Rerun the Exchange Schema, Forest and Domain Prep and it will re install the arbitration mailboxes.  In Exchange 2010 the arbitration mailboxes do a lot of the backend processing.

etcMWI takes advantage of the Exchange impersonation capability to monitor a user’s mailbox for new unread voicemails.  This impersonation has to be applied to all Exchange 2007 mailbox and client access servers and mailbox databases.  So when you add any of these servers or databases to your existing Exchange organization you must apply the impersonation to these new resources.

This applies to all versions of etcMWI and Exchange.  Instructions on how to accomplish this can be found in the etcMWI Installation Guide provided to you.  It is detailed early on in the document and an easy to use Powershell script is provided in the Appendix.  If you need help, contact Enabling Technologies Tech Support

1. Stand up E2010 core
2. Stand up E2010 UM
3. Configure E2010 UM dial plans & policies
4. Link the dial plans between the 2010 UM servers and 2007 UM servers
5. Reconfigure VoIP gateways to point to 2010 UM servers

•  If a UM 2010 server finds itself handling a call for a user whose mailbox is Exchange Server 2007, it will immediately try to hand off the call to a UM 2007 server in the same Dial Plan. Therefore, client must retain at least one UM 2007 server in any Dial Plan in which users have Exchange Server 2007 mailboxes.

1. Reconfigure/test 2010’s native MWI
2. Migrate all mailboxes to 2010
3. Shut down 2007 UM and other roles

The prerequisites to installing an R1 CWA server in an R2 SIP environment you first need to create a UC certificate with the necessary subject name and subject alternate names. Then create a 32 bit Windows Server with Service Pack 2 or above and of course the Office Communication Server 2007 code, and a BES server 4.7 and above. 　　　To integrate Blackberry Enablement Services (BES) with Office Communication Server 2007 R2 you first need to do a Forest Prep, Domain Prep with the R1 code. In this prep you are preparing the SIP domain to except the install of an Office Communication server 2007 R1 Role into an Office Communication server 2007 R2 domain environment. You then need a 32 bit Windows Server installed with service pack 2 and above, then using the R1 code you can click setup, then Other Roles, then Communicator Web Access role and install that role on the 32 bit Windows Server.

On the BES server it first must be installed with the 4.7 and above BES code and you must then install the Collaboration Services on the BES server and make sure the services are started.　Through the BES server menu you navigate to the Collaboration Service and in the server or URL text space type in the subject name or one of the subject alternate names in the certificate along with a DNS "A" record setting for the name used.

The reason for this article is because if the R1 forest and domain prep is not done you cannot in install the R1 CWA role. You will get an error that the forest prep is reading "partial" and the role will not install.

It should be very clear that the costs increase with the increase in number of communication modes that a company enables for their employees. In addition to the costs associated with the use of different modes of communications, you should be aware that the costs of management, monitoring and compliance also increase exponentially. 

Obama BlackBerry

The 44th president of the United States, President Barack Obama hoped that he could continue to use  his BlackBerry after his inauguration. The security of his communications through the BlackBerry became a major concern to the U.S. Secret Service. On January 21, 2009, it was reported that the President would be allowed to use his favorite toy, the BlackBerry.

 Let's look at how Mr. Obama communicates:

* He has been using the same e-mail address for many years

* He also uses a laptop and Twitter

* He receives briefing books and memos on his BlackBerry

Even our current president can't live without his BlackBerry. It gives him the capability to communicate and collaborate from anywhere, anytime, and with whomever he pleases, including you and me. It's really become the norm in our society for folks to always carry a BlackBerry, a smart phone, a PDA, or other such device with them.

If a company provides its employees with five modes of communications, there are five ways for someone to break in and steal the company's most critical data. Each mode of communication is a potential security hazard. How often have you heard about a bank employee's laptop being stolen with clients' secure financial data, making those clients vulnerable to someone doing something detrimental with the confidential information? Additionally, in light of major government bailouts, taxpayers are demanding that financial and healthcare providers be extremely transparent. The federal, state, and local governments, as well as the FDA, SEC, HIPPA, and Sarbanes-Oxley, all require companies to be compliant and archive their communications data for three to ten years.

Increased efficiency is the first of many UC advantages.

A hospital in California was looking for a solution to improve its capability to communicate and collaborate more efficiently, especially in the radiology department. The radiology department recently had expanded to include radiologists from India and Australia to provide round-the-clock coverage. When patients were x-rayed in California, their x-rays were loaded in a Kodak system that allowed authorized colleagues access to them over the Internet. So a patient's x-rays taken at 5 p.m. in California by the technician and then loaded on the Kodak system were available to the radiologist in India or Australia after the California radiologist  went home. Now the radiologist in India or Australia could analyze the x-rays and enter the diagnosis into the system. In the morning, when the California radiologist and technicians returned to work, those x-rays were viewed and the diagnosis was entered.

If the radiologist or the physician in California had questions or wanted to consult with the radiologist in India or Australia, or with an expert, he or she could participate in a Web conference (Microsoft Office Live Meeting or Cisco Webex, for example), open and view the x-rays online, consult with the other participants, annotate a specific area on the charts by using annotation tools, and reach a well-informed conclusion. Just for an example, 4-6pm in California would mean 9-11am in Australia (next day) and 5:30-7:30am in India (next day).  UC is a brilliant innovation that improves efficiency in communications and collaboration.

When a user connects to an OCS conference call using the web client, they can also see the instant messages that are typed in the conference call.

I recommend starting a back channel conversation in a different window rather than assume only OCS federated users can see the IM traffic.

 
While this can be very convenient, it can also lead to some big problems particularly with some of the Office Communications Server roles.  It will likely lead to all manner of problems and it’s not easy to find the cause!

The problems show up on the Office Communications Server roles that use more than one IP address: the Mediation Server and the Edge Server.  The Best Practice on the Mediation server is to install two network interface boards with individual IP addresses on each one.  It also specifies that these addresses should be on separate subnets.  For the Edge, the Best Practice is to install 4 network interface cards with individual IPs.  OK, before you start letting me have it, this is Best Practice and I’m not getting into a discussion about straying from Best Practice regarding network cards and IP addresses etc. on these boxes.  The bottom line is when using more than 1 IP address on either of these roles the DNS autoregistration will register all the IP addresses.  You will end up with several (1 for each IP on the box)  “A” records in the DNS zone with the same hostname but different IP addresses.   

As an analogy, think of this as being a dynamic rather than static DNS address.  Would you ever give a server device a dynamic TCP/IP address?  Not likely!  I use the same reasoning when I am installing OCS servers.   I always disable the autoregistration on the Mediation and Edge servers and usually do so on the other OCS roles as well.  I like my servers to have static IP addresses and static DNS entries! 
Remember; once you disable the autoregistration you must manually create the “A” record for the IP or IPs.  Typically, I will create the “A” record for the IP representing the network interface card facing the OCS Pool on the Mediation Server and not create a record for the gateway facing interface.  For the Edge server, I create “A” records for the network cards/IP addresses facing the public side within the external DNS zone and create an “A” record for the internal interface in the appropriate internal zone.  Again, this is just an example, I’m not saying its right or wrong for every install of Mediation or Edge servers.  Let’s keep our eye on the prize!

The moral of the story, and some of my Best Practices:

1. Don’t let Office Communications Server hardware autoregister themselves to DNS.  Sidebar: Don’t let servers in general autoregister themselves to DNS, especially those with multiple IP addresses.

2. All Office Communications Servers have “A” records manually/statically created for them within DNS.  Do not create “A” records with identical host names and different IP addresses in the same DNS zone.

3. At several points during an installation, double check the Office Communications Server DNS records.

4. If needed, the “hosts”  file on a machine is your friend!

As the Unified Communication Solutions, IM/Presence Partner of the Year, I was overwhelmed with the amount of recognition we received for our accomplishment from Microsoft and our peers. We believe it will translate into future opportunities and growth in our business throughout the year. We’ve already experienced a tremendous surge in people visiting our website to learn more about our company since the announcement was made public about us winning the award. Getting back to the WPC event. Not only were we recognized on stage during a keynote presentation by Allison Watson, we were celebrated during an Award Winner Executive Luncheon hosted by Allison and Steve Ballmer. Suffice to say, I was very proud of what our employees enabled us to achieve through their dedication and efforts. The award brought us recognition that no amount of advertising could buy. Of course, the opportunity to network with Microsoft associates, SI partners, and vendors each evening at parties and dinners in New Orleans was very exciting and gained us maximum exposure throughout the Microsoft ecosystem. Coming away from WPC, I am more energized and dedicated than ever to make our company a repeat winner at WPC 2010 in Washington, DC.

If it hasn’t happened yet, it will! One day something bad will happen in Active Directory and damage will be done to a User or an entire Organizational Unit, Domain or Forest. After you’ve realized the scope of the problem and taken a few breaths you will certainly be headed down the road to recovering the user, Organizational Unit, Domain, etc. However, Office Communications Front End servers(this article applies to LCS 2005 and OCS 2007 equally well! ) will still be up and running! Naturally your user(s) who no longer exist in Active Directory won’t be doing much Instant Messaging but the Front End services will still be running. Of course, with no users who can log on, the Front Ends won’t be doing anything…..or will they?

User Replicator

Actually they will! By default, your Front Ends (in an Enterprise Pool it will be a specific Front End) will periodically(every 2 hours by default) trigger the User Replicator process. What is User Replicator? Its responsibility is to ensure that Active Directory and the RTC(Real Time Communications aka OCS) databases are synchronized. Any time an OCS user is created or modified in Active Directory it is the User Replicator’s job to make sure the changes are pushed into thedatabase. OK, so we have a User Replicator that can’t add or modify user data to the database ….So what? My Active Directory is messed up! That’s the least of my problems!...

The bad news is that it can still delete user data from the RTC database! When User Replicator runs, it will notice any and all users in the RTC database who are not in Active Directory. It will take the data for these users in the database and delete it! That includes any and all of their contacts and block/allow lists! Not good, but User Replicator is doing exactly what it is supposed to do! It doesn’t know the reason that it can’t find the user in Active Directory, nor does it really care! So now you’ve had some sort of Active Directory catastrophe and once the User Replicator runs your user’s contacts and block/allows have been wiped out….Now What?

The answer to this is pretty simple and covered in the Office Communications Server Administration and Backup and Restore Guides. Recovering this data is neither rocket science nor a big deal if you’ve done your backups and you are the DBA and/or can reach the DBA! However, why not prevent the problem in the first place? All you have to do is shut down the Office Communications Server services at the onset of the Active Directory catastrophe! Most folks in this situation will shutdown Exchange’s services, just add shutting down the OCS services to your disaster protocol and you will save yourself some time while recovering your systems! One more Best Practice suggestion…..

DB Import/Export - DBIMPEXP

Take advantage of DBIMPEXP (you can find it on the installation media or via the Internet-use the version specific to your servers; i.e. use the LCS dbimpexp if you have and LCS server). It will save your bacon some day! DBIMPEXP is a tool used to backup/export and restore/import user information from the RTC database usually via an XML file. Predominantly this information is the user’s contacts and block/allow list. Anyone ever have a user say, “I accidentally deleted one of my contacts and for the life of me, I can’t remember his IM address! Any chance you can help me out?”

Solution A: Just restore/import the RTC and RTCConfig databases….NOT! This will overwrite ALL the contacts for ALL your users! The guy with the missing contact will be real happy but the rest of your users…..

Solution B: Figure out a way to run a DBIMPEXP backup/export on a regular basis (Ideally, the Office Communications Server services should be stopped when running this backup/export but not required).

The best place to run this is from the Front End Server itself because the ID you are using to run the Front End Server probably has the appropriate permissions to do both a backup/export, restore/import using DBIMPEXP. Of course, periodically test the backup/export and restore/import!

Once you have this backup/export you can either restore/import the info for all users (do this with the services shut down!) or restore/import the info for a single user! You cannot directly restore/import the info for a specific contact for a specific user but it’s easy enough to look at the XML file created by DBIMPEXP and lookup the contact for the user and provide him with the appropriate information. Realistically, it won’t take you very long to configure a daily run of DBIMPEXP to backup/export everything and overwrite the backup file. If you want to get fancy and rig it to store the information to a different file name every day or something more exotic, go for it!

The Bottom Line:

· Make sure you have a good, current SQL backup of your RTC database.

· Run DBIMPEXP on a regular basis. Backup/export your user data to someplace you can get to it!

· When you have an Active Directory catastrophe, shut down Office Communications Serverservices as quickly as practical!