Hyper-V-active Exchange

As most folks have already seen, Hyper-V - the newest Microsoft Virtual Server Solution - released to market yesterday.  This is a pretty big step forward for MSFT, adding a datacenter-ready virtualization solution to the Server 2008 platform.

I'm doing a little research now on the Exchange front with Hyper-V, but I was able to get an un-named source (ooooh spooooky) from MSFT to confirm that some MSFT Exchange 2007 servers at their facilities are running on Hyper-V, so support on that platform will most likely be forthcoming.  More info on this front as I can get my hands on it.

From the limited contact I've had with the betas, I can safely say that this product is a different animal than Virtual Server 2005.  The overall performance in both the Host and Guest components has ratcheted up by a factor of multiples, The management interface is also re-designed to allow for more fine-tuned control over the vitalized systems, and cluster support means the ability to move VM's between hosts is built-in.

We're still in the very early stages of this technology, so expect more reports as I run into it more and more in the field.  Within a short period of time, I'm also going to get to put Exchange 2007 through its paces on Hyper-V, and I'll be happy to report on that too!

It goes without saying that Double-Take is going to be supporting Hyper-V in the very near future with some amazing tool sets.  If course we can already replicate and fail over Exchange from Physical or Virtual (of any flavor) into a Hyper-V VM, but keep your eyes peeled to find out what new stuff we're working on over here especially for the Hyper-V platform.

Legally important

Providing Disaster Recovery (DR) for Exchange servers has always had strong arguments in its favor.  As an overall requirement, being able to get the email systems, calendars and contacts back up and running ranks pretty high up the list.  But there are other reasons to look at Dynamic Infrastructure solutions for Exchange that go beyond the convenience of Exchange end-users.

When email is stored only on the production Exchange server, it can be altered or destroyed by anyone with access to that server, which means anyone with an email account that allows them to see that particular user's mailbox.  This leaves you with a whopping legal liability (check local listings on exactly what that is for you), but one that can be avoided in most cases.

Using an Operational Recovery tool, like Double-Take TimeData (see disclaimer below), will allow you to ensure that another copy of the data is not only held off-site, but held in a repository that tracks all changes to the email, calendars, contacts and all other information.  This way if a critical change is accidentally applied, or if someone maliciously attacks the data on the production server, you have the ability to revert either individual items or the entire Store or Storage Group as required.

This means that if the data is required for a legal requirement (like court-ordered discovery), you can be sure that not only will the system be available, but also that any information that could be lost without impacting the server can also be quickly and efficiently restored.  Of course, being able to get that information back to some other server or even to a desktop or laptop is best, so look for tools that give you that flexibility.  After all, if you don't happen to have the original server running to perform the restore of information, you'll need to be able to designate someplace else to receive the data instead.

This doesn't change anything you might be doing with DR and Dynamic Infrastructure in your Exchange Environment, but gives you a deeper level of protection and flexibility that most standard DR tool-sets just can't natively offer.

No backup for you!

Of course we all all knew there would be rough spots for early-adopters of Exchange 2007 on Windows Server 2008, but this one is a bit hard to handle.

Based on the information in This Blog Entry, there is no built-in online backup tool for Exchange 2007 on Server 2008.  That's right, none.

The suggested options are to use either 3rd Party VSS-based backup tools or 3rd-Party ESE-based backup tools, or else use Microsoft's Data Protection Manager to handle it. While this solves the problem, it does mean you're forced to throw a lot of money at the system to fix it.

Some may ask, "Why not just use a replication tool?" Good question.  Replication systems like Double-Take effectively protect the data, but do not truncate the log files that have already been committed to the database.  That means that your log files will continue to grow - often at an alarming rate - essentially forever. You could turn on circular logging, but NO ONE RECOMMENDS THIS, including Microsoft.  So you are left with two choices, get a LOT of disk for logs, or buy a 3rd Party backup tool or DPM.

The SBS version of Exchange 2007 on Server 2008 *will* have these tools, so it can be done with the native 2008 backup solution, and Microsoft has confirmed that the API's to do it are in Exchange 2007 SP1, they just didn't build it for the full Server 2008 platforms.

Hopefully, that just means it will be in Server 2008 SP1, and that we won't be expected to shell out more cash to fix Exchange 2007!

Technorati Tags: Exchange 2007,backup,DPM

Come listen to me babble!

For those looking at protecting Exchange 2007 and the other tools that usually surround it (like Blackberry, SharePoint, etc), we're doing a webinar on the morning of June 25th.  I'll be presenting on what the built-in Exchange 2007 tools cover, what they don't, and how to fill in the gaps for total DR protection.

Join us, the link to register can be found here: http://www.doubletake.com/innercircle/resources.aspx

Best of all, it's free!

Eliminate Virtual Migration downtime using Double-Take Virtual Recovery Assistant

If you are like many customers that I have worked with over the years there has always been a need to migrate their Exchange servers to new and better hardware but scheduling the downtime for the migration was always problematic. Now with virtualization technologies, performing the P2V process without downtime has become even more of a requirement. Below Double-Take has announced this capability and a webinar June 24th, on how to accomplish these migrations while eliminating downtime.

Southborough, Mass. -Double-Take^® Software (NASDAQ: DBTK) will showcase its latest innovation, Double-Take^® Virtual Recovery Assistant, in an upcoming webinar on June 24, 2008. Double-Take Virtual Recovery Assistant is a new enhancement to Double-Take that simplifies the process of migrating to VMware^® ESX virtual machines while online, with virtually no downtime.

Webinar participants will learn how to leverage Double-Take Virtual Recovery Assistant for real-time physical-to-virtual (P2V) conversions, reduce end-user downtime associated with traditional P2V migration products, and automate the protection of workloads running on physical servers by moving them to virtual machines automatically in the event of a server outage or site-wide disaster.

The Double-Take Virtual Recovery Assistant webinar will discuss:

• The features of Double-Take Virtual Recovery Assistant and its use for P2V or virtual-to-virtual (V2V) Protection
• The effortless migration from physical systems or other virtualization platforms to virtual machines running on VMware Infrastructure
• Lower total cost of ownership by reducing the number of software licenses required at the disaster recovery site when used for protection of physical systems
• Ways to reduce hardware expenses and improve manageability by virtualizing existing physical production and disaster recovery servers on VMware Infrastructure

Webinar Information: To register for the webinar, click here.

Date: Tuesday, June 24, 2008

Time: 11am — 12pm ET

Nicholas Schoonover, senior solutions architect, Double-Take Software

Nicholas has spent the past fifteen years solving problems for some of the world’s largest financial institutions, internet sites, and small/mid-sized enterprises. He has designed and implemented recoverability strategies for Windows and Unix platforms and over the past eight years has architected hundreds of projects for high availability and disaster recovery using virtualization.

Hotfix info for Exchange 2007

Great link for Exchange Admins, Anderson Patricio recently posted a link to Microsoft's Central Listing of Exchange 2007 Hotfixes.  This not only lets you know what Service Packs and fixes are available, but has link information for the Update Rollups, which will make life a lot easier for any Exchange ranger out there!

There is also a ton of information on how you can approach patching (individual fixes versus rollups, etc) at This Link.

Just remember that the rollups contain ALL previous patches, so if you didn't want to apply a patch, don't apply the rollups that come after it either.

Happy patching.

Trying out Live Writer

The Live team at Microsoft has updated Live Writer (their blogging software) to accept non-Microsoft blogs, like the Blogger system I use.   So far so good, the integration seems pretty slick, and it was able to figure out all my setting just from my URL.  No small feat since the actual site is hosted on a different provider than Blogger!

As with everything else, you can get it for free at Live.com

Post TechEd update

Well, the show is over, and as usual Microsoft put on one hell of an event. There were quite a few different vendors this year, most likely owing to Microsoft splitting up the event into two differnent shows, one for Devs and one for IT Pros like me. I did get to meet the guy who owns Sperry Software (http://www.sperrysoftware.com/), who makes some of the worlds best utilities for Outlook. Personally I've used a few, but love their Outlook Contact Duplicate Eliminator. For those already using it, Mr. Sperry assured me that the current version doesn't have the annoying plug-in issues on Outloook 2007/Vista anymore!

I was impressed by the overall show, and especially by the depth of different subjects covered both in the Microsoft pavillions and in the sessions. However, I was really upset with the tiny showing of the Messaging and Collaboration (i.e. Exchage) team. The few kiosks that they had were all dedicated to the Unified Messaging and Outlook Office Communication Server functions, and there were few Exchange-dedicated experts to be found among the staff. I guess if they're not launching a new version, it doesn't make a lot of sense to dedicate resources, but still there were a ton of us Exchange folks there, and no one to talk to.

All of the other product lines seemed to be at least decently represented, though. SQL Server, Office Communicator and - of course - the new Server 2008 systems were the highlights of the show, with pleanty of staffers availble to answer questions.

The show itself had one major impact on me. I switched off the iPod and onto a Zune. Long story short, my nearly new Touch from Apple apparently is incompatable with the previous TV Out cables (which I paid a ton for!). That was the final straw for me. No all-you-can eat systems, proprietary file formats, sneaky installation tactics (Safari? You bet, even if you didn't want it...) and all the other recent crap from Apple pushed my tolerance level beyond acceptable, and so I took advantage of the show-only pricing and grabbed a Zune. They have come a long way with ti, and if they get a movie rental partner, the iPod is in serious trouble.

So along with the Vista laptop, Windows Mobile phone and other Microsoft products, now I'm on their portable player as well.

Hmm, I wonder if you can jury-rig it to get your email from Exchange...

Cluster-Funk ROCKS out!

TechEd 2008 has been a great show so far, with dozens of sessions on everything from Exchange to the new 2008 Hyper-V solution set. Of course, last night Double-Take blew the roof off Margarittaville in Universal Studios with our own house band, Cluster-Funk. About 700 folks joined us for the festivities, which has got to make us one of the biggest TechEd parties this year.

Of course, the show would be a lot less fun without the band, so I snapped some pix of the guys up on stage. Beleive it or not, the whole crew works for Double-Take except the drummer and the violinist. Our drummer is a friend of the family though, and the fiddler works for Microsoft!

By the way, for those who are wondering, THIS is our CEO. Some days I just love my job...

CCR display leaves a bit to be desired

I recently got a chance to play with Continuous Cluster Replication (CCR) hands-on at the Microsoft Technology Center in my home-town of New York City, and I can honestly say it is a very solid-performing technology for High Availability between two clustered Windows 2003 servers. I have yet to physically see it on Server 2008, but from everything I’ve been reading, it would appear that CCR is even better on that platform. Next week, I’ll be at TechEd with Double-Take Software (booth 809, if you’re heading to the show), and I’ll be spending at least a little time over at Microsoft’s pavilion checking out their new 2008 systems, including the Exchange 2007 integration.

Meanwhile, there was one thing I found lacking in CCR, the information that is available in the Exchange Management Console (EMC) and the Exchange Management Shell (EMS) is very limited, and doesn’t really give you that much detail on what’s going on.

In the EMC, you can clearly see the current status of the CCR replication for each storage group, which is good, but that’s about all you can see. You cannot configure or manage CCR through the EMC, though you can set it up via the Exchange Installation Wizard. Once it is running, you really can’t do much of anything except move the resources from one node to another with the GUI. While I understand Microsoft’s desire to move Exchange Administrators to the EMS platform and use cmdlets for everything, asking us to drop to the shell just to find out how many logs are in queue is a little much.

As for the shell, EMS does give you much more information on what’s going on with any of the three forms of Exchange 2007 replication. However, trying to get the hang of the several PowerShell commands required to find that information is going to take a bit of time, even for the old MS-DOS hands among us.

Please, Microsoft, give us a single, clear, easy to use readout for this vital resource. Just seeing “Healthy” doesn’t really say everything we need to know. It doesn’t show us how many logs are queued up for transmission due to bandwidth issues that are only slowing down the network, not stopping it. It doesn’t tell us what the replay depth is (though EMS can), and it doesn’t give most engineers the “happy, fuzzy feeling” that we really would like to have with Exchange clustering in general, and especially with a relatively new form of it.

I will send off dispatches from TechEd during next week, so keep an eye on the blog!