Viewing Extensions in the Lync Client

In companies that use extensions off the main number, I’ve always recommended they format their numbers using the main office number for the “base” and followed by the extension (ie. tel:+14165551111;ext=222 for North America, or tel:+4420555111;ext=222 for elsewhere).  In Active Directory, I’ve recommended they use the same type of format (ie. +1.416.555.1111 X222), and use the Company_Phone_Number_Normalization_Rules.txt to enforce the same formatting.

Background

For those unfamiliar with how Lync displays phone numbers, the Lync client will only display numbers it can parse from Active Directory.  It doesn't show the actual TelURI defined in the Lync Control panel, which is a good thing, because there are often additional settings applied there that administrators don't want to have appear in the address list.  The Company_Phone_Number_Normalization_Rules.txt file is helpful in re-formatting phone numbers in AD to the E.164 format we like to see in Lync.  For a good overview of the Company_Phone_Number_Normalization_Rules.txt, see Jeff Schertz's blog entry on the topic.

The Problem

I've found that when using the Lync 2010 client, North American phone numbers would always display with the typical North American formatting properly along with the extension when dialed, (ie. +1 (416) 555-1111 X222).  However, when applying similar rules to international numbers, the extension would never display.  It would show +4420555111 instead of +4420555111 X222.  The number would be dialed properly, but it would be confusing to the user, because the extension wouldn't appear.

Fast forward to Lync 2013 client, and the situation appeared even worse.  Even North American phone numbers with extensions wouldn't appear properly when dialled.  I assumed this was a client bug that was getting worse with each client revision.

The Solution...Sort of

Fellow Lync MVP Tom Pacyk of ConfusedAmused.com brought to my attention a little known (well, little known by me) feature added with the October 2012 update for the Lync 2010 client.  In short, the client update finally fixed the bug for international extensions, but with a catch....you had to first add a new client policy entry to your Lync client policy.  If only the global client policy is being used, the commands are:

$x = New-CsClientPolicyEntry -Name "ShowExtensionInFormattedDisplayString" -Value "True"
$y = Get-CsClientPolicy -Identity Global
$y.PolicyEntry.Add($x)
Set-CsClientPolicy -Instance $y

Since this policy setting was a part of the Lync 2010 client updates, it would be very easily missed by someone whose focus has changed to Lync 2013.  I can only assume the same settings are required in a pure Lync 2013 environment, although there is very little information on the topic.

If you're running Lync Server 2010, it appears as though you need at least the October 2012 Lync 2010 server updates for this to work.  I tried to apply this client policy entry to a client's deployment whose got Lync 2010 and 2013 in parallel, but the Lync 2010 environment is a bit out of date.  The policy entry showed up properly when I ran the following:

$y = Get-CsClientPolicy -Identity Global
$y.PolicyEntry

...but neither my Lync 2010 or 2013 clients would show extensions properly.  Applying the policy entry to a 2010/2013 server mix where the 2010 environment was up-to-date worked fine...at least for 2010 clients...mostly.

So, make sure your Lync 2010 infrastructure is up-to-date before attempting to apply this setting.

What's Still Not Working

I setup a few test users with extensions off a fictitious Toronto, Canada and London, UK main office number.  The Toronto user's office number in AD was set to +1.416.555.1111 X111, and the London user's number in AD was set to +44.20.444.2222 X999.  Both users were setup in Lync with their TelURI properly defined as tel:+14165551111;ext=111 and tel:+44204442222;ext=999 respectively.

A Company_Phone_Number_Normalization_Rules.txt file was placed in the ABFiles folder on the shared folder for the Lync server pool with the following rule to ensure Lync would be able to parse phone numbers  with extensions in Active Directory properly:

\+(\d{6,14})\D+(\d+)
+$1;ext=$2

When testing with the Lync 2010 client, North American numbers showed up properly in both the list view and the dialing view, as shown below:

However, international numbers did show numbers a bit differently between the views, using the ;ext= from the TelURI instead of the X:

Interestingly enough, the Lync 2013 client doesn't seem to fully honour the extension setting for either North American or international numbers.  The number shows up properly when hovering over the phone icon, but only shows the main number when dialing:

North America

United Kingdom

Needless to say, this can be very confusing to the average user.

This appears to be a bug introduced with the Lync 2013 client.  I've brought this up with the Lync product managers, and we should see a future patch that will address the issue for Lync 2013 clients.

Location-Based Routing in the Lync Optimizer

The Lync Server Update for February 2013 included a much asked for feature: location-based routing. Location-based routing basically ensures the routes that phone calls can take are dependent on where the user is currently located.

Before the February 2013 update, the path calls would take were solely determined by the assigned Lync voice policy.  Normally,  call routes are somewhat local to the user (at least within the same country). If a user were to go to another office on the other side of the planet, their calls would still route via the same paths as if the user were at their home office.

Normally, this isn't a very big issue, but some countries have very strict rules about how calls may enter and exit the country. India is an example where they don't want to see any calls leave the country via the Internet.  All calls must leave the country via the PSTN.  Toll bypass or least-cost routing is not allowed.

Fellow Lync'er Richard Brynteson wrote up a great post on how to setup location-based routing in Lync 2013 CU1.  He even put me on the spot by saying that I was working on adding location-based routing to the Optimizer.  So, that pretty much meant that I had to do it.

So, in the spirit of the Lync Optimizer doing all the work required to setup Enterprise Voice in Lync, I've added an option to configure location-based routing when the Optimizer is run for a given Lync deployment.

There is a single prerequisite in order to make it all work. The Lync environment must have at least one Lync network region, site and subnet configured (preferably/obviously for the location where location-based routing is to be enabled).  If call admission control has already been deployed, then this should already be done.  If not, the Lync network site/subnet can be setup manually by following the guide on Technet. If AD Sites & Services has been fully populated with site/subnet information, I've created a script that will transfer this into Lync.

When the Optimizer runs, it will check to see if the deployment is running Lync 2013 and has at least one network site defined.  It will then prompt the user to apply location-based routing to that site (or others if more than one site is found).

If you are unsure whether or not to apply location-based routing, then DON'T DO IT.  It will ensure that all calls made from within the network site use only the routes defined for that site.  It may have unintended consequences for users visiting the site.  It can be undone, but if you are not comfortable with the inner workings of Lync, its best not to open up that can of worms.

As a final note, the Lync Optimizer now includes dial rules for India, which is one of the main reasons why location-based routing was added to Lync in the first place.

Best of luck out there!

Automatically Creating Lync Sites and Subnets using AD Sites

To use call admission control or location-based routing in Lync, one of the first steps is to define all the sites and subnets used in the company and to assign the sites to network regions.  In larger companies, this can be a tedious process if this information is not stored in a central, easily accessible location.

One place where network site/subnet information can sometimes be centrally found is within Active Directory itself.  Active Directory and other applications such as Exchange uses site/subnet information to determine where to direct clients for authentication services (among other things).  Ideally, every site and subnet should be defined using the Active Directory Sites and Services snap-in.  If the Active Directory administrator has diligently defined all the sites/subnets, this information can be easily parsed and used in Lync.

I've created a short script that will automate the creation of network sites and subnets using the information stored in AD Sites and Services.  Again, this script is only useful if all the sites/subnets have been defined in AD. 

Copy the below script into a .PS1 file on a Lync server and run it.  As always, be careful and test this in a lab first.  I can't be responsible for any damage this script will cause.
# Load Active Directory Powershell module if its not already loaded if(-not(Get-Module -name ActiveDirectory)){Import-Module ActiveDirectory} # If a Lync network region doesn't exist, create one and name it the same as the first Lync site # If one or more does exist, select the first one associated with the first Lync site if(-not(Get-CSNetworkRegion)) { $SiteName = (Get-CSSite 1).DisplayName New-CSNetworkRegion $SiteName -CentralSite 1 } else { $SiteID = (Get-CSSite 1).Identity $SiteName = (Get-CSNetworkRegion | Where {$_.CentralSite -eq $SiteID} | Select-Object -First 1).NetworkRegionID } # Get the ConfigurationNamingContext from AD and go to the Sites container $ConfigDN = (Get-ADRootDSE).ConfigurationNamingContext $SiteContainerDN = ("CN=Sites," + $ConfigDN) # Select all the site objects $Sites = Get-ADObject -SearchScope OneLevel -SearchBase $SiteContainerDN -filter {ObjectClass -eq 'Site'} foreach ($Site in $Sites) { # Create Lync network sites based on AD site $SiteObj = Get-ADObject $Site.distinguishedName -properties "siteObjectBL", "description" New-CSNetworkSite -NetworkSiteID $SiteObj.Name -Description $SiteObj.Description -NetworkRegionID $SiteName # Add each defined AD site subnet to the Lync network site object foreach ($SubnetDN in $SiteObj.siteObjectBL) { $Subnet = Get-ADObject -Identity $SubnetDN -properties "description" $SubnetName = $Subnet.Name #Subnet name is in format 192.168.100.0/24 $SubnetSplit = $SubnetName.Split("/") #Need to split subnet name into IP and mask for command $SubnetID = $SubnetSplit[0] $SubnetMask = $SubnetSplit[1] New-CSNetworkSubnet -SubnetID $SubnetID -MaskBits $SubnetMask -Description $Subnet.Description -NetworkSiteID $SiteObj.Name } } #Create subnet object with 32-bit mask for all public edge A/V IP addresses $EdgeServerPools = Get-CSService -EdgeServer | Where {$_.SiteID -ne "Site:BackCompatSite"} foreach ($EdgeServerPool in $EdgeServerPools) { $EdgeServers = (Get-CSPool $EdgeServerPool.PoolFQDN).Computers foreach ($EdgeServer in $EdgeServers) { $NetworkInterfaces = Get-CSNetworkInterface -ComputerFQDN $EdgeServer | Where {$_.Interface -eq "External" -and $_.IPAddress -ne "0.0.0.0"} if (($NetworkInterfaces).Count -eq 3) { $NetworkInterface = $NetworkInterfaces | Where {$_.InterfaceNumber -eq 3} } else { $NetworkInterface = $NetworkInterfaces } if ($NetworkInterface.ConfiguredIPAddress -ne $NULL) { $IPAddress = $NetworkInterface.ConfiguredIPAddress } else { $IPAddress = $NetworkInterface.IPAddress } $Description = "A/V external edge IP for " + $EdgeServer New-CSNetworkSubnet -SubnetID $IPAddress.IPAddressToString -MaskBits 32 -Description $Description } }

The script will first check to see if a Lync network region already exists. If so, it will select the first one it sees as the base for all the subnets.  If one doesn't exist, it will create one using the first Lync site as the base. If need be, you can define additional network sites later.

The script will then create all the sites/subnets based on information from AD. All the sites will be assigned to the one network region selected according to the rules in the previous paragraph. Since there is no way for the script to know which site belongs to which region, this will have to be done manually if multiple Lync network regions exist.

IMPORTANT: The script will also create the required subnets for all the Lync edge A/V public IP addresses with a 32-bit subnet mask as per the requirements for call admission control.  You will have to manually assign these subnets to the sites where you deployed the edge servers.

While this script doesn't try to do everything associated with setting up call admission control, it does go a long way towards automating a good chunk of the process. A future revision of the Lync Optimizer will hopefully close the loop and help completely automate the work required to setup location-based routing.

As a final side note, one thing I don't understand is why Lync can't leverage AD Sites & Services for this information directly.  Exchange versions since 2007 have used AD Sites & Services for mail routing, so its not like it hasn't been done before.  Maybe a future version of Lync will add this functionality.  Until then, this script should help those who have well-defined site/subnet information in Active Directory.

Assigning a Throttling Policy to all iOS 6.1 Devices in Exchange 2010

UPDATE 14-Feb-2013: After being linked to from Ars Technica, I felt the need to improve the script. My original version of the script tried to use the UserDisplayName field that is returned from Get-ActiveSyncDevice.  However, that field had a maximum number of characters, and a number of longer user DNs got cut off, which would make the script fail for that user. I figured out a way around it, and the script has been updated.

If you're an Exchange administrator, you have probably heard about the issues caused by Apple's iOS 6.1 on Exchange servers.  It seems that screwy calendar code in iOS 6.1 ends up generating tons of transaction logs in Exchange, and can bring CAS servers to its knees.

As an aside, its rather timely that this has come up just after the new Blackberry 10 came out.  Coincidence??????  Heh.

The temporary workarounds that MS recommends were to either apply a throttling policy to affected users, or block iOS 6.1 devices completely.  Since an all-out block is generally a Bad Thing, throttling policy is the way to go.  However, determining who is using iOS 6.1 and applying a throttling policy to them was not provided.

A company I do work for experienced the exact issue.  Exchange transaction logs were being generated at a prodigious rate.  So, rather than blocking them, we went the throttling route.  First, I created a throttling policy called iOS61 and set the ActiveSync throttling to the lowest recommended values:

New-ThrottlingPolicy -Name iOS61 -EASPercentTimeInCAS 10 -EASPercentTimeInAD 10 -EASPercentTimeInMailboxRPC 10

Then, I created a short Powershell script to apply the policy to any user with a device using iOS 6.1:

$DeviceList = Get-ActiveSyncDevice -ResultSize Unlimited | Where {$_.DeviceOS -match "iOS 6.1"}
foreach ($Device in $DeviceList)
{
$DeviceDN = $Device.DistinguishedName
 $MBName=$DeviceDN.SubString($DeviceDN.IndexOf("CN=ExchangeActiveSyncDevices,")+29)
Set-Mailbox $MBName -ThrottlingPolicy iOS61
}

The script will likely throw up some warnings about completing the command but not making any changes.  This happens because users may have more than one device, and the script tries to update the mailbox for every device it finds.  Nothing to be alarmed about.

Quick and dirty, but gets the job done.

Once Apple comes up with a solution, you can remove the throttling policy like this (assumes the users didn't have a throttling policy before this whole debacle):

Get-Mailbox -ResultSize Unlimited | Where {$_.ThrottlingPolicy -eq 'iOS61'} | Set-Mailbox -ThrottlingPolicy $NULL

Limiting Call Forwarding and Simultaneous Ringing in Lync 2013

In Lync 2010, users were able to forward calls or setup simultaneous ringing to any number that they were able to call themselves.  This could lead to people forwarding all their calls to a long distance or even international number (assuming their voice policy allowed international calling).  This can lead to increased telephony costs to the company, and there wasn't very much the administrator could do to control this, short of disabling the feature entirely.
A new feature in Lync Server 2013 Enterprise Voice allows administrators to limit where users are able to setup call forwarding/simultaneous ringing.  You'll see the new option at the bottom of the voice policy screen in Lync 2013 Control Panel (as highlighted in red).

The selectable options are:

Route using the call PSTN usages (default) - essentially means to allow users to forward calls to any destination phone number they are allowed to call (the same behaviour as in Lync 2010).  So if a user is allowed to call international numbers, they will be allowed to forward/simultaneous ring international numbers. The option wording here could really use some improvement, because to me it doesn't make much sense as it currently stands.
Route to internal Lync users only - allows call forwarding/simultaneous ringing only to other Lync users within the company.
Route calls using custom PSTN usages -  allows administrators to limit call forwarding/simultaneous ringing to only the PSTN usages defined by the administrator.

When the last option is selected, the Control Panel will display a box where administrators can select the appropriate PSTN usages.  For instance, to limit call forwarding/simultaneous ringing to only local numbers, select a local PSTN usage, as shown below:

Administrators can add multiple usages to allow calls forwarding/simultaneous ringing to any combination of PSTN usages.  Since voice policies can be applied either globally, site-wide or down to the user level, this should allow administrators to grant different policies to any number of users.

This assumes the Enterprise Voice deployment has been designed granular enough to break out local, national and international calling.  If there is only a single catch-all usage for all calls, then this won't work and the Enterprise Voice configuration will have to be re-designed.

Luckily, the Lync Dialing Rule Optimizer (shameless plug) creates usages for local, mobile, national, premium, and international usages, which should be enough for any administrative need.

Unfortunately from the user's perspective, they won't get a notification in the Lync client if they try to forward/simultaneous ring a phone number that is outside the dialing areas allowed by their voice policy.

For example, assume a user's voice policy only allows forwarding/simultaneous ringing to local numbers. If the user sets up simultaneous ringing to a number outside the local dialing area, they won't get any feedback that it isn't allowed by the policy.  When someone phones the user, their Lync devices will ring as usual, but the simultaneous ring number simply won't ring.  This could generate an unnecessary helpdesk call, which could also confuse the helpdesk, if they haven't been made aware of the policy.

Similarily, the user doesn't get any feedback in the Lync client when setting up call forwarding to a number outside the allowed dialing areas.  In this case, calls will simply go straight to Exchange voicemail without attempting to forward.  However, the user will get an email stating the forwarding isn't allowed.  Slightly more helpful, but still likely to generate a helpdesk call.

Curiously, the interface still refers to Office Communicator, but since I'm still running Exchange 2010, maybe this is to be expected.

The situation would be greatly improved if the Lync client would be more informative in this respect.  In my perfect world, there would be no war, global warming would turn Canada into a tropical paradise, gin and tonic would flow freely from taps all around my house, and the Lync client would provide feedback like this when users try to forward to a number that isn't allowed:

A man can dream, can't he?

Either way, the new control administrators have to limit call-forwarding/simulring is a much needed new feature, and I'm glad it made it into Lync 2013.

Least-Cost/Failover Routing in the Lync Dialing Rule Optimizer

The Lync Dialing Rule Optimizer tries to take care of all the Enterprise Voice configuration necessary for all deployments.  On most fronts, it does a pretty good job (IMHO).  One area that has been lacking is the ability to automatically arrange PSTN usages to provide least-cost/failover routing for multi-site Lync deployments. Once an administrator has run the Optimizer for all their sites, they have to manually configure the voice policies to provide least-cost/failover routing.

If you're not familiar with how least-cost/failover routing works in Lync, please refer to my post on the subject in my Lync Enterprise Voice Best Practices series.

With version 9.0 of the Lync Dialing Rule Optimizer, Lync administrators now have the option to create least-cost/failover routing between all sites that have been deployed using the Optimizer (or at least follows the same naming conventions used by the Optimizer).

When the Optimizer-generated script detects multiple Lync sites, it will prompt the user if they want to apply least-cost/failover routing to the voice policy being generated.  If the user allows it, the Optimizer will add PSTN usages for all the other sites to the new voice policies.  This assumes the existing PSTN usages are named with the country abbreviation first, and ends with either Local, National, International etc in any of the languages currently supported by the Optimizer as in the following examples:  UK-Leeds-113-Local, IT-Rome-06-Nazionali.

The output can be best summarized by way of example.  Assume a company has Lync Enterprise Voice deployed in the following locations:

Toronto, Canada

Vancouver, Canada

London, UK

Rome, Italy

Belém, Brazil

After running the Optimizer script against all sites, the PSTN usages for the Toronto International user voice policy would look like this (you might want to sit down for this):

NA-ON-Toronto-416567-Local

NA-BC-Vancouver-604678-Local

NA-ON-Toronto-416567-National

NA-BC-Vancouver-604678-National

NA-ON-Toronto-416567-Premium

NA-BC-Vancouver-604678-Premium

NA-ON-Toronto-416567-Service

UK-London-20-Local

IT-Roma-06-Locali

BR-Belém-91-Local

UK-London-20-Mobile

IT-Roma-06-Cellulari

BR-Belém-91-Celular

UK-London-20-National

IT-Roma-06-Nazionali

BR-Belém-91-Nacional

NA-ON-Toronto-416567-International

NA-BC-Vancouver-604678-International

UK-London-20-International

IT-Roma-06-Internazionali

BR-Belém-91-Internacional

The Optimizer will group PSTN usages from the same country near the top, using the assumption that most of the calls will be made to in-country locations and that the fewer PSTN usages the system needs to evaluate for each call, the better.

The ordering is such that least-cost routing and failover routing will occur.  PSTN usages from other countries will be placed lower in the list while still providing least-cost and failover routing for all calls.  If there are multiple Lync sites out-of-country, the ordering of PSTN usages within the National and International usages will be somewhat random. Administrators should review the PSTN ordering to make sure it suits their needs.

It is assumed that premium numbers can only be dialed from in-country locations, so out-of-country premium PSTN usages are not assigned. Also, only the local service PSTN usage is assigned for similar reasons.

All calls will use the least-cost route where possible. If a route assigned to a specific usage is unavailable, calls will use a route in another site, again keeping the call in-country if possible.

With the inclusion of least-cost/failover routing in the Lync Optimizer, administrators now have a tool that can completely build the Enterprise Voice environment for even the largest enterprise-level Lync deployment.

As always, feedback is greatly appreciated. Let me know if you come across any issues with functionality or scalability as I haven't tested if there is a limit to the number of PSTN usages that can be assigned to a voice policy.

Blocking International Calls to Specific Countries in Lync

A question that often comes up in my travels is that Lync administrators want an easy way to allow users to dial internationally but exclude specific countries for some reason.

This is very easy to accomplish once you understand the ins and outs of regular expressions, and assuming you follow all my best practices regarding number normalization, and Enterprise Voice setup.  To summarize, every number a user enters in Lync should be normalized to E.164 standards, which starts with a + followed by the country code, then the area/city code and finally the local subscriber number.  A Canadian example (country code 1) would be +14165551111.  A UK example (country code 44) would be +442033334444.  If you use the Lync Dialing Rule Optimizer to create your Enterprise Voice configuration, you'll be all set.

When all numbers are normalized to E.164, and you already have separate usages and routes for local, national and international calls, it makes it very easy to design regular expressions that meet any specific criteria required by your business.

Say your company is based in North America, and is required to block international calls to Iran (98) and North Korea (850).  The Lync Dialing Rule Optimizer routing rule for international calls is this:

^\+[2-9]\d{6,14}$

This rule allows any number that doesn't start with a 0 or 1 to be routed out. Only North American countries (US/Canada and some Caribbean countries) use the country code 1.  All other country codes start with the digits 2-9.  So essentially, this rule allows calls to any international destination.

To block calls to Iran and North Korea, modify the rule to look like this (new stuff highlighted in yellow):

^\+(?!98|850)[2-9]\d{6,14}$

This will allow any number starting with 2-9, but excludes numbers that start with 98 or 850 as required.

If you're not in North America and used the Lync Dialing Rule Optimizer to create your Enterprise Voice setup, your international rule looks a little different, because we want to make sure that North American numbers are formatted correctly (as highlighted in blue), while making sure that it does not try to route national numbers as international (using UK +44 as an example, shown in green):

^\+((1[2-9]\d\d[2-9]\d{6})|(?(?!(44))([2-9]\d{6,14})))$

To block calls to Iran and North Korea using this example, modify the rule as follows (new stuff highlighted in yellow again):

^\+((1[2-9]\d\d[2-9]\d{6})|(?(?!(44|98|850))([2-9]\d{6,14})))$

Now, anybody who tries to dial a number in Iran or North Korea will be met with a notice that the call couldn't be completed.

This method is pretty granular as you could create separate routes to block countries selectively or for a specific group of users.  If you want to just block those numbers globally, you can use Unassigned Numbers to provide an announcement to users that those numbers are not allowed to be dialled. The only downside to that method is that you need to know in advance how long phone numbers are in the country you want to block.

Happy Enterprise Voicing everyone!

Localized UK Dialing Rules for the Optimizer

Using data from a number of web sources and some helpful users, I've been able to create customized Lync dialing rules for over 30 countries worldwide and made them available to all via the Lync Dialing Rule Optimizer.  These customized dialing rules account for country-specific dialing patterns for local, national, international, toll-free, mobile, premium and service numbers.

However, true localized dialing rules have been out of reach for countries other than those in North America until now.  For all non-North American countries, I have been forced to create generic dialing patterns for local numbers simply because I don't have access to any source of local dialing rule data.  This isn't generally a problem because most local dialing rules are different enough from other patterns that conflicts are rare.  For example,  a local dialing rule for a country might specify that the number start with digits 2-9 and are between 5 and 6 digits in length.  As long as there aren't any other dialing patterns in that country that use that same pattern (ie national, international, mobile, service, toll-free or premium), there won't be a problem.  I try to make sure that is the case for all dialing rules created for the Optimizer, but it is sometimes difficult.

The United Kingdom has more than 60 different local dialing rules. Local numbers can be 4 to 8 digits in length and area codes have from 2 to 5 digits. Some 41 area codes have mixed length numbers where some numbers have a total 10 digits and others have only 9. In some area codes the local number cannot begin with certain digits, or certain digits will not come into use for many decades.  The generic UK local dialing rule used by the Optimizer tries to account for all these in a single regular expression, but obviously some generalizations had to be made.

Enter Ian Galpin of the UK.  Ian has made it a personal mission to ensure that the UK has a complete set of dialing rules written as regular expressions.   Over the past year, he's helped me ensure the regex for UK dialing rules are correct.  He's also created a localized set of dialing rules for every area code in the UK as an XML file along with a list of all the area codes used in the UK.  He's done this to help various telephony vendors create the appropriate rulesets for PBX deployments. I've been able to incorporate that into the Lync Dialing Rule Optimizer, so that UK users now have true customized local dialing rules for Lync that will differ depending on which area code you select in the Optimizer.

I've modified the Optimizer code to be able to use local dialing rules for other countries if available.  If anybody can point me to a similar source as provided by Ian for other countries, it would be greatly appreciated.

Lync 2013 is Code-Complete

For those who haven't heard through Twitter or other channels, Lync 2013 is code-complete, with a general availability target of first quarter 2013 as part of the Office 2013 suite.  It isn't clear from the announcement, but this is for the Lync 2013 CLIENT, not server.  Since Exchange 2013 and Sharepoint 2013 have been announced, I'm sure we'll be hearing about Lync Server 2013 achieving the same milestone fairly soon.

UPDATE: Well, confusion reigned among some of the MVP mailing lists.  While some were saying the Lync 2013 announcement was for the client, others were saying it was for the server.  Seeing how all the other servers in the "suite" have been announced, I think its safe to say that Lync Server 2013 is part of it.

Lync Branch Site Options and Recommendations

A large-scale Lync deployment typically consists of at least one central site that contains either a Standard or Enterprise Edition pool and one or more branch sites that don't warrant having a full-blown Lync deployment, but still need local voice connectivity in case of a WAN outage.

You have 4 options when considering what to deploy at a branch site:

1. Survivable Branch Appliance (SBA) - an all-in-one solution that consists of a PSTN gateway with the appropriate interface required for connectivity to the PSTN and a "Lync-Lite" installation that has only the Lync registrar and mediation server roles.
2. Survivable Branch Server (SBS) - A "Lync-lite" installation containing the Lync registrar and mediation server roles.
3. A PSTN gateway and a standalone Lync mediation server
4. A full-blown Lync deployment (typically Standard Edition)

Microsoft doesn't give much guidance as to which option is right other than saying that an SBA is appropriate for branches up to 1000 users, an SBS or 2 SBAs is good for 2000 users, and anything beyond that should be a Standard Edition Server.

What a lot of people don't realize is that there is more to consider than simple user counts.  Let's start by looking at where an SBA is appropriate, and where it isn't.

Survivable Branch Appliance

As noted earlier, a Survivable Branch Appliance is an appliance that contains a PSTN gateway along with a small Windows Server 2008 R2 deployment with the Lync registrar and mediation server roles. SBAs are purchased from vendors such as NET, Audiocodes, Dialogic, or Ferrari, among others. An SBA is paired with a central site that has a full-blown Lync deployment and is designed to be dropped into an office that doesn't have much local IT support.

Lync clients are homed on an SBA, and will register/login directly against it. PSTN calls are routed through the mediation server role and onto the PSTN gateway component of the SBA.

While clients register against the SBA, their contact list is still homed on the central Lync deployment.  Also, all conferencing features and response group functionality is provided by the central Lync deployment.  So, if the WAN link between the central site goes down, clients will lose their personal contact list, the ability to do multiparty web/video conferencing and any response groups whose phone numbers terminate on the SBA won't work.  This is beautifully illustrated by VOIPNorm on his blog post on the topic.  What DOES work is one-to-one IMs/telephone calls, PSTN calling, searching for contacts and any other features that don't require the full Lync deployment's involvement.  Again, see the blog post by VOIPNorm for full details.

When to Use a SBA and When Not To

Use an SBA when you your site meets the following criteria:

• Up to 1000 users (2000 if you use 2 SBAs)
• Have a T1/E1 or analog connection to the PSTN
• Won't miss conferencing or response group functionality during a WAN outage
• Have limited IT staff on site

Don't use an SBA if your site meets any of these criteria:

• Does a lot of conferencing and has a slow/expensive WAN pipe to the central site.  Conferencing services come from the central pool, so this can be a strain on WAN resources.
• Requires high availability for conferencing features and/or Response Group services
• Use a SIP trunk for PSTN connectivity, unless the SBA can be configured as a SIP gateway and doesn't have PSTN interfaces you don't need (they aren't cheap)

Survivable Branch Server

A Survivable Branch Server (SBS) has the same capabilities as an SBA, but without the PSTN gateway component.  It isn't something that you purchase from a vendor.  It's simply a "Lync-lite" server you define in the topology and install yourself on a regular Windows Server 2008 R2 (or Windows Server 2012 for Lync 2013) server.  Like the SBA, it only has the Lync registrar and mediation server roles.

Incidentally,  you won't see an option to define an SBS in the Lync topoology.  Your only option is an SBA.  Since the functionality of the Lync portion of the SBA is identical for an SBS, it doesn't need to be defined separately. Lync doesn't care if the PSTN gateway is part of the same chassis as in an SBA or a separate component as with an SBS.

When to Use a SBS and When Not To

Use an SBS when your site meets the following criteria:

• Have up to 2000 users
• Have the local infrastructure to support a full-blown Windows Server deployment (either VM or physical server)
• Have a SIP trunk connection to a Lync-certified SIP provider, or you already have a standalone PSTN gateway
• Won't miss conferencing or response group functionality during a WAN outage
• Have at least some local IT staff

Don't use an SBS if your site meets any of these criteria:

• Does a lot of conferencing and has a slow/expensive WAN pipe to your central site. Conferencing services come from the central pool, so this can be a strain on WAN resources.
• Requires high availability for conferencing features and/or Response Group services
• Have either a T1/E1 or analog service for PSTN access. You would still need to purchase a PSTN gateway, and would probably be better served by using an SBA (unless either of the above 2 points apply)

PSTN Gateway and Mediation Server

This option is an interesting one in that I've not seen it "in the wild".  A site with just a mediation server role is entirely dependent on the WAN link to the central site. If the WAN goes down, then so do your clients. You need the infrastructure to support a Windows installation, something many small branch sites don't have.  If you have a SIP trunk connection, you may not need a PSTN gateway at all.  Conversely, you may not require a mediation server if you use media bypass to send client audio traffic directly from the client to the PSTN gateway.

When to Use this Option and When Not To

Use a PSTN Gateway/mediation server when your site meets the following criteria:

• You have a 100% reliable WAN connection to your central site or your users will be OK with no functionality in the event of an outage

Don't use a PSTN Gateway/mediation server if your site meets the following criteria:

• Your WAN isn't reliable

Standard Edition Deployment 

Sometimes, you will need a full-blown Lync deployment at your branch site.  With even just a Standard Edition server, local clients will get all the functionality Lync has to offer, so WAN outages won't be a big issue.

When to Use this Option

Use a Standard Edition server when your site meets one or more of the following criteria:

• You have more than 2000 users in a site
• Your WAN link reliability is low
• Your users do a lot of conferencing with each other
• You need local Response Group functionality

Hopefully, this post will give you some better guidance as to what branch role to deploy at your sites. If you have any questions or comments, please let me know. By the way, the above is relevant for both Lync 2010 and Lync 2013.