demo network,
|
|
0:15:00
|
you can get away with the minimum that we talk about earlier.
|
|
0:15:05
|
But, for VMWare, you need two virtual CPUs,
|
|
0:15:09
|
you need 6GB RAM and two 80GB hard disks
|
|
0:15:14
|
specifically at SCSI 0:0 and 0:1 respectively.
|
|
0:15:24
|
Only a UCS version 7, this is not CUCM version 7
|
|
0:15:28
|
but a UCS version 7 of VM is supported or Virtual Machine is supported
|
|
0:15:36
|
And this supports up to 7500 users per VM.
|
|
0:15:40
|
There is no USB support so this is important for Secure keys
|
|
0:15:46
|
or Music on Hold live feeds, in VMWare it is not currently supported.
|
|
0:15:51
|
All backups for the Virtual Machine must be done via SFTP or Secure FTP.
|
|
0:16:00
|
Taking a look at Server Roles in Unified Communications Manager,
|
|
0:16:05
|
the CUCM database server has two different,
|
|
0:16:08
|
and really this is SQL databases in general,
|
|
0:16:12
|
have a primary database, the publisher database
|
|
0:16:16
|
and then they have many subscriber databases
|
|
0:16:20
|
and these are synchronized from publisher databases,
|
|
0:16:21
|
this is really SQL in general.
|
|
0:16:23
|
But we do refer in a lot of the modules in teaching
|
|
0:16:28
|
that you'll end up going through about the CUCM pub
|
|
0:16:35
|
meaning the database publisher server
|
|
0:16:37
|
or the first server in the network, the first node in the cluster
|
|
0:16:41
|
and then various CUCM subserver or database subscriber servers.
|
|
0:16:49
|
So, we typically run a publisher server
|
|
0:16:54
|
strictly for database changes or synchronizations.
|
|
0:16:57
|
Now, when I'm saying some of this stuff, this is for larger deployments.
|
|
0:17:02
|
If you have a smaller deployment, if you have a very small deployment,
|
|
0:17:06
|
maybe 150 phones or less but planning on growing,
|
|
0:17:13
|
otherwise you might just want to install CME
|
|
0:17:15
|
but if you have 200 or maybe even 500 phones, something like that,
|
|
0:17:22
|
It's possible that you might have only two servers in your cluster.
|
|
0:17:25
|
You never wanna have less than two.
|
|
0:17:28
|
And I'm more talking about real world now than I am talking about the actual exam.
|
|
0:17:34
|
But it's possible that you might just have a database publisher server
|
|
0:17:40
|
and one subscriber server
|
|
0:17:41
|
and that you're running a lot of the services such as call publishing such as call processing engine
|
|
0:17:46
|
and other things on the publisher server as well as the subscriber
|
|
0:17:50
|
But for decent size deployments, you know, 500 phones and up or whatever,
|
|
0:17:56
|
but really in general and specifically for the lab or test exam,
|
|
0:18:03
|
that you'll be taking regarding this,
|
|
0:18:06
|
we want to keep in mind that the publisher server,
|
|
0:18:08
|
its general role is not to participate in call processing, digit analysis
|
|
0:18:15
|
and call processing engine functions and various functions like this.
|
|
0:18:19
|
Its primary role is as a database publisher, dealing with keeping all the subscribers in sync.
|
|
0:18:26
|
It also, there are certain things that only run on the CUCM pub server such as MVA which is Mobile Voice Access,
|
|
0:18:38
|
it is something that we get into and deal with when we talk about Unified Mobility or Mobile Connect,
|
|
0:18:44
|
which is essentially the single number reach feature, Mobile Voice Access is an extension of that.
|
|
0:18:51
|
Things such as Directory Synchronization with LDAP but again it can be used as a CPE.
|
|
0:18:58
|
We can have up to 8 subscriber servers running as CPE's, Call Processing Engines.
|
|
0:19:05
|
We can have up to three Call Processing Engines per Unified CM Group,
|
|
0:19:12
|
so one of the things that we do in CUCM when we're provisioning it and configuring it,
|
|
0:19:17
|
is we define Groups and these Unified CM Groups are groups of CPE's or Call Processing Engines.
|
|
0:19:26
|
So these are servers with which the Call Processing Engine service has been activated.
|
|
0:19:31
|
Not all servers will have the Call Processing Engine service activated.
|
|
0:19:36
|
Such as the next two, we can have two subscriber servers running as TFTP servers
|
|
0:19:41
|
and then we can have additional servers used as Media Servers.
|
|
0:19:44
|
For instance, we might have two Music on Hold servers.
|
|
0:19:48
|
We have two TFTP and two MoH, Music on Hold servers,
|
|
0:19:53
|
each to provide redundancy for another.
|
|
0:19:56
|
That way, if one should fail, we still have a backup and phones aren't left wondering what there configuration is, TFTP.
|
|
0:20:05
|
Or wondering what music file should be being played, MoH.
|
|
0:20:19
|
An Annunciator and possibly Conference and MTP that we'll talk about why we typically don't want to do that on a server.
|
|
0:20:21
|
Just the Conference and MTP is typically what we don't want to do.
|
|
0:20:24
|
MoH and Annunciator we will almost always, in fact, those will be pulled from the CUCM servers.
|
|
0:20:35
|
But, servers such as MoH and Annunciators, the Media servers, the TFTP servers and the Publisher,
|
|
0:20:42
|
those, let's see, two MoH, two TFTP and the Publisher, so those five servers
|
|
0:20:47
|
will almost, guaranteed, not be in a larger cluster running any CPE or Call Processing Engine service.
|
|
0:20:57
|
And this is actually known as the Call Manager service by the way, that's the Call Processing Engine service.
|
|
0:21:07
|
But again, we are going to cut these up to the eight Subscriber servers that we can have running Call Processing Engine or Call Manager Service,
|
|
0:21:17
|
we're going to cut this up or divide this up into groups.
|
|
0:21:20
|
And the reason is, is that we're going to assign these groups ultimately to phones through an intermedia, intermediate mechanism known as the device pool.
|
|
0:21:33
|
But we're going to ultimately assign this to phones and gateways and other entities and these will define which Call Processing Engine servers
|
|
0:21:43
|
a phone should register to as a primary, and then if the primary is down, as a secondary, and then if the secondary is down, as tertiary or third server.
|
|
0:21:54
|
And, there's also a fourth option which would be something that we'll talk about later called SRST,
|
|
0:22:01
|
which is know as Survivable Remote Site Telephony and this is a local router running something called SRST
|
|
0:22:09
|
and a few different formats that we'll talk about
|
|
0:22:10
|
server, but there are three primary or three CPE servers, primary, secondary and tertiary,
|
|
0:22:18
|
that can be presented to anyone for Registration, Call Processing, and that's where we can have up to these three CPE's per Unified CM group.
|
|
0:22:29
|
So if we can have eight, then in order to have these eight servers, we really have to have three separate groups
|
|
0:22:35
|
and with this where we'll cross between already use servers
|
|
0:22:40
|
Looking at CUCM database roles, the IBM uses SQL or Structure Query Language
|
|
0:22:47
|
It is owned and managed by replications that are performed, that is from the Publisher to the Subscribers.
|
|
0:22:54
|
Most configurations are made on the Publisher database server, however there are some database fields that can be changed on the Subscriber servers, local to where phones are registered,
|
|
0:23:06
|
and these are called User Facing Features or UFF's
|
|
0:23:10
|
and then those UFF's are then replicated up to the Publisher server if its actually alive and running.
|
|
0:23:18
|
So some examples of User Facing Features a Call Forward All, possibly Message waiting indicator,
|
|
0:23:26
|
whether it's on or off, whether privacy was enabled or disabled by the user,
|
|
0:23:32
|
whether Extension Mobility, which we'll talk about later, has been logged in or logged out of,
|
|
0:23:39
|
whether a Hunt Group has been logged in or logged out of the phone,
|
|
0:23:43
|
Device Mobility Status, something called Certificate Authority Proxy Function dealing with certificates and security,
|
|
0:23:51
|
and their status for end users and application users and Credential Authentication whether has a phone or even a user has been authenticated in terms of their credentials.
|
|
0:24:04
|
So those are User Facing Features that can be changed or updated on a Subscriber server and then replicated back to the publisher.
|
|
0:24:15
|
Now, if the Publisher is down, that is to say for some reason the server went down, crashed, taken down for upgrade, maintenance or anything,
|
|
0:24:28
|
then, users can still, of course they can still do everything that they could normally do in terms of all of their normal functionality
|
|
0:24:39
|
because the subscribers possess a complete copy of the publisher database.
|
|
0:24:41
|
But they're also have the ability to change standard User Facing Features.
|
|
0:24:45
|
So in other words, if there call is forwarded, when the publisher went down, they would have the ability to unforward all of their calls.
|
|
0:24:55
|
Now while an administrator could not change any of the General Administrative features or configuration of CUCM while the Publisher is down,
|
|
0:25:04
|
the user still, obviously as I mentioned, have the ability to change those features, the User Facing Features specifically,
|
|
0:25:12
|
and then when, first of all those will be synchronized between all of the subscribers even while the publisher is down,
|
|
0:25:20
|
and then once the publisher comes back on line, the subscriber who was responsible, the CPE who was responsible for making those changes,
|
|
0:25:28
|
will in fact update or synchronize with the publisher and update its tables to reflect those new values.
|
|
0:25:42
|
Inside of a CUCM cluster of servers, there is cluster communication and this is dealt with the ICCS Protocol, the Intra Cluster Communication Signaling Protocol.
|
|
0:25:54
|
This runs on TCP ports 8002-8004.
|
|
0:25:59
|
Anytime we have TCP ports that begin with 8000 something, this is typically a Java Tomcat Server type port
|
|
0:26:08
|
So, Cisco does rely fairly heavily on Tomcat and the ICCS is sent on Tomcat port 8002-8004.
|
|
0:26:21
|
This replicates all the run time data between all servers and a cluster that are CPE's.
|
|
0:26:27
|
Again that are running the Cisco Call Managing Service or processing calls.
|
|
0:26:32
|
Then looking at Call Detail Records and Call Management Records, sound fairly similar but they're slightly different.
|
|
0:26:41
|
This are all logged by the actual CPE taking the call, so where the phone or where the gateway that the call comes in from is registered to,
|
|
0:26:52
|
and then data is periodically pushed up to the Publisher server.
|
|
0:26:56
|
Cisco CDR Analysis and Reporting also known as the CAR tool, or any third party billing applications servers,
|
|
0:27:05
|
always point to and collect data from the publisher.
|
|
0:27:11
|
And actually they typically point to even a third party different server that really aggregates all that information
|
|
0:27:21
|
but in terms of what we need to look at, they point to the publisher.
|
|
0:27:28
|
So taking a look at a General Overview of Traffic Flow as it relates and pertains to CUCM.
|
|
0:27:36
|
So, you can see at the top here, that we've got our CUCM cluster and we've also got individual IP phones over to the left.
|
|
0:27:48
|
And we've got a Voice Gateway, this router with the V symbol typically specifies a Voice Gateway of some sort.
|
|
0:27:58
|
So in this case a Voice over IP gateway that takes Voice over IP from one side and converts it to or from TDM signaling out to the PSTN or Public Switch Telephone Network.
|
|
0:28:10
|
So you can see my little legend down here, we've got a great TDM signaling here, between the Voice Gateway and the PSTN.
|
|
0:28:19
|
The orange symbol or orange dashed, arrowed line is referring to either SIP or Skinny signaling
|
|
0:28:29
|
so that is to say that to and from CUCM and a Voice Gateway, this could be SIP or Skinny
|
|
0:28:39
|
and the truth is it could also be MGCP signaling as well as a VOIP signaling protocol.
|
|
0:28:46
|
Now, not too often would it be Skinny signaling. I'm sorry.
|
|
0:28:52
|
It also could be H323 as well, so this could really be anyone of the four.
|
|
0:29:05
|
It could be H323 or SIP or Skinny in the case of analog ports only or it could be MGCP.
|
|
0:29:13
|
Obviously, any of the phones can run the SIP or Skinny Protocol in the case of a lot of the newer generation of phones
|
|
0:29:16
|
such as the 9971, the 8941, 8945, 8961, those are SIP-only protocols.
|
|
0:29:29
|
However a lot of the older phones can be SIP or Skinny.
|
|
0:29:34
|
And then RTP media actually flows between the endpoints.
|
|
0:29:40
|
So, in a typical call, this first phone that we see here might have Skinny signaling over the CUCM cluster to indicate that wishes to call the second phone.
|
|
0:29:52
|
The CUCM cluster will send, maybe the second phone is SIP-based so it will send SIP signaling down to the second phone
|
|
0:29:58
|
to indicate that it should begin alerting or ringing.
|
|
0:30:03
|
Once all the communication has passed between these phones and the CUCM cluster that needs to be
|
|
0:30:08
|
in order to setup the call and negotiate all of the particulars, the Codec, the DTMF relay, things of that nature.
|
|
0:30:21
|
Then, the RTP, the actual media will ensue and that will be directly between the two phones.
|
|
0:30:29
|
The only reason that would not be between the two phones is if something third-party was in vote
|
|
0:30:34
|
such as maybe a conferencing server, maybe they were dialing into a conference
|
|
0:30:39
|
then the media would not be between the two phones but the RTP would be to a conference, DSP farm or conference point.
|
|
0:30:49
|
Or possibly one of the phones possible supports one protocol and the other phone only supports another protocol.
|
|
0:30:58
|
And therefore maybe a transcoder would be in vote in order to allow communication between those two.
|
|
0:31:03
|
in which case the RTP media would not go between them.
|
|
0:31:07
|
But if both phones pick the same Codec and there's no conferencing or MTP (Media Termination Point) involved,
|
|
0:31:16
|
then the RTP stream will stay directly between these two phones.
|
|
0:31:22
|
If these two phones happens to be on the same switch, physical switch,
|
|
0:31:28
|
then all the RTP media traffic which is the, most of all the traffic will be staying there local on this, on that switch.
|
|
0:31:40
|
And just be switched back and forth rather than routed.
|
|
0:31:43
|
Remember the signaling is very low over head typically under 1 kilobit per second,
|
|
0:31:50
|
per call anywhere from 150 bits per second to maybe 1 kilobit per second.
|
|
0:31:56
|
But the RTP media is where most of the traffic is.
|
|
0:32:02
|
Same goes for if a call came in from the PSTN and T1 or E1 CAS or Comment Channel, PRI signaling or even FXO, FXS, Analog Signaling
|
|
0:32:18
|
if a call came in from the PSTN, that TDM link would be between the PSTN and the Voice Gateway
|
|
0:32:26
|
and the Voice Gateway would be what uses its PVDM's, it's Packet Voice/Data Modules or DSP's
|
|
0:32:33
|
to terminate that call and reoriginate it in a VOIP Codec
|
|
0:32:39
|
it would do all its signaling to CUCM, CUCM would signal the phone to was be rung
|
|
0:32:46
|
and then once the media and everything was, all the call specifics were negotiated,
|
|
0:32:50
|
then the RTP would be directly be between the phone and the Voice gateway.
|
|
0:33:01
|
So taking a look at Unified Communication Manager Express and a Hi-Level overview of CME.
|
|
0:33:09
|
So capabilities, it's a fully featured Voice and Video solution running inside of a Cisco Router, so inside IOS
|
|
0:33:18
|
typically it's for smaller deployments, it can be up to 350 phones.
|
|
0:33:23
|
It also can be used to provide high availability 'fallback' services for large and small branch offices normally using CUCM, also known as SRST.
|
|
0:33:35
|
And there are two types of SRST, there is Unified SRST which I sometimes refer to as traditional SRST.
|
|
0:33:42
|
And then there is CME as SRST.
|
|
0:33:45
|
Since we're talking about CME or Communication Manager Express,
|
|
0:33:50
|
this is where I say that it can be used as a high availability 'fallback' service.
|
|
0:33:56
|
and by the way, both Unified SRST which is the Call-Manager-Fallback command,
|
|
0:34:03
|
or CME which is under the Telephony-Services command.
|
|
0:34:09
|
Both are really the same engine.
|
|
0:34:15
|
It does have the ability to provide for third party CTI or Computer Telephony Integration support.
|
|
0:34:22
|
CME can deal with Single-site or Multi-site Deployments.
|
|
0:34:28
|
It can provide for Directory and Presence Services Integrated
|
|
0:34:33
|
It has the ability to interwork with obviously other CME sites in the Multi-site Deployment as well as interwork with CUCM.
|
|
0:34:42
|
And it does provide support for remote phones.
|
|
0:34:47
|
Some of the Hi-Level features, it also is a Call Processing Engine.
|
|
0:34:52
|
It can deal with device control via VOIP, SIP or Skinny signaling.
|
|
0:34:59
|
In CME, configuration can be done via a command line, CLI (Command Line Interface)
|
|
0:35:05
|
or Web User Interface
|
|
0:35:08
|
It should be noted that in CME, that only a small subsection of the configuration can be done using the Web User Interface.
|
|
0:35:21
|
Whereas using the Command Line, which is not only what we will, through most of the learning modules that you go through from INE.
|
|
0:35:29
|
Not only what we'll use most of the time but it has the ability to unlock far, far, far, far ,far more
|
|
0:35:38
|
not only CME specific features but also all of the rest of the features that IOS has to offer that aren't specific to CME,
|
|
0:35:47
|
but become very critical and crucial to CME proper operation.
|
|
0:35:54
|
As an Integrated Directory, it has integrated or ability to provide Music on Hold files, so it's a Music on Hold server in essence.
|
|
0:36:06
|
It can be a Conferencing Server, both software and hardware based.
|
|
0:36:11
|
So software meaning it's, the conferences are muxed or multiplexed together in IOS software versus hardware based
|
|
0:36:18
|
where we actually involve and invoke PVDM's in order to have conferences with different Codecs, different rates.
|
|
0:36:27
|
And it has the ability to run really fully featured rich media applications
|
|
0:36:35
|
using a TCL or Tool Command Language, scripting language.
|
|
0:36:40
|
And some of the ones that were included by default are B-ACD or Basic Automatic Call Distribution.
|
|
0:36:47
|
pretty much anything you can write in TCL as well as VXML or Voice XML
|
|
0:36:53
|
can be run in a Cisco Router.
|
|
0:36:58
|
So General Traffic Flow, we see the signaling again is typically always between the endpoints and router itself.
|
|
0:37:09
|
Or also, actually not pictured here between the router, the CME, and the PSTN
|
|
0:37:14
|
that would be a TDM signaling but the VOIP signaling is gonna be between the endpoints
|
|
0:37:19
|
The CME is the Voice Gateway, so it doesn't really signal between the Call Processing Engine and the Voice Gateway since it is itself.
|
|
0:37:28
|
And then RTP stream go between the endpoints.
|
|
0:37:31
|
They also would go between the phone and the gateway in a sense of a call going out to the PSTN
|
|
0:37:38
|
or in a sense of a conference which uses either software or hardware or transcode or something of that nature.
|
|
0:37:46
|
or also, if the router was providing Music on Hold or some sort of custom rich media application like B-ACD,
|
|
0:37:56
|
where the router is actually providing and serving voice files,
|
|
0:38:01
|
or audio files then there would be RTP media between the router as well.
|
|
0:38:08
|
Dealing with specs, for CME in sizing, routers that support CME will look first at the ISR-G2's
|
|
0:38:17
|
and then, which are the Generation 2 of the Integrated Services router and then we'll look at what the older Integrated Services router, the Generation 1 just plain called the ISR's
|
|
0:38:29
|
what they can support, so with the 2901 ISR-G2, this can support 35 phones
|
|
0:38:39
|
It has zero Service Module Slots and it can support up to 8 T1 or E1's.
|
|
0:38:48
|
2911 can support up to 50 phones, it has 1 Service Module, the Integrated Service Module Slots used to be called the Network Module Slot.
|
|
0:39:01
|
So in the ISR's, they're know as NM's or Network Modules.
|
|
0:39:04
|
In the ISR-G2's they're know as SM's or Service Modules.
|
|
0:39:09
|
And it can support up to 12 T1 or E1's.
|
|
0:39:13
|
The 2921 router can support
|
|
0:39:19
|
and by the way, ISR-G2's always have a 9 as the second digit.
|
|
0:39:24
|
The older ISR's always have an 8 as the second digit so the 2801 and the 2901 are generation 1 and 2 respectively.
|
|
0:39:36
|
But anyhow, 2921 can have up to 100 phones, 1 Service Module and 12 T1 or E1's
|
|
0:39:43
|
2951, up to 150 phones, 2 Service Module Slots, up to 16 T1 or E1's
|
|
0:39:50
|
The 3925 ISR-G2 can have up to 250 phones, 2 SM Slots and 16 T1/E1's
|
|
0:40:02
|
The 3945 can have up to 350 phones, 4 SM Slots and 24 T1/E1's
|
|
0:40:10
|
By the way, these are not the numbers that represent SRST, CME as a source T failover capability
|
|
0:40:18
|
These are just the numbers that represent CME as a plain default.
|
|
0:40:24
|
So looking at the older ISR, there was an 1861 that can support 15 phones with no Network Module Slots
|
|
0:40:33
|
The 2801 can support 25 phones with no Network Module Slots
|
|
0:40:38
|
2811, 35 phones and 1 Slot
|
|
0:40:42
|
2821, 50 phones and 1 slot
|
|
0:40:45
|
2851, 100 phones and 2 slots
|
|
0:40:49
|
The 3825, 175 phones and 2 slots
|
|
0:40:54
|
The 3845, 250 phones and 4 slots
|
|
0:40:59
|
And then we have the RISR, those are the Rugged ISR's
|
|
0:41:04
|
These are for Police and Law Enforcement or Military-use
|
|
0:41:08
|
or really just any harsh environment that you might need to have an ISR
|
|
0:41:12
|
maybe for some sort of disaster, True Disaster Recovery site, for enterprises
|
|
0:41:19
|
the Rugged ISR 3250 and 3270 can have 20 and 48 phones respectively but don't have any Network Module Slots.
|
|
0:41:33
|
Taking a look at Cisco Unity Connection at a Hi-Level Overview
|
|
0:41:38
|
Unity Connection is an integrated messaging server, voicemail with email retrieval in other words.
|
|
0:41:46
|
This is typically for larger deployments, we can have actually up to 20,000 users or voicemail boxes per server
|
|
0:41:54
|
with the ability to cluster or failover in an Active/Active configuration so we can have two servers in a cluster
|
|
0:42:02
|
both running Active/Active so we can have 20,000 on one, 20,000 in the other with the ability to have a level of failover between the two
|
|
0:42:13
|
It features a fully featured Auto Attendant or you could also refer to it as a Voice Menu server
|
|
0:42:21
|
It does provide Directory Services with Text to Speech and with the ability to synchronize from LDAP
|
|
0:42:29
|
It supports Microsoft Exchange integration for Calendaring and IMAP integration for Integrated or Unified Messaging
|
|
0:42:39
|
It does support remote SRSV support, so Survivable Remote Site Voicemail
|
|
0:42:48
|
And it does have the ability to perform Digital Networking with other Unity Connection servers for large deployments.
|
|
0:42:56
|
So in the same way that I could have multiple clusters of CUCM for phones and regular Call Processing interconnected with each other,
|
|
0:43:08
|
through various things that we'll talk about later, refer to as Trunks
|
|
0:43:12
|
I can also link Unity Connection with other Unity Connection systems in order to provide fully integrated or clustered or remotely accessible voicemail boxes between clusters.
|
|
0:43:30
|
This is known as Digital Networking
|
|
0:43:33
|
Then there's something called the VPIM or Voice Profile for Internet Mail, this is an industry standard Digital Networking is proprietary to Unity Connection and Unity.
|
|
0:43:46
|
But, VPIM is an industry standard and I can run VPIM on Unity Connection to network with other voicemail servers.
|
|
0:43:56
|
Some of these servers being Unity Express
|
|
0:43:59
|
Unity Express and Unity Connection don't use Digital Networking as Unity Connection to Unity Connection would
|
|
0:44:06
|
but Unity Connection to Unity Express use VPIM.
|
|
0:44:09
|
This give the ability to lookup users in other clusters or other locations as they're called
|
|
0:44:19
|
and send a message across to another server even forward messages that are in my voicemail box to another user on another server.
|
|
0:44:29
|
In Unity Connection, messages are stored in a local file system
|
|
0:44:34
|
Whereas with the larger Unity, temporarily larger Unity, messages were actually stored in Microsoft Exchange.
|
|
0:44:46
|
So with clustering for Unity Connection, as I mentioned,
|
|
0:44:49
|
the two server configuration, Active/Active in terms of Failover
|
|
0:44:53
|
so it's not an Active Standby.
|
|
0:44:56
|
So both servers process the Real Time Protocol
|
|
0:45:01
|
HTTP request and IMAP request
|
|
0:45:04
|
and it really uses the same Informix database structure as CUCM
|
|
0:45:09
|
in fact it runs the same operating system structure as CUCM.
|
|
0:45:14
|
In fact, all of the base underlying system is the same.
|
|
0:45:19
|
In fact that's the reason that you
|
|
0:45:21
|
if you're looking at the DVD and doing an installation,
|
|
0:45:25
|
there is the ability to install Unified Communication Manager,
|
|
0:45:30
|
Unified Communication Manager Business Edition
|
|
0:45:33
|
or Unity Connection all from the same disc.
|
|
0:45:36
|
And, very soon you're going to be able to
|
|
0:45:39
|
install CUPS from the same disc as well
|
|
0:45:42
|
because that would pretty much be merged with CUCM
|
|
0:45:44
|
but that's not quite, not quite yet
|
|
0:45:49
|
So, the Publisher server is a master
|
|
0:45:52
|
and it owns both the database and the Message Store.
|
|
0:45:57
|
And, the Subscriber server subscribes to both of those.
|
|
0:46:01
|
Not only the database for configuration
|
|
0:46:03
|
but also the Message Store.
|
|
0:46:06
|
Typical traphic flow in Unity Connection
|
|
0:46:09
|
notice I've kind of separated Skinny as dark blue
|
|
0:46:17
|
and SIP as light blue
|
|
0:46:18
|
for each respective protocol,
|
|
0:46:22
|
I can integrate Unity Connection with either Skinny or SIP
|
|
0:46:27
|
into a Unified Communication Manager cluster
|
|
0:46:31
|
of course the signaling between the voice gateways
|
|
0:46:35
|
and the phones still stay as SIP or Skinny.
|
|
0:46:39
|
Voice Gateway could be SIP, Skinny, H323 or MGCP
|
|
0:46:44
|
but the RTP stream goes directly with between Unity Connection
|
|
0:46:48
|
and the phone or the gateway.
|
|
0:46:52
|
Looking at Cisco Unity Express as a Hi-Level Overview
|
|
0:46:59
|
So some of the capabilities of Unity Express,
|
|
0:47:04
|
obviously, Voice Messaging, this is built on a Linux OS
|
|
0:47:10
|
no it's not the same structure as CUCM and Unity Connection is.
|
|
0:47:16
|
but it is a Linux Operating System underneath
|
|
0:47:18
|
with the pseudo IOS-like command line
|
|
0:47:22
|
so, Unity Express does give you the ability
|
|
0:47:24
|
to configure everything via Command Line
|
|
0:47:29
|
or everything via Web UI (Web-based User Interface).
|
|
0:47:34
|
So unlike CUCM, Communication Manager or Unity Connection,
|
|
0:47:42
|
where we can only do things with the web User Interface
|
|
0:47:46
|
the Command Line is useful for operating system stuff
|
|
0:47:48
|
but not for primary configuration of the product.
|
|
0:47:52
|
And unlike CME, Communication Manager Express
|
|
0:47:57
|
where we can configure everything via Command Line
|
|
0:48:01
|
and a subset of that via Web User Interface.
|
|
0:48:06
|
With Unity Express, we can configure everything via Web User Interface
|
|
0:48:11
|
or everything via Command Line
|
|
0:48:14
|
and in fact, if Unity Express is integrated with Communication Manager Express (CME),
|
|
0:48:23
|
then we can not only configure everything in Unity Express
|
|
0:48:27
|
but we can also configure the same subportion of CME
|
|
0:48:32
|
that we were able to do in the CME Web User Interface
|
|
0:48:35
|
where now we're able to configure the same subportion in Unity Express Web Interface as well
|
|
0:48:41
|
now that's only a few to the Express is integrated with CME
|
|
0:48:45
|
It could primarily be integrated with CUCM.
|
|
0:48:50
|
Likewise, Unity Connection can primarily be integrated with CUCM
|
|
0:48:55
|
or it could be primarily integrated with CME.
|
|
0:48:58
|
But for either one of those messaging systems,
|
|
0:49:00
|
they can only be integrated with one of the Call Processing engine type systems
|
|
0:49:08
|
as a primary method.
|
|
0:49:12
|
Unity Express again is typically for smaller deployments.
|
|
0:49:15
|
We can have up to 300 mailboxes.
|
|
0:49:18
|
We do have the ability to support fax messages.
|
|
0:49:22
|
just like Unity Connection.
|
|
0:49:24
|
It also can be used to provide high availability failover or fallback services
|
|
0:49:30
|
for medium and small branch offices
|
|
0:49:32
|
that would normally use Unity Connection Voicemail
|
|
0:49:36
|
and this is known as SRSV.
|
|
0:49:40
|
So, CME can be used as high availability for Call Processing
|
|
0:49:44
|
known as Survival by Remote Site Telephony (SRST).
|
|
0:49:49
|
CUE (Unity Express) can be used as a high availability mechanism
|
|
0:49:54
|
for Voice Messaging known as SRSV or Survival by Remote Site Voicemail.
|
|
0:50:01
|
It also has a fully featured 'Auto Attendant' or 'Voice Menu'.
|
|
0:50:07
|
Also has actually an 'Interactive Voice Response' server
|
|
0:50:08
|
with the ability to queue callers.
|
|
0:50:10
|
Now, not in the lower level CCNA and CCNP exams,
|
|
0:50:17
|
not in there is the UCCX server
|
|
0:50:20
|
or the Unified Contact Center Express server.
|
|
0:50:30
|
Basically a call center server
|
|
0:50:30
|
but the Express version.
|
|
0:50:33
|
So not in there, in those lower levels certifcations
|
|
0:50:36
|
are, is that UCCX contact server.
|
|
0:50:39
|
It is on the CCIE Voice.
|
|
0:50:42
|
However, Unity Express has the ability to do a number of things
|
|
0:50:50
|
that are similar in fashion
|
|
0:50:52
|
and certainly a subset of the abilities of UCCX.
|
|
0:50:57
|
So, and because it's actually built on just about the same Java classes
|
|
0:51:02
|
and it's got almost identical editor as the Unified Communication
|
|
0:51:10
|
I'm sorry, Unified Contact Center Express server.
|
|
0:51:15
|
The visual drag and drop editor is almost identical
|
|
0:51:19
|
except for two small things which is like schedules
|
|
0:51:24
|
schedules is actually all I can think of.
|
|
0:51:26
|
It's not a bad idea to,
|
|
0:51:31
|
specially if you're planning on going on
|
|
0:51:32
|
for the CCIE Voice certification to begin looking at
|
|
0:51:37
|
the Contact Center Express server in scripting
|
|
0:51:41
|
because it really will teach you all you need
|
|
0:51:43
|
to know about Unity Express scripting
|
|
0:51:46
|
as well for that IVR (Interactive Voice Response) and for queuing callers.
|
|
0:51:55
|
So looking at Unity Express traffic flow,
|
|
0:51:59
|
again, we have our signaling between the phones
|
|
0:52:04
|
and the CME or phones in CUCM and then the signaling between CME
|
|
0:52:10
|
and unity Express is SIP only.
|
|
0:52:14
|
Now, if Unity Express was integrated into CUCM,
|
|
0:52:19
|
this would not be SIP signaling.
|
|
0:52:21
|
It would actually be JTAPI (Java Telephony Application Programming Interface).
|
|
0:52:31
|
So it's a form of CTI or a form of Computer Telephony Integration.
|
|
0:52:37
|
But when CUE is integrated with CUCM, big CUCM uses JTAPI
|
|
0:52:42
|
when it's integrated with CME,
|
|
0:52:44
|
it uses SIP and only those protocols.
|
|
0:52:48
|
And then of course the media, the RTP goes directly
|
|
0:52:51
|
between the voicemail server and the endpoints itself.
|
|
0:52:59
|
Now looking at Unity Express if it's using
|
|
0:53:04
|
or if it's embeded in an older ISR Generation 1 router.
|
|
0:53:09
|
So a 2800 or a 3800 series router, it can be run on one of two devices.
|
|
0:53:15
|
One is called the AIM-CUE and this is a Compact Flash based storage.
|
|
0:53:23
|
This is actually a module that,
|
|
0:53:27
|
so that the CF Flash is actually use as the hard drive
|
|
0:53:30
|
and this is actually a square little module
|
|
0:53:33
|
that you have to open the router up
|
|
0:53:35
|
and you actually plug it into a pin section on the motherboard itself,
|
|
0:53:43
|
a connector that's already open in the motherboard
|
|
0:53:45
|
and you screw it down to the motherboard and then replace the lid.
|
|
0:53:49
|
This includes the ability to have 6 ports,
|
|
0:53:54
|
not only included but that's the maximum
|
|
0:53:58
|
up to 6 IVR sessions or Interactive Voice Response concurrent sessions.
|
|
0:54:03
|
Up to 65 Voicemail boxes and up to 14 hours of Voicemail storage.
|
|
0:54:10
|
Then there is the Network Module Express, sorry, Network Module Enhanced (NME-CUE).
|
|
0:54:19
|
So this is actually a hard drive based storage,
|
|
0:54:22
|
2 and a half inch laptop hard drive surprisingly only 80GB
|
|
0:54:26
|
and 5400 RPM but this is on a network module
|
|
0:54:29
|
that slides into the router without having to take the lid off.
|
|
0:54:33
|
8 ports were included but up to 24 can be licensed.
|
|
0:54:38
|
Up to 24 IVR sessions, up to 275 Voicemail boxes
|
|
0:54:44
|
and up to 300 hours of Voicemail storage.
|
|
0:54:48
|
So those are the only two that can be used on the older ISR's.
|
|
0:54:51
|
On the new ISR-G2's, the 2900 or 3900 series routers,
|
|
0:54:59
|
there are only two modules that can be used
|
|
0:55:02
|
and these are service modules or what are known as SRE (Service Ready Engines).
|
|
0:55:09
|
So the Integrated Services Module-Service Ready Engine
|
|
0:55:14
|
is what that stands for -300 and -700.
|
|
0:55:17
|
They're both hard drive based,
|
|
0:55:19
|
they both slide into the router.
|
|
0:55:21
|
There's no aim port accessible inside.
|
|
0:55:25
|
With the 300, two ports come included,
|
|
0:55:29
|
10 max, 10 IVR sessions, 100 Voicemail boxes,
|
|
0:55:34
|
and 60 hours of Voicemail storage.
|
|
0:55:37
|
And with the SRE-700, 4 ports are included,
|
|
0:55:40
|
32 max, 32 IVR sessions, 300 Voicemail boxes
|
|
0:55:45
|
and up to 600 hours of storage.
|
|
0:55:50
|
And finally we come to the last server in our overview,
|
|
0:55:54
|
this is the Unified Presence Server or CUPS.
|
|
0:55:58
|
Again, for capabilities this is a presence add-on to CUCM.
|
|
0:56:02
|
We have the ability to cluster
|
|
0:56:06
|
in order to provide high availability for Presence.
|
|
0:56:10
|
This adds presence capable application support for clients.
|
|
0:56:15
|
We can have up to 15,000 users per cluster, not per server.
|
|
0:56:21
|
This gives us the ability to add things like instant messaging,
|
|
0:56:24
|
something called the IP Phone Instant Message Service
|
|
0:56:28
|
or IP Phone Messenger (IPPM).
|
|
0:56:31
|
This gives us the ability to also use instant messaging
|
|
0:56:36
|
between CUPC (Cisco Unified Presence Client).
|
|
0:56:44
|
This also gives us the ability to have 3rd party Presence integration
|
|
0:56:49
|
through SOAP and AXL as well as,
|
|
0:56:54
|
so in terms of being able to query the server
|
|
0:56:57
|
for Presence for maybe live intranet or call center website
|
|
0:57:03
|
or things like that.
|
|
0:57:05
|
Also gives us the ability to provide interdomain federations
|
|
0:57:08
|
with other presence servers via SIP and either
|
|
0:57:12
|
using the Extensible Message Presence Protocol (XMPP).
|
|
0:57:23
|
Or also the Jabber Extensible Communication Platform.
|
|
0:57:27
|
So if you happen to remember or maybe you didn't know
|
|
0:57:31
|
but Cisco actually bought the intellectual property
|
|
0:57:34
|
for the open source Jabber and they have added
|
|
0:57:40
|
that Jabber Extensible XCP (Extensible Communications Platform)
|
|
0:57:47
|
into their Presence server and their using Jabber quite a bit now.
|
|
0:57:53
|
In fact that's what they use for CUPC communications,
|
|
0:57:56
|
back to CUPS if you are using an 8.0 client.
|
|
0:58:00
|
If you are using a 7.0 client with a 8.0 server,
|
|
0:58:02
|
it's still using the SIP or Simple signaling.
|
|
0:58:08
|
We also have the ability to, or not really also
|
|
0:58:12
|
but adding on to that, the ability to provide these interdomain federations.
|
|
0:58:16
|
So what these are networking between or presence information
|
|
0:58:20
|
between other CUPS servers, other CUPS clusters.
|
|
0:58:23
|
Also between WebEx connect, between Google Talk
|
|
0:58:27
|
which uses Jabber and CUPS and then also between Microsoft link,
|
|
0:58:32
|
what was formerly know as the OCS server.
|
|
0:58:38
|
So some of the features again,
|
|
0:58:40
|
This is built on the Cisco Unified Client Services Framework known as CSF.
|
|
0:58:45
|
This can deal with or provide support for the Unified Presence Client
|
|
0:58:52
|
the UPC or CUPC as a standalone soft phone or as a computer based control
|
|
0:59:04
|
over your actual hardware deskphone,
|
|
0:59:10
|
There is SRST failover for support but only for soft phone mode
|
|
0:59:15
|
There is instant messaging support for direct messages,
|
|
0:59:19
|
group messages and even persistent chat.
|
|
0:59:22
|
We do have voice and H264 video chat both supported
|
|
0:59:28
|
for the CUPC (Cisco Unified Presence Client).
|
|
0:59:32
|
Conferencing can be done via Cisco MeetingPlace or Cisco WebEx
|
|
0:59:37
|
whichever you happen to have installed or contracted.
|
|
0:59:41
|
And IMAP integration can actually be performed with Unity
|
|
0:59:45
|
or Unity Connection but not Unity Express for MWI.
|
|
0:59:50
|
So the CUPC client can actually indicate not only
|
|
0:59:55
|
if you have a messge but actually how many messages
|
|
0:59:57
|
and it does this through IMAP integration.
|
|
1:00:00
|
And then also LDAP integration really is crucial
|
|
1:00:05
|
in order to provide the full features of the actual CUPS server
|
|
1:00:11
|
for directory services.
|
|
1:00:15
|
So looking at traffic flow between CUPS.
|
|
1:00:19
|
So the CUPS server, the CUPC, the CUP Client,
|
|
1:00:22
|
the Cisco Unified Presence Client on a laptop
|
|
1:00:25
|
and a CUCM cluster and a hardware phone.
|
|
1:00:28
|
First of all the CUPS server,
|
|
1:00:30
|
one of the things that many people don't recognize
|
|
1:00:32
|
or realize is that this actually joins in a sense the CUCM cluster
|
|
1:00:38
|
and there actually is one way database replication between the publisher
|
|
1:00:43
|
pushing database information about the CUCM cluster one way.
|
|
1:00:48
|
Notice all my other arrows have arrows on both sides
|
|
1:00:52
|
but the database replication is one way pushed down to CUPS.
|
|
1:00:56
|
and actually if you do a show,
|
|
1:00:58
|
i think it show network cluster on the CUCM server,
|
|
1:01:03
|
let's say on the publisher,
|
|
1:01:05
|
you'll see not only the publisher
|
|
1:01:06
|
but all the other subsscriber servers here in the cluster.
|
|
1:01:10
|
One of those subscriber servers will actually be the CUPS server
|
|
1:01:14
|
or multiple if you have multiple CUPS servers.
|
|
1:01:17
|
There is CTI or JTAPI information that is sent back and forth between CUPS
|
|
1:01:26
|
and CUCM and this is primarily if the CUPC
|
|
1:01:28
|
the Cisco Unified Presence Client running on a laptop
|
|
1:01:33
|
or desktop is doing deskphone control.
|
|
1:01:37
|
So it's got the software based client
|
|
1:01:39
|
but audio RTP streams are not coming to the client
|
|
1:01:43
|
when in deskphone mode instead it is simply signaling
|
|
1:01:49
|
to the CUPS server and the CUPS server uses CTI over the CUCM server
|
|
1:01:53
|
and the CUCM server uses SIP or skinny signaling down to the phone
|
|
1:01:57
|
to tell it to go off-hook or to dial this number or to answer a call, etc.
|
|
1:02:04
|
That can be done in deskphone control mode.
|
|
1:02:08
|
In soft phone mode, and you can only run one at a time
|
|
1:02:12
|
and possibly you're an adminisitraror
|
|
1:02:14
|
or you would only provision one or the other for the phone
|
|
1:02:19
|
but normally you provision both and then you just switch back and forth
|
|
1:02:23
|
between soft phone mode and deskphone control mode.
|
|
1:02:27
|
But if i'm running soft phone mode on the CUPC,
|
|
1:02:30
|
then RTP media does come to my laptop.
|
|
1:02:35
|
So if i go off hook on the CUP client on my laptop in deskphone control mode
|
|
1:02:48
|
then i don't hear any dial tone.
|
|
1:02:53
|
My deskphone hears the dial tone because i told cups server
|
|
1:02:57
|
to tell CUCM to tell the phone to go off hook.
|
|
1:03:00
|
If I'm in soft phone mode and i go off hook,
|
|
1:03:04
|
my desk phone does nothing except it shows
|
|
1:03:06
|
that i have a call on that first line
|
|
1:03:08
|
beacuse it's a shared line between the two
|
|
1:03:11
|
and all the audio actually comes the RTP media,
|
|
1:03:14
|
the audio, the dial tone or ringback or actual voice communication
|
|
1:03:20
|
if i established a call comes to the actual laptop.
|
|
1:03:25
|
Now i mentioned this already, XMPP is what is used
|
|
1:03:29
|
between CUPC and CUPS if running the 8.0 client,
|
|
1:03:35
|
of course the server has to be 8.0 as well.
|
|
1:03:38
|
If the CUPC client is the 7.0 client,
|
|
1:03:42
|
then it uses the SIP or more specifically it uses Simple
|
|
1:03:47
|
which is the SIP markup language for Presence.
|
|
1:03:52
|
CUPC and CUCM do talk via SIP and Simple
|
|
1:04:01
|
and this is used for soft phone registration to the actual CUCM
|
|
1:04:08
|
especially if it's in soft phone mode and CUPS and CUCM cluster
|
|
1:04:17
|
also speak through SIP or Simple.
|
|
1:04:21
|
In fact, so we have the CTI control that's only for deskphone control mode.
|
|
1:04:26
|
We have the one way database replication but all presence information
|
|
1:04:32
|
between the CUCM cluster and CUPS is done through SIP/Simple
|
|
1:04:38
|
and specifically with the SIP publish or SIP notifying method.
|
