FYI – the Lync Server 2010 180-day evaluation bits are slated to be available on November 17th which is Lync launch day. You can grab them here on the 17th.
The question I get asked is can you upgrade these bits to production bits at a later point. The answer is yes.
How do I do this?
1) Install Lync Eval RTM bits
2) When the volume license bits for Lync RTM production bits become available, download them and install them overtop of the trial bits.
3) From command line run: “msiexec.exe /fvomus server.msi EVALTOFULL=1 /qb where server.msi = licensed Lync Server media
4) From powershell run: “enable-cscomputer”
Thanks to Andrew Sniderman for the answer!
Updated info for this post can now be found in teched. Please disregard this blog entry. Please refer to http://technet.microsoft.com/en-us/library/gg398833.aspx
Recently I was onsite with a customer and was asked about DNS Load Balancing. How does it work? Why the change from OCS 2007 R2?
DNS Load Balancing
Microsoft Office Communications Server 2007 requires a Hardware Load Balancer (HLB) to provide resilience for the Enterprise pool. This configuration is both expensive and difficult to configure for SIP load balancing. Lync 2010 introduces DNS load balancing as an alternative to hardware load balancing.
How it Works
The front-end servers register their fully qualified domain name (FQDN) as A records in DNS. When the Enterprise pool is created, the pool FQDN is registered to return from DNS the list of IP addresses of all the front-end servers. The client attempts to connect to one of the IP addresses that were returned. If this connection fails, the client attempts to connect to the next IP address in the list until the connection succeeds.
Server Failure and Recovery
When a server fails, the physical registrar sequence is updated to show the server as unavailable and shared amongst all surviving servers by using a server-server heartbeat. Users are redirected to the next server in their logical registrar sequence and are then connected in backup mode. The server will be recovered returning the physical registrar sequence back to its original state.
Server Commission and Decommission
When topology changes occur, the logical registrar sequence is recalculated for all users. Some users are re-homed to a different front-end server in the same pool. When the server is fully operational, the heartbeat process updates the physical registrar sequence. This results in the batched re-registration process. Decommission is very similar to server failure, with the exception of the re-home to a new primary registrar being part of the decommission process. The topology change results in the recalculation of the logical registrar sequence. This step doesn’t happen in a server failure.
You can use DNS load balancing for the SIP traffic on Front End pools and Director pools. With DNS load balancing deployed, you still need to also use hardware load balancers for these pools, but only for HTTP and Distributed Component Object Model (DCOM) traffic. The hardware load balancer is used for HTTP traffic from clients over ports 443 and 80, and for DCOM traffic over port 135 from administrators performing user moves.
Although you still need hardware load balancers for these pools, their setup and administration will be primarily for HTTP traffic, which the administrators of hardware load balancers are accustomed to.
DNS Load Balancing Decision Guidelines
DNS load balancing supported?
DNS load balancing recommended?
Hardware load balancer (only) recommended?
All or most users homed in the pool run Lync Server 2010 clients.
Many users homed in the pool still running older clients.
Interoperates only with other Lync Server 2010 servers.
Interoperates with many servers running earlier versions of Office Communications Server.
Running Exchange UM with Exchange 2010 SP1 (or not running Exchange UM)
Running Exchange UM with earlier versions of Exchange
Before you can use DNS load balancing, you must:
1. From the Lync Server 2010 program group, open Topology Builder.
2. From the console tree, expand the Enterprise Edition Front End pools node.
3. Right-click the pool, click Edit Properties, and then click Web Services.
4. Below Internal web services, select the Override FQDN check box.
5. Type the pool FQDN that resolves to the physical IP addresses of the servers in the pool.
6. Below External web services, type the external pool FQDN that resolves to the virtual IP addresses of the pool, and then click OK.
7. From the console tree, select Lync Server 2010 , and then in the Actions pane, click Publish Topology.
1. For each Front End Server in your pool, create a DNS A Host record that maps the pool FQDN to the IP address of that Front End Server.
For example, if you had a pool named pool1.contoso.edu and three front-end servers, you would create the following DNS entries:
A new feature called server draining enables you to take a server offline without any loss of service to users. When a server is drained it stops taking new connections and calls. These new connections and calls are routed through other servers in the pool. A server being drained allows its sessions on existing connections to continue until they naturally end. When all existing sessions have ended, the server is ready to be taken offline.
Lots of useful information announced from the Communications Server product team at VoiceCon in Orlando has been released:
Microsoft Office Communications Server has been renamed to Microsoft Communications Server. Final naming is still being determined so for now it is code named Communications Server “14”.
What are some new Communicator14 client features?
New Communicator14 client allows for rich capabilities such as:
More social networking features - Rich photos, real time status updates
Ability to have keyword search on skills/specialty within your organization
Activity feeds – similar to other social networking feeds like Facebook
Selectable OCS devices such as headset vs. USB phone, etc. which is nice if you have multiple UC endpoints connected. Swappable during conversation.
Simultaneous ring other devices (cell phone) based per user or caller ID lookup.
Contextual calling – can type in context prior to the call so the recipient can have context about the incoming call
Location awareness – based on subnet or custom location – tied to E911 location awareness
Listen to voicemails right from client
When is CS14 shipping?
CS14 slated to be released by the end of this calendar year
Is there a deep dive of CS14 voice features I can review?
Yes, there is a sample CS14 IP Telephony RFP response released today at VoiceCon that you can refer to here
What are some of the new voice features of CS14?:
Call Admission Control
Malicious Call Trace
3rd party Survivable Branch Appliances
New 3rd party partnerships announced for CS:
Common area, desktop, and conference room IP phones optimized for Microsoft Communicator “14”
Call Accounting and Recording
Nice Systems Quest
Survivable Branch Appliances that provide local telephony services in the event of a wide area network failure
AudioCodes Dialogic Ferrari electronic AG HP NET
Enhanced 9-1-1 call routing services
911 Enable Intrado
We will post more as more CS14 information is released.
New Conversion Tool
A new conversion tool just came out that takes a Live Meeting recording package (service or OCS server) and converts it to a WMV file. This allows you to stream your OCS meeting to playback. Benefits include multi-cast support, multiple bit-rate playback, easier SharePoint integration using “Links” web part.
Coverts standard LM recording such as this:
To a single video file:
Download the tool here.
I tested a OCS 2007 recording yesterday (audio only, no video) and came up with the following results:
Meeting length: 1 hr 16m
Audio-only recording with some app-sharing:
· 1024x768 741mb (default), 10mb/min
· 640x480 371mb, 5mb/min
· 320x240 128mb, 1.7mb/min
Lecture On Demand
I have seen a lot of interest in Education for playback of Live Meeting lectures and meetings. Some schools are posting the Live Meeting lectures to SharePoint and and using technologies like VidiTalk or Windows Media Services to stream the lecture to a PC or a mobile device on demand.
I have also seen schools incorporate recorded Live Meeting Lectures with the free Pod Casting Kit for SharePoint (above) to enable students to download the lecture to their iPod or Zune players.
Exchange Online will be migrating to Exchange 2010 code after we RTM Exchange 2010.
Exchange 2010 in the cloud will address some limitations we currently have with Exchange 2007 Online. Frankly, there are some features of an Exchange 2007 on-premises deployment that you just can’t get in the cloud. With Exchange 2010, the full features of the server are available in the cloud as well.
Second, there will be improved coexistence between hosted and on-premises Exchange deployments. Many schools are considering a hybrid model for e-mail, where they move some e-mail services to the cloud and keep some on-premises. Recognizing this, we’ve designed Exchange 2010 on-premises deployments so that they are easy to extend to the cloud.
Third, a new level of management and control will become available for the administrator with hosted Exchange environments. New web-based management capabilities will be added, and familiar Exchange administration tools, like the Exchange Management Console and PowerShell, will be able to connect to Exchange Online for the first time.
The bottom line here is that because the capabilities of Exchange 2010 are provided as a service as well as a server, you have the freedom to choose the right deployment option for your school without sacrificing functionality. Whether you deploy Exchange Server 2010 on-premises, host your mailboxes with Exchange Online, or combine these two options in a hybrid deployment, your users get business-class messaging and collaboration tools that they need to be productive.
I mentioned that Exchange 2010 brings new features to Online for the first time. There are some Exchange Server features that are not available with today’s Exchange Online Standard offering. They range from relatively minor features, such as customizing the OWA login page, to marquee Exchange 2007 features like Transport Rules, Managed Folders, and Unified Messaging. These features light up in Exchange Online when the service is upgraded to Exchange 2010. Other features, like IMAP access, POP access, SMTP relay, mail forwarding (server-side), “send as” capabilities for shared mailboxes, journaling to on-premises archive, footers, and disclaimers, are scheduled to be implemented in the Exchange Online service even before the Exchange Server 2010 update. Naturally, these will be carried forward when the service is upgraded to Exchange 2010. Contact your sales team for details on today’s deployment of Exchange Online and features.
In addition to eliminating old feature gaps, Exchange Online adds new features from Exchange Server 2010. Because Exchange 2010 was built with services in mind, these new features available in the service right from the start. I don’t have time all the new features for users and service administrators that come in Exchange 2010, but I’ve highlighted a few of them here:
• E-Discovery capabilities across mailboxes
• MailTips to help users prevent email mistakes
• Archiving capabilities to eliminate the need for PSTs
• Transport rules that can apply RMS policies to prevent valuable or sensitive data from leaking out of the organization
One of the great things about having an online deployment, is that you get new features like this much faster and with less effort. The datacenter staff does the heavy lifting of migration and upgrade tasks, leaving you with the ability to roll out up new features that will move your business forward and delight end users. And, it allows you to focus your time and attention on academic priorities.
Here are some of the other features: Conversation view, ignore/move conversation, POP account aggregation, server-side RSS feeds, IM and presence in OWA, SMS in OWA, Firefox and Safari support for OWA premium, OWA side-by-side calendars, shared nickname cache across OWA/Outlook, federated calendar sharing, federated contact sharing, mobile free/busy status, mobile read/reply state, mobile SMS sync, over the air update for Outlook Mobile, mobile device block/allow list, UM message waiting indicator, voicemail preview, call answering rules, protected voice mail, journal decryption, Messaging Records Management 2.0, litigation hold, moderated DLs, self-service distribution groups, self-service message tracking.
In my next blog I’ll discuss management, coexistence and migration to Exchange Online in Exchange 2010.
This was a question coming from an education customer in Minnesota deploying OCS R2 Group Chat Server:
Their Group Chat client was working but the Group Chat administrator console was not. It was getting this error:
“Cannot sign in because of a problem with the chat room service….”
Server 2 received error while subscribing to peer 1, <1> <net.tcp://ocsgroupchat.campus.xxx.edu:8011/MGC/PeerService> <ChannelServer>. Details: Identity check failed for outgoing message. The expected DNS identity of the remote endpoint was 'ocsgroupchat.campus.xxx.edu' but the remote endpoint provided DNS claim 'ocscontent.xxx.edu'. If this is a legitimate remote endpoint, you can fix the problem by explicitly specifying DNS identity 'ocscontent.xxx.edu' as the Identity property of EndpointAddress when creating channel proxy.
Here are some support steps I found to check on your Group Chat Server installation that can relate to this error:
1. Verify the certificate assigned to the group chat server it should have both Server and Client Authentication. If you have only server authentication, sign in to admin console will fail with above error.
2. Above service accounts must be part of RTCUniversalServerAdmins group and also Administrators group of the group chat server. Also add the user name (admin account) with which you are going to sign in to group chat.
3. Enable the admin account along with the above five service accounts for SIP communication on OCS 2007 server R2. Configure them for Federation,PIC,Remote User Access and Enhanced presence.
4. From SQL server management studio ->Security->Logins, Make sure all the above service accounts and admin account, are there. Then here Login properties, General-> default database for each of the account should be "GCDB", Under user mapping->check db_owner for all the service accounts.
5. Under GC admin sign in console->Edit Accounts Settings->Automatic Configuration->Uncheck "Use my Windows credentials to log in automatically" , then under Office communications Server leave Host "blank", select encrypted radio button, under Group Chat Server Settings leave Use default server address box unchecked and server address as "OCSchat service account uri"
6. C:\Documents and Settings\All Users\Application Data\Microsoft\Crypto\RSA\MachineKeys re-add the Lookup Service account and the Channel Service account and give them full control to this folder and re-apply full control to all the files in it and try to restart the services.
In their case, Step 6 resolved the Group Chat Admin Console error above.
For more on what is OCS R2 Group Chat Server read my other post here.
Here are some questions I received from a school district in Hawaii. Aloha! and a university in Ft. Collins, Colorado:
Is a Director server required for R2?
A Director is not required but recommended – for most school sizes I have seen they typically don’t deploy a Director.
What does ISA 2006 reverse proxy publish with R2?:
What Communicator clients can coexist with R2?
LCS 2005 SP1 client – can connect if the users is in an R2 pool
R1 (RTM) Communicator client can connect if the users is an R2 pool.
Communicator R2 client can connect to an R1 pool.
With a consolidated Edge in R2 do I need multiple NICs?
You don’t need multiple nics for each Edge role in a consolidated scenario but it is recommended to have a NIC for external traffic and one NIC for internal traffic. The only requirement is a unique IP and unique A record for each role. The public facing IP requirement for A/V Edge has been removed as well when using consolidated Edge.
What is the largest size an R2 response group can be?
A response group is targeted at small departmental ACDs around the size of 20 to 25 users. Most schools I talk to this is their average size.
For larger needs, you may look at a partner like Aspect which provides Contact Center like functionality with OCS.
Can Communicator Mobile use Wifi or data only similar to SkyPe, etc?
I know CoMo 3.0 combines data and voice minutes typically when using Single Number Reach. In Europe, the OCS call back feature should not use any voice minutes after OCS calls you back. As far as Wifi or Data only, this will work for IM only conversations but not for the Single Number Reach as OCS dials you back based on your cell number.
Does Group Chat console work on a Mac?
At this time it only works on a PC.
How large can I scale my conferences in R2?
I haven’t seen new performance data on this but I would imagine, since it is 64-bit now, it would be more than the 250 user soft cap in RTM. We will post once more R2 performance data becomes available.
What new R2 UC devices are there?
There are many new UC devices coming along with the R2 launch. There is a new Tanjay device, a new Catalina device, new headsets (some starting at $44), a new candybar USB phone, a new IP conference phone and a new conference usb speakerphone. Additionally, there are new HD cameras (720p @ 30 frames a sec). I have also heard rumors there will be a VGA Roundtable.
If you missed the R2 launch event you can still watch it here.
In my next blog post, I will talk about migration strategies to R2.
I thought this was pretty amazing to watch with some very simple items you can have an amazing smartboard for any surface.
I love gadgets so I am taking this project on. I figured it may be an interesting way to kick off a demo of OCS/Exchange with schools when I am onsite. :)
Here is a great link on how to set this up for your school here.
They are selling whiteboard IR pens here already and if you want to make your own IR pen go here.
Some cool free whiteboard software for download here.
Each AD site where Exchange 2007 exist needs a GC that is at least Windows 2003 Sp1 or later.
The following applies to domain controllers:
RODC (Read-only domain controllers)
No version of Microsoft Exchange uses read-only domain controllers or read-only global catalog servers. However, Microsoft Exchange works in environments that include read-only domain controllers or read-only global catalog servers, as long as writeable domain controllers are available. In these environments, Exchange 2007 effectively ignores read-only domain controllers and read-only global catalog servers.
Domain functional level
You should use at least Windows 2000 Server native for all domains in the Active Directory forest where you will install Exchange 2007 or that will host Exchange 2007 recipients.
Forest functional level
If you plan to use any of the following advanced features, the forest functional level must be Windows Server 2003 in each forest that contains Exchange servers:
A disjoint namespace is the scenario in which the primary Domain Name System (DNS) suffix of a computer does not match the suffix of the domain name where that computer resides. Limited tests were performed to validate Exchange 2007 on a computer that has a disjoint DNS namespace. These tests showed that any issues resulting from this configuration may be resolved by ensuring that the DNS suffix search list on an Exchange server references all DNS namespaces that are deployed within the organization. The list of namespaces should include not only Active Directory and Exchange servers, but also the namespaces for other servers with which Exchange may interoperate, such as monitoring servers or servers for third-party applications. For detailed information about supported scenarios with disjoint namespaces, see Understanding Disjoint Namespace Scenarios with Exchange 2007.
Single-label DNS names
Single-label DNS names are not recommended for use with Exchange 2007 or Exchange 2007 SP1. For additional information about single-label DNS names, see Knowledge Base article 300684, Information about configuring Windows for domains with single-label DNS names.
AD Ratios to Mailbox Servers
Customer today asked me about Outlook 2007 clients. They are planning on doing there schema update for Exchange 2007 and curious what this does to the Outlook 2007 clients. The 'service-connection-point' class is defined in the schema and the SCP objects published in AD contain information that various applications can use to direct clients to bind to a particular service. Exchange 2007 makes use of SCP's to advertise autodiscover service information specifically.
Clients, such as Outlook 2007, will search against a GC to locate SCP's in the forest by querying AD for objectclass=serviceconnectionpoint. In E2K7 during the setup of the Client Access Server role autodiscover will create an SCP object (obviously with an objectclass of serviceconnectionpoint) under the container:
By default the serviceBindingInformation attribute of this object will be updated during setup with the autodiscover service url:
The client will follow this url to determine configuration information for some mobile devices for example, or to provide access to the OAB.
This url can be changed using the set-clientaccessserver cmdlet with the parameter -AutodiscoverServiceInternalURI.
Figure 1 The Autodiscover service process for internal access
For external access, the client locates the Autodiscover service on the Internet by using the primary SMTP domain address from the user's e-mail address.
Depending on whether you have configured the Autodiscover service on a separate site, the Autodiscover service URL will be either https://<smtp-address-domain>/autodiscover/autodiscover.xml or https://autodiscover.<smtp-address-domain>/autodiscover/autodiscover.xml. Figure 2 illustrates a simple topology with a client connecting from the Internet.
Figure 2 The Autodiscover service process for external access