Ramblings of an Exchange-Lync PFE

Upgrading your MCITP to 2013 (MCSE)

2013-04-19T15:23:00Z

When I started thinking about my career goals for 2013 one of them is to upgrade my MCITP: Enterprise Messaging (2010) to MCSE: Messaging (2013). When I started to research the exams that I need to complete I saw a requirement that wasn’t there for previous MCITP certifications. In order to be certified as MCSE: Messaging or Communications you must have your MCSA for 2012. The MCSA 2012 involves passing 3 tests (70-410, 70-411, and 70-412). When I saw this my first thought was….ehhhhh! Then I read further down and I saw that I could expand “How to upgrade to the MCSE: Messaging certification (see below). After clicking on “Upgrading your skills to MCSE Windows Server 2012” I found that if you have your MCITP you can gain your MCSA 2012 with a single test (70-417). See the image below that shows the requirements for taking the 70-417 test.

After you have achieved your MCSA 2012 you can then take 2 tests and get your MCSE. The tests for both Messaging and Communication (Lync) have hyperlinks below so you can review what is involved in gaining these certifications. I have completed my MCSA 2012 and my first Exchange 2013 test (70-341) and I will tell you it seems the difficulty level of Microsoft certifications is definitely going up. In my opinion that is a good thing. If anyone is interested (comments here or send me an email) I can do a write-up on each certification test. Good Luck!!!

MCSE Messaging

MCSE Communications

Exchange Team Blog: Exchange 2013 RTM CU1 Status

2013-03-25T19:26:48Z

The Exchange Product Team has posted a blog about the details of when CU1 will be released. Check it out (http://blogs.technet.com/b/exchange/archive/2013/03/25/exchange-2013-rtm-cu1-status.aspx)

Question the PFE: March 2013 Edition

2013-03-24T22:59:00Z

**Update - Per the Exchange Team Blog (http://blogs.technet.com/b/exchange/) Exchange 2013 RTM CU1 is currently planned for an April 2nd release.

Question 1:

I want to deploy Exchange 2013. Can I do this now that 2010 SP3 is out?

PFE Answer1:

Short answer is “Not if you want to be supported”. Although Exchange 2010 SP3 is prerequisite in the existing environment Exchange 2013 CU1 is still required to be supported. This should be released by the end of 2013 Q1. If you are deploying Exchange 2013 in a Greenfield (No existing Exchange products) environment then you don’t have to wait for CU1.

Question 2:

How do I take advantage of High Resolution photos with Lync 2013?

First of all let’s take care of the prerequisites. You must have an Exchange 2013 mailbox and also Lync Server 2013. Once your users are on both of these versions the process is fairly simple. You basically just upload the picture via OWA or script (below). This will upload the picture to the Root of the users mailbox (648x648 pixels) and also to AD (48x48 pixels). Once you do this we can verify by either opening Lync 2013 client, Options, My Picture or go to the EWS url (https://fabcm1.fabrikam.com/ews/Exchange.asmx/s/GetUserPhoto?email=rschwendiman@fabrikam.com&size=HR648x648).

Upload Photo

$photo = ([Byte]] $(Get-Content -Path "C:\Photos\RSchwendiman.jpg" -Encoding Byte -ReadCount 0))

Set-UserPhoto -Identity "Richard Schwendiman" -PictureData $photo -Confirm:False

Set-UserPhoto -Identity "Richard Schwendiman" -Save -Confirm:False

http://technet.microsoft.com/en-us/library/jj688150.aspx

***Please feel free to submit questions about anything Lync or Exchange related.***

Exchange 2013 Mail Flow Demystified…Hopefully!

2013-03-13T16:07:00Z

After installing Exchange 2013 for the first time I noticed some new changes in the architecture that made it confusing in regards to mail flow. One of the bigger changes in Exchange 2013 is the changes of Exchange server roles. In Exchange 2013 we now have just Mailbox and the CAS server role. There is no longer a standalone hub server role. The transport components that where traditionally on the hub role are now spread out on both the Exchange 2013 mailbox and CAS server roles. Let us first go ahead and talk about different transport services in Exchange 2013.

FrontEnd Transport service - This service runs on all Client Access servers and acts as a stateless proxy for all inbound and outbound external SMTP traffic for the Exchange 2013 organization. The Front End Transport service doesn't inspect message content, only communicates with the Transport service on a Mailbox server, and doesn't queue any messages locally.
Transport service - This service runs on all Mailbox servers and is virtually identical to the Hub Transport server role in previous versions of Exchange. The Transport service handles all SMTP mail flow for the organization, performs message categorization, and performs message content inspection. Unlike previous versions of Exchange, the Transport service never communicates directly with mailbox databases. That task is now handled by the Mailbox Transport service. The Transport service routes messages between the Mailbox Transport service, the Transport service, and the Front End Transport service.
Mailbox Transport - This service runs on all Mailbox servers and consists of two separate services: the Mailbox Transport Submission service and Mailbox Transport Delivery service. The Mailbox Transport Delivery service receives SMTP messages from the Transport service on the local Mailbox server or on other Mailbox servers, and connects to the local mailbox database using an Exchange remote procedure call (RPC) to deliver the message. The Mailbox Transport Submission service connects to the local mailbox database using RPC to retrieve messages, and submits the messages over SMTP to the Transport service on the local Mailbox server, or on other Mailbox servers. The Mailbox Transport Submission service has access to the same routing topology information as the Transport service. Like the Front End Transport service, the Mailbox Transport service also doesn't queue any messages locally.

Next let’s talk about each of the ports mentioned above. Then finally we will discuss the flow of mail through this pipeline.

Port 25 – This port just like in previous versions of Exchange is used for SMTP. Used by both External SMTP into the Front End Transport Service (FET), SMTP with Exchange 2007\2010 hub servers, between MBX servers, and also from the FET to the Transport Service. There is a receive connector named Default Frontend <servername> that listens on this port.
Port 587 – This port just like previous versions of Exchange is used for Client Connections (POP\IMAP). The CAS Server has a receive connection listening on this port name Client Frontend <servername>.
Port 717 – used for outbound proxy connections from the Transport service to the FET Service. When you create a Send connection you have the option to send mail destined for the Internet directly from the Transport Service to the Internet\Smart Host or relay that mail through the Front End Transport Service. There is a receive connector named Outbound Proxy Frontend <servername> that listens on this port.
Port 465 – used to accept proxied connections that were received on port 587 by the FET service for client connections. There is a receive connector named Client Proxy <servername> that listens on this port.
Port 475 – the Mailbox Transport Delivery Service listens on this ports for connections either from the transport service SMTP Send connector or SMTP from the Transport Service on other Mailbox Servers that need to send mail to users on this server.
Port 2525 – if the CAS and MBX servers are collocated on the same server the SMTP Receive connection for the Transport service will listen on 2525 instead of 25. This is because two services (FET and Transport Service) can’t listen on the same port.

Here is a diagram that I put together to help me understand these 3 services and also what each of these ports are used for. Next we will discuss now mail flows through these services.

Now let’s talk about Exchange 2013 Mail Flow.

Sending Mail to both Internal\External recipients

Step 1: Mailbox Transport - This process starts with the user typing a message in Outlook\OWA and clicking the send button. In Exchange 2007\2010 it was the responsibility of the Store Driver in the Transport Service on the HT Server to send the message to the next hop. Now the Store Drive has been relocated to Mailbox Transport and split into two services (Submission and Delivery Services). The Mailbox Transport Submission service will pick up the message from the users Outbox, Run the Hub Selector process (in order to select the best Transport service which could be local or another server), and forward the message to the Default Receive connector in the Transport service (SMTP 25 or 2525).

Step 2: Transport Service - After the message has been accepted by the Transport service it will get put into the submissions queue. The submission queue will process the message and hand it off to the Categorizer which will do recipient resolution (expansion and bifurcation) and routing resolution. Next it will be place into the correct delivery queues. If the message is going to an external recipient it will use the correct send connector and either send directly to internet or proxy through the FET Service (Set-SendConnector <name> -FrontEndProxyEnabled $true). If the message targets an internal user the message will be send from the Transport Service to the Mailbox Transport Delivery service on the destination mailbox server. Once the Mailbox Transport Delivery service receives this message it will use local RPC to place the message in the users Inbox.

Receiving mail happens in the reverse order of Sending mail. Note that bifurcation always happens at the Transport Service level on the Mailbox Server. After bifurcation the message is sent via SMTP directly to the Mailbox Transport Delivery service over port 475 on the Mailbox Server where the mailbox of the recipient is currently mounted.

I hope this has been helpful in understanding how mail flow now happens in Exchange 2013. If you have any questions please feel free to contact me to discuss further.

Exchange 2013 Logging and Space Requirements

2013-02-21T12:10:00Z

There are a lot of changes that were incorporated into Exchange 2013 both from an Architecture and Operational standpoint. One of these changes is something that can be overlooked very easily so I wanted to point out the reasoning behind this change and what to expect. If you look at the disk space requirements for Exchange 2013 (see below) you will notice we now require 30GB free on the Exchange Install drive. This is a drastic increase from 2010 which only required 1.2GB (which I think was a little on the small side). Although 30GB nowadays isn’t much if you don’t plan for this space increase you could get caught off guard and possibly run into space issues very quickly.

So why do we require all of this extra space? It’s because we now have logging of various Exchange components turned on by default. If you open the logging folder (<Install Drive>\Microsoft\Exchange Server\V15\Logging) on a Mailbox Server check out the amount of folders compared to Exchange 2010.

In all of these folders you will find the corresponding logs for each component. Also if you look in <Exchange Install Drive>\Microsoft\Exchange Server\V15\Logging\Diagnostics\DailyPerformanceLogs you will find Perfmon files (see below) will all of the counters pertaining to Exchange and the Server role (think ExPerfWiz).

Although these log files and Perf data will take up a decent amount of disk space the reasoning behind turning these things on be default was to help customers. If you have had to call into support to troubleshoot an issue you could almost expect the same response every time. After gathering more info on your problem the Engineer would say let’s turn on logging\performance monitor and reproduce the problem. For various reasons customers might not have been able to do this right away. So the Engineer would give the customer an action plan and the customer would call back when they gathered the data. This could be very time consuming and delay getting the problem resolved.

After seeing what kinds of delays this caused the Exchange team decided to collect these logs and performance data by default so it can be used for historical troubleshooting. Instead of having to turn logging on and repro the problem these logs already contain the data we might need. I think this is a great improvement for Exchange 2013 and a welcomed change. I bet most companies are willing to sacrifice a little disk space for hours of support this will save in troubleshooting issues.

Exchange 2010 SP3, SP2 RU6, and Exchange 2007 SP3 RU10 released!

2013-02-13T21:14:00Z

The much anticipated Exchange 2010 SP3 has been released along with SP2 RU6 and 2007 SP3 RU10. There are links below to the actual KB articles below. Note the excerpt below about coexistence between Exchange 2010 SP2 and Exchange 2013.

NOTE: Exchange 2010 SP3 allows Exchange 2010 servers to coexist with Exchange 2013 CU1, which is also scheduled to be released in Q1 2013. Customers can test and validate this update in a representative lab environment prior to rolling out in their production environments as an important coexistence preparatory step before introducing Exchange Server 2013 CU1.

Exchange 2010 SP3

Features: Coexistence with Exchange 2013, Support for Server 2012, and IE10

Documentation: http://support.microsoft.com/kb/2808208

Download: http://www.microsoft.com/en-us/download/details.aspx?id=36768

Exchange 2010 SP2 RU6

Documentation: http://support.microsoft.com/kb/2746164

Download: http://www.microsoft.com/en-us/download/details.aspx?id=36716

Exchange 2007 SP3 RU10

Documentation: http://support.microsoft.com/kb/2788321

Download: http://www.microsoft.com/en-us/download/details.aspx?id=36708

Update iOS 6.1 Causing problems with Exchange (Resolution iOS 6.1.2)

2013-02-12T17:46:00Z

**Update** - Apple has released iOS 6.1.2 which apparently resolves this Exchange Calendar Issue (http://support.apple.com/kb/DL1639).

In case you haven’t read any blogs or forums about this issue I wanted to communication our KB article that discusses the problem. Basically we are seeing problems around devices that have been upgraded to iOS 6.1 causing transaction log growth, CPU usage, and Memory consumption problems on CAS & MBX servers.

If you notice excessive IIS log growth on the CAS server you should look for the text below which keeps looping continuously.

2013-02-06 23:41:59 192.168.137.11 POST /Microsoft-Server-ActiveSync/default.eas
User=user1@contoso.com&DeviceId=ApplDMPFXZABDFHW&DeviceType=iPad&Cmd=Sync&Log=[truncated]
443 user1@exchlab.com 192.168.137.246 Apple-iPad2C1/1002.141 200 0 0 187

2013-02-06 23:41:59 192.168.137.11 POST
/Microsoft-Server-ActiveSync/default.eas User=user1@contoso.com&

DeviceId=ApplDMPFXZABDFHW&DeviceType=iPad&Cmd=MeetingResponse&Log=[...Error:WrongObjectTypeException…]

443 user1@exchlab.com 192.168.137.246 Apple-iPad2C1/1002.141 200 00 140

Also if you have enabled EAS Mailbox logging you might see the text below (Status 4) in some meeting requests that suggest the problem is happening.

Workaround

The workaround can include recreating the device partnership, custom throttling policy, ABQ policy, and possible other Load Balancer rules. Take a look are the link below which gives all the details and also different ways to collect logs.

http://support.microsoft.com/kb/2814847

KBs: Exchange 2010 SP2 RU4 Blackberry\RIM Issues

2013-02-08T14:13:00Z

These are some BB\RIM existing issues that were caused by SP2 RU4. Below are KB articles written with details around the issues and resolutions\workarounds.

Exchange 2010 SP2 RU4 Blackberry\RIM Issues

KB:2797529
Email message delivery is delayed on a Blackberry mobile device after you install Update Rollup 4 for Exchange Server 2010 SP2

KB:2787500
Declined meeting request is added back to your calendar after a delegate opens the request by using Outlook 2010

KB:2779387
Duplicated email messages are displayed in the Sent Items folder in a EWS-based application that accesses an Exchange Server 2010 Mailbox server

KB:2800080
ErrorServerBusy response code when you synchronize an EWS-based application to a mailbox in an Exchange Server 2010 environment

Question the PFE: January 2013 Edition

2013-01-31T20:13:00Z

I did get a couple of questions from users this month which hopefully I can answer. Again please feel free to contact me through “Email Blog Author” with any Exchange\Lync questions you may have. I will post these questions at the end of every month.

Question 1:

What steps should I be taking to further my career working with Exchange? I am currently an Exchange Administrator but would like to take my career to the next level (Exchange Engineer).

PFE Answer 1:

Most of us go through this level sometime throughout our careers and hopefully I can give you some tips. As an Administrator most of your responsibilities live at the User\Server level. This can include things like creating mailboxes, tracking messages, Exchange server updates, scripting, etc… To get to the next level (Engineer) you really need to focus on troubleshooting and core infrastructure.

If you have Exchange engineers above you in your current job ask questions. It is important to build a network of people that you can bounce questions and ideas off. The key here is to ask intelligent questions. Don’t go the Engineer with questions that you could have easily went to Technet or Binged the answer. If you keep doing this they are eventually going to be reluctant to discuss technical issues with you. Along with this you really need to gain deep technical knowledge of your product and in this case Exchange. This means reading things like blogs, Exchange help file, TechNet articles, and any other training you can get your hands on. But I think the single most important step in expanding this knowledge beyond the book is to build a lab.

Usually this will include some type of workstation\server with a lot of memory and disk that you can virtualize. Once you have the hardware throw Hyper-V on there and build an Exchange environment. I don’t mean just 1 or 2 Exchange servers but build out a production environment that a normal mid-sized company might deploy. This should include things like DAGs and Multiple Sites with all the different roles deployed. This way you can get the full experience. After you have the infrastructure in place deploy Exchange technologies (archiving, database copies, delegation, etc..). Run through different scenarios and troubleshoot any problems you might run into. Don’t just use Bing as a first resort to troubleshooting really try to understand what is going on and where to look for different problems. Even if you have to eventually get the answer from the internet don’t just make the change and move on but again try to understand why this fixed the problem.

Hopefully this helps and if you have any questions please feel free to ask. Also here are some links to help get started.

http://blogs.technet.com/b/exchange/

http://technet.microsoft.com/en-us/library/bb124558.aspx

http://exchangeserverpro.com/

Question 2:

Can we migration from OCS 2k7R2\Lync 2010 to Lync 2013? If so what are the steps involved.

PFE Answer 2:

The simple answer is Lync 2013 doesn’t support tri-existence. You can migrate from OCS 2k7R2 to Lync 2013 OR Lync 2010 to Lync2013. If you already started to migration to Lync 2010 you need to complete this before you should introduce 2013 into your environment. Here is a link to the Technet documentation about supported migration scenarios (http://technet.microsoft.com/en-us/library/gg425764.aspx).

Question 3:

Why did Microsoft consolidate the roles in Exchange 2013 down to the CAS and MBX?

PFE Answer 3:

This was largely based on hardware specs now compared to when Exchange 2k7 and 2k10 were released. It was found that with newer hardware platforms resources were being underutilized. With the role consolidation in Exchange 2013 we will be able to take full advantage of our resources.

Lync 2013 Persistent Chat (Lab Environment)

2013-01-29T15:13:00Z

I started to extend my lab to include Lync 2013 last week and ran into a problem. After getting a Standard Edition pool installed everything was working great. Then I decided I wanted to do some testing on a new Persistent Chat pool since some of my customers are interested in this updated service. Here is where I ran into a very simple problem that took me awhile to troubleshoot.

So I configured the Persistent Chat pool in Topology Builder and published the topology. Since this this a lab I am going to collocate the PChat pool on the SE Server (not supported but this is a lab environment). I also collocated the PChat database on the Default SQL instance I had on the SE server. When I went to the Deployment Wizard, Install\Update Server System, Setup\Remove Lync Server Components and clicked Run. Everything was installing properly and then when it went to Install-CSDatabase it failed. There error said something about database file locations and please check that the SQL instance is online. I don’t have a screenshot because I closed the web window and didn’t save the
file :(

At this point I checked that all the SQL components were running and I was able to connect. Then I verified all services were healthy on the SE server. Checked all the log files and just saw the same vague message as above. I also check disk space which I had about 14GB free which is enough right? Wrong. I did some research and it turns out that in order to install this DB there needed to be at least 16-18gb free on the DB drive. The crazy thing is the install of the DB took up very minimal space. I increased the DB drive by 20GB and the install went through no problem. There isn’t much documentation on this besides the help file stating the hardware requirements for Front End Servers, Back End Servers, Standard Edition Servers, and Persistent Chat Servers need 72GB of space.

Hopefully this saves someone the 2hrs I spent messing around with this issue.

FYI...I will be doing a post on Lync 2013 Persistent Chat in the near future.