This is the eulogy that I will give later today for my best friend. He was born on 7 September 1963 and passed away on 26 April 2013 at 10:40 AM. He fought kidney cancer the past year and had to endure a slow, yet painful death that I talked about in my last article. I tried my best to capture the spirit of our conversations this past year. He knew early on that he had at best six to twelve months to live. He also knew that the medications he would take had about a 10% chance of succeeding. He did his best to make peace with it and made right in everyone’s eyes. In the end, as he took his last breath, he was surrounded by his family and everyone knew how much he loved them.

I will miss him and here is my story…

Tim would be very happy to see so many of us here today. It is times like this one that clearly let us know how both fragile and precious life is for each one of us. I believe that God only takes those, no matter what their age, when he feels they are ready to come back home to Him. I am here today to tell you what both Tim and his family have meant to me. More importantly, I want to share with you how Tim Locke made me a better person.

I would like to start with a few quotes, before I begin with my own words. These words are from Steve Jobs’ 2005 Stanford Commencement Address:

No one wants to die. Even people who want to go to heaven don’t want to die to get there. And yet death is the destination we all share. No one has ever escaped it. And that is as it should be, because Death is very likely the single best invention of Life. It is Life’s change agent. It clears out the old to make way for the new.

Samuel Johnson wrote that “It matters not how a man dies, but how he lives.” I find both strength and wisdom in each of those quotes.

Tim loved Leah, his wife of 27 years and his children (Karen, Geena and Erik) very much. He always talked about them, since they were first and foremost on his mind. He lived for his family and that is what many of us remember about him. It is also what made him so special.

As for me, I met Tim in 2008, when we were working together in Las Vegas and fate would bring us together via a software product called Mariposa. Like most relationships, we started off as strangers and it did not start smoothly.

The first time we met, I was checking mail on my smart phone and it just annoyed him. When we started talking however it become clear that it was a great fit for both of us. He needed someone with a strong technical background and I needed a personable manager, who could provide clear direction.

I can clearly recall what started our friendship. It was a simple act of kindness by me, where I had no expectations I offered to take him to the airport and pick him up when he flew back from San Diego every week. It was during this time when the economy had really taken a dive and money was tight for everyone. That act of kindness started opening a lot more doors. I do not think that either one of us knew it at the time, but our friendship would grow due to the little things we would do for one another.

I read a book a few years ago; where they said a relationship is like a bank account. You need to maintain a positive balance for it to hold a meaningful value. Over time, you will need to deposit into this account and the system only works if you put in more than you take. I see relationships the same way and I agree that it is better to give than receive.

Before too long, we become roommates and it started out rough for Tim. I have a dog named Foxy Brown and at this time she was a three-month old puppy, who was not yet house broken.

On the night we moved in, Foxy had pooped the entire length of the hallway from his bedroom to the front door. Let’s just say that started a special relationship for both of them. A few weeks later, Foxy would strike again. This time she had barged into his bathroom and started ramming the glass door with her head. Needless to say, Tim was just trying to defend himself with a bar of soap and a shampoo bottle. Of course he had to tell everyone at work about it. Tim loved playing the victim and nothing was ever his fault. That is a trait of his that I dearly miss about him.

The best moment was when he fell asleep on the couch and Foxy had jumped up there to snuggle with him. I was in the other room, but this moment would soon catch my attention, when they both started to snore. Of course, I had to email Tim and a bunch of close friends, where he did not know a single person. Foxy loves Tim was the title of the email I sent that day with those pictures.

Here is another photo:

I will never forget the time when I first met Leah. She quickly saw that we basically had no food and went grocery shopping. I think it was the first time that we actually felt human in that apartment. I would not be at all surprised if that trip was not motivated by the fact Tim admitted to eating Foxy’s dog treats. Not only did he admit that he ate a few of them, but they also tasted really good.

When I look back, I see the reason that our friendship grew was not because of some major events. The real significance of our relationship is that it simply continued. We both went in different directions where could have easily lost touch, but we understood that we had a special bond and kept it alive. As I speak to you today, I realize what Tim and his family has taught me regarding relationships, especially great ones, they are not based on needs. Instead they are built over time and one you start down this continuum, you realize that all you really did was accept and uplift one another.

I was raised a Christian, but I know some people have their doubts with religion. They look at the Bible as a book of rules that tell you what you can and can’t do. Sadly, I think a lot of people overlook a more important lesson, the ability to love one another. If you just look at Jesus’ travels, you will see that it was all about the people he met and the actions he took to make a difference. I am not sure why so many people fail to see that important teaching, but I wonder if that is why so many people are unhappy today. We desire to be accepted, but yet some get hung up on a perception that causes them to feel alone, angry and betrayed.

It was great to see so many neighbors offer love and support to the Locke family this past year and especially these past few months. I saw a lot of love and acts of kindness to a family that really appreciated it.

I do hope that more people will find this simple truth. Life is short and you should live it with no regrets. Stop focusing on what you do not have and remind yourself how blessed you are with what you have today. Do not doubt the power you have as a person to make a difference in the others. You never know when you might say or do an act of kindness to someone that inspired them to keep fighting onward.

I could not imagine my life today without Tim being such a key part of it. I find comfort that God has Tim snug closely in his arms and is welcoming him home. I will long to hear his constant rants that usually led him to picking on me, which showed me how much he cared.

As I close this speech, I would like to share another quote from Steve Jobs.

When I was 17, I read a quote that went something like: “If you live each day as if it was your last, someday you’ll most certainly be right.” It made an impression on me, and since then, for the past 33 years, I have looked in the mirror every morning and asked myself: “If today were the last day of my life, would I want to do what I am about to do today?” And whenever the answer has been “No” for too many days in a row, I know I need to change something.

Remembering that I’ll be dead soon is the most important tool I’ve ever encountered to help me make the big choices in life. Because almost everything — all external expectations, all pride, all fear of embarrassment or failure – these things just fall away in the face of death, leaving only what is truly important. Remembering that you are going to die is the best way I know to avoid the trap of thinking you have something to lose. You are already naked. There is no reason not to follow your heart.

I find comfort in those words as I also start to embrace my own mortality. Live each day as if it is your last. Then remember that we can all make a difference when we choose to give a little bit of ourselves without taking away more than we gave.

God bless you all and I will you miss you Tim. Thank you for being a great friend.

As stated in my last post, I am taking a bit of a break from technical writing as I am spending (what little) time (is left) with my best friend Tim J. Locke. As most things in life, time wins and we must accept our destiny that one day we will share the same fate and that is death. Tonight, I write this article with a heavy heart; since it is clear that I only have days with him. Talking with the hospice care, we doubt that he will ever wake up again and to make matters worse, his skin is turning yellow and gray from his liver shutting down. I read a book a few days ago called Making the most of the time we have: Caring for a loved one at life’s end.

It describes some of the symptoms as the body starts to shut down and let go of its gripe on life. Here is the stage that we are currently at right now:

Days Before Death:

• Decreased intake of food and fluids
  
• Dry mouth
  
• Restlessness
  
• Labored breathing
  
• Fever
  
• Darker urine; reduce output
  

This is a very tough process to watch when it goes so slowly. You see the person you know slowly turn into someone who is not the same person you first met. They start to deal with a loss of appetite  fatigue and restlessness. It is almost as if the body knows it needs to prepare for a journey and they start trying to go somewhere and anywhere. You see their thoughts go all over the place and then you start to see how the body begins to slip away. The only true blessing that you have is that you had a chance to make peace with him and let them know how much he meant to you.

In the end, it is weird how simple life becomes for someone and how all the things you chose to achieve in life is pushed aside. As an adult, you worry about finding a good job, a place to live, food in your stomach and then as that becomes easier to obtain, then you start to focus on a nicer car, a home of your own and a person to love. It is never-ending series of pursuits to make your life feel more enriched during this ride of life.

In the last few months of your life, your focus changes drastically in what you seek. You want to resolve any past conflicts; you want people to know how much they meant to you, if you have family you want to ensure they are taken care of financially. I was very fortunate (and thankful) to tell Tim how much of a better person I became and because of him.

Sadly, the last remaining weeks of a person’s life is really simple. Goals are to eat and drink, plus keep it down. You also hope for bowel movements to rid the body of toxins, but as the weeks pass, the body is less capable of performing those simple tasks. Then it becomes a battle to keep the person calm and reassured, since the mind is now dealing with less blood flow and battling toxins that should have been flushed from the system.

The last part that we are dealing with now is to reassure him and tell him everything is fine, in hopes that his brain will relax and allow him to end the suffering. Sad as it may seem, we are hoping that he will let go, even though we do not want to see him leave us. We know it is time and we want him to be at peace. I get the sense that he has stayed on this long because he is very worried about how his children will be during their adulthood. He loved them very much and he really wished he could stay around longer to help them with words of wisdom as they become adults.

I met Tim J. Locke five years ago in Las Vegas, while working for a casino gaming company that is well-known for making slot machines. Looking back, I am amazed how fast the time flew by and it just goes to show you that each decade that passes, the faster the next one seems to go by. If those 5 years flew by, imagine how fast this past year went by when we found out that he had stage 4 kidney cancer.

It just seems like yesterday, that we were both working in Las Vegas, drinking and griping about how frustrating the job was and next thing I know I offered him a ride to the airport. The economy had just crashed and I knew that parking your car was expensive. Before I knew it, we become roommates and I will never forget the time he moved in. My dog Foxy Brown (5 months old min-pin) pooped from his room all the way down the hallway. What a great way to start a friendship! It seems that my dog and Tim would always be at odds from that point onward. For example, one day as he was taking a shower and I had already left for work. My dog had barged into his bathroom and started ramming the glass door with her head, trying to get in the shower with him. He was so freaked out that he almost slipped and fell in the shower. He had a dog, but it was a low energy breed (Pug) and he was not ready for a high energy breed, especially as a puppy.

Quickly over time, he became my best friend and we would take trips to his home in San Diego. We used it as an excuse to leave Las Vegas I would meet his mother, all four of his brothers and sisters. I would then meet his 2 daughters, his son and his wife. (I will always cherish how she made both of us feel when she came to visit us and went grocery shopping!) Somewhere along the way, I became family and it became an annual event to spend time with them over the fourth of July. It is amazing to look back and see how this friendship has evolved. I can truly say that I have been blessed to have known him. He was always supportive of me and encouraged me to pursue my dreams, plus he always had great advice as I navigated through life. He was the person I always longed to hear from and I could never wait until we spoke again.

As I watch him start to pass away, I become quickly overwhelmed by emotions, knowing that he is in a deep sleep with very little chance of waking up. I am however thankful for letting him know many times beforehand that I loved him. In the end, I think that is really all that matters is let others know how much they mean to us and even more important is to make an effort to make a difference to the people we meet along this journey. We are not here forever and the biggest reward in life is when you make a difference in others, because it is what truly makes us happy.

I believe in God and was raised a Christian. I know some people look at religion and have their doubts. I know some people look at the Bible as a book of rules that tell you what you can and can’t do. Sadly, I think a lot of people overlook a more important lesson, the ability to love other people. It is written all over the place and if you know Jesus’ journeys, it was all about the people he met, plus the actions he took to make a difference. He healed a lot of people, he spoke with a lot of people, and he passed on a lot of guidance. He also did not care about social statuses of people and he crossed all boundaries. He saw the rich as well as the poor. I am not sure why so many people fail to see that important teaching, but I wonder if that is why we see so many people unhappy today. I look at the recent violence in all places of the world and I wonder how many people feel left out and isolated to the point that they blindly start killing others. (i.e. like we just saw earlier this week with the Boston Marathon.) I wonder what a difference it would have made if we took time to get them know them and accept them as they are, instead of trying to mode them as we want.

I hope and wish that more people find that simple truth. Life is short and you should live it with no regrets. Stop focusing on what you do not have and start focusing on what you do have, then remind yourself how blessed you are with what you have today. Do not get so caught up in what you are pursuing that you forget what you have around you. It sucks to be taken for granted. Last but not least, do not doubt the power you have as a person to make a difference in someone you meet. You never know when you might have said hello to someone who was at the last bit of strength to continue and your act of kindness inspired them to keep fighting onward.

No matter what you believe, I want to say God bless you and I hope you will find the happiness that you really deserve in life. I hope you find some people in this life and make a difference for them by your actions.

I know I have been quiet for the past month, but I have been dealing with a personal issue. My best friend is slowing passing away and has “‘weeks if not days” to live. About a year ago, he found out he has kidney cancer and sadly he found even worse news… it was stage 4, which means it had spread to other organs. This past year, he has been on the losing end of this battle as it has aggressively spread throughout his body. It is now to a point where it has spread to his brain.

Needless to say, each day has gotten worse and he is now rapidly losing his memories as well. I am now preparing to go down for the last time to be there for his family and help them get through this. If it is not too much to ask, if you could send some prayers or best wishes their way… I would really appreciate it.

When I come back, I have a lot of articles that I plan to write, some on SQL and Windows, plus some networking stuff. I just have not had the time and also my mind has been on my friend and his family.

In my last post, I told you about moving an existing AMI to another region by exploiting the copy snapshot feature. On 13 March 2013, Amazon announced a new feature called Copy AMI:

Dear Amazon Web Services Customer,

We are excited to announce the immediate availability of a new feature: Amazon Machine Image (AMI) Copy. AMI Copy enables you to copy your AMIs across AWS regions, thus making it easier for you to leverage multiple AWS regions and accelerate your geographical expansion and help increase application performance and availability.

To use AMI Copy, simply select the AMI to be copied from within the AWS Management Console, choose the destination region, and start the copy. AMI Copy can also be accessed via the EC2 Command Line Interface or EC2 API as described in the EC2 User Guide. Once the copy is complete, the new AMI can be used to launch new EC2 instances in the destination region.

There are no additional charges for using AMI Copy, but you will be charged to transfer the AMI out of the source region and to store the copied AMI in the destination region.

We recently launched Amazon EBS Snapshot Copy, which enables you to copy EBS Snapshots across AWS regions, and last month we significantly lowered the cost of transferring data between AWS Regions (by 26% to 83%). Together with AMI Copy we hope that these updates make it easier for you to take advantage of the AWS global footprint.

Sincerely,

The Amazon Web Services Team

Now I am going to show you how simple it is to perform:

1. Select your AMI section under Images on the left side of the EC2 Dashboard:

   

2. Right click on your AMI and select Copy

   

3. Then select your destination region, give it a user-friendly name to help you quickly identify it and change anything in the description as needed. (Again, to help you quickly identify it)

   

   After a while, depending on your root volume activity/size (and other volumes that might be part of your image) it will show up in your new region and then you will deploy it as you would in the original region. Right click the AMI and select Launch:

   

   Now you do not need to worry about launching another instance, just so you can detach and swap out a root volume. This is a great new feature and I like the direction this is headed, as it will allow for true redundancy with AWS by allowing you to be in multiple regions easier and taking advantage of their global foot print.

When I used to work at Amazon Web Services (AWS) one of the pain points was the ability to migrate your fault tolerant strategies to another region. Amazon had these great technologies to help you fail away from trouble areas. A great overview of these strategies is in a video, that I did last year, which shows you how to create an AMI and snapshots plus move them around inside of a region. You can watch that video here.

For example, you have a recent image of your web server or you have a snapshot of an EBS volume that host a key database. You are located in a region of AWS, let’s say US-East with all of these backups and now there is an outage that is like this one from April 2011 that was called “AWS Cloud Apocalypse 2011.” There were even jokes that Skynet from Terminator had taken AWS out as seen on their forums. For more articles, you can search on your favorite search engine like this: “skynet site:forums.aws.amazon.com.”

Sometime in the past 6 months, Amazon has finally allowed this feature and it is long overdue! Here is a great whitepaper from AWS on this topic and here is the support document.

Let’s walk through the steps of creating an AMI with an EBS root volume, which most Windows instances should be now:

If the image is too small in your browser, then click on the image and you can see the full version.

1. Create a current image of your machine, which is called an Amazon Machine image (AMI):

   Right click on your EC2 instance and then select Create Image (EBS AMI)

    

   

2. Provide a name and description to help you easily identify this new image

   

3. View the pending status

   

4. Verify the Status

   

5. Verify the status is available

   

Now you are ready to migrate this image to another region. Note the AMI-eee570de and go to snapshots.

Right click on your snapshot with the AMI-eee570de and select copy snapshot

Select a region (I choose US East) and then Yes, Copy

Notice the advice:

Then select the region:

Then click on snapshots for a status:

They even added a nice progress bar as well:

Completed:

Now you will need to register the snapshot as an AMI, by right clicking it and selecting Create Image from Snapshot:

Then it will give you the properties and sadly, it looks like you still can’t use Windows snapshots to create an image.  The underlying software that AWS uses will not  allow Windows volumes to be created from a snapshot, since the software assumes all volumes are Linux. However, these steps will work for Linux, I may as well show you how that will work.

Give a name  and description, then match then architecture  and if need be you can change the Kernel and Ramdisk ID to help you. Select Yes, Create when ready.

Check the status:

Now you have a new image (again this will not work for Windows machines):

Another thing to note is that there are 2 different forms of virtualization. Hardware-assisted virtualization and Paravirtualization. It seems that AWS assumes most systems will be Paravirtualization, which may be causing the issue where you can’t create an AMI from snapshot on Windows EC2 systems.

I am wondering if you could do the following:

• Create a volume form Snapshot as /dev/sda1
  
• Launch a new instance with the same parameters (Windows 2008 R2) and wait for it to load
  
• Stop the new instance
  
• Detach the current root volume and then attach the volume we created from snapshot
  

So I tried it

Then I verified the new volume existed:

Launched a new instance:

Waited for the instance to launch and then tried to login (did not bother to get the password, since I want to log in with my instance):

Then I stopped the instance and detached the root volume:

Then attached my volume that I created from my Snapshot as dev/sda1:

Now I started it and tried to RDP into my instance and damn my guess was correct:

Here you can see it with my volume:

So that is how you do it in these simple steps:

• Create an AMI
  
• Copy the snapshot to the new region
  
• Create a volume form Snapshot as /dev/sda1
  
• Launch a new instance with the same parameters (Windows 2008 R2) and wait for it to load
  
• Stop the new instance
  
• Detach the current root volume and then attach the volume we created from snapshot
  

As stated in an early article of mine, there is a great free book from Microsoft Press that talks in good depth about Windows Server 2012. Today, I wanted to share some of the information that I learned from that book. (Limits and new features)

Here is a simple comparison of the different limits between Windows Server 2008 R2, Windows Server 2012 Release Candidate 1 and Windows Server 2012:

NOTE: Please click on image for a larger view.

I verified this information on TechNet with Hyper-V Scalability in Windows Server 2012 and Requirements and Limits for Virtual Machines and Hyper-V in Windows Server 2008 R2.

Another great bit of information is the TechNet article called What’s New in Hyper-V which outlines the new changes. Cool new features, like the following:

Hyper-V Replica
Resource metering
Improved Storage migration (No more downtime)
Virtual Fibre Channel
Virtual NUMA

In my last article, I spoke about a new cool feature of Windows Server 2012. The ability to go from a server GUI to Server CORE (or vice versa) without having to re-install the Operating System! You may ask yourself, “Why would I care about that?” To be blunt, Server CORE is very similar to the way most Linux distributions are now working… no Graphic User Interface. Most web servers are running this way in world today with some distribution of Linux with an Apache Web Server. It decreases the attack surface area, has less code to run the OS and takes a lot less CPU or memory resources to run. This means that older machines can stay in production longer. Another good case might be you have to ship a domain controller to a branch office where there is no IT staff or a secure place to lock down the server. So now you can send them a Read Only Domain Controller (RODC) that is also running as Server Core. Both of those roles/features were introduced with Windows Server 2008.

In Server 2012, you can flip back and forth much easier and without having to re-install the OS. This is really helpful. Think about this scenario, I am told at the last-minute to ship a server that is already running a GUI version of Windows Server 2012 to a branch office, where our current policy is to ship Server CORE. In Windows 2008 or R2, that meant a new install of the OS, whereas in Server 2012 that is a few PowerShell commands and I am done! J

I have been reading William R. Stanek’s book Windows Server 2012 Pocket Consultant. I highly recommend it. (Hint hint)

Today, I am going to talk about Minimal Interface, which is a new GUI mode in Server 2012. TechNet has a really good article on this topic as well, located here. I want to be clear, that this article will not be a step by step, like yesterday, but it will give you all the commands to go in any direction with any feature.  With that being said, let’s get started.

What does a minimal interface look like? It looks a lot like Server CORE, but with Server Manager and MMC snap in support (as seen below):

What does the GUI look like? This is my full GUI installation before doing any commands

What does Server CORE look like? This is my CORE installation before doing any commands

PowerShell commands:

To convert a full GUI mode to a minimal interface:

Uninstall-WindowsFeature server-gui-shell -restart

TIP: To see what will happen before it takes action use the –Whatif (See below)

TIP#2: Add-WindowsFeature is an alias of Install-WindowsFeature and Remove-WindowsFeature is an alias of Uninstall-WindowsFeature. You can use Add instead of Install and Remove over Uninstall.

This command did not run, but you should not always trust what you hear, since someone could be lying to you. (Lord forbid a mistake happens as well) To verify it did not complete in the GUI, open Server Manager, then Dashboard, then Add roles and features, click next (4 times) and then on Features drill down to User Interfaces and Infrastructure

When you do execute the command, you will see a status bar, just like when you apply a Windows Update patch:

To convert a minimal interface to full GUI mode:

Install-WindowsFeature server-gui-shell –restart

As you can see it did not add the Desktop Experience:

To add it type Install-WindowsFeature desktop-experience –restart and hit enter

To convert a minimal interface to full GUI mode with the Desktop Experience:

Install-WindowsFeature server-gui-shell, desktop-experience –restart

Converting to Server Core Installations

To convert a minimal interface or full GUI mode to Server CORE:

Uninstall-WindowsFeature server-gui-mgmt-infra –restart (Full)

Uninstall-WindowsFeature server- gui-shell –restart (Minimal)

To convert a Server CORE to minimal interface or full GUI mode:

Install-WindowsFeature server-gui-mgmt-infra –restart (Full)

Install-WindowsFeature server-gui-shell –restart (Minimal)

Here is an example, open PowerShell in server CORE by typing powershell and hit enter

To go to a Server 2012 Minimal Interface, type in the command Install-WindowsFeature server-gui-shell –restart and hit enter

You will see a progress bar as it starts to add the feature and then it will restart

In closing, I want to say that I am really impressed with Windows Server 2012 and more importantly with PowerShell 3.0. It very simple and has straight forward commands. I am really glad to see these new features and I believe that Server 2012 Minimal Interface will allow people to get away from the GUI, while still using the tools that have gotten used to using over the years with many different versions of Windows. This is great and I am glad to see that we are moving more towards command line tools and relying less on the GUI tools. This is great, because it has a significant smaller attack surface area and the reason is that there are less components installed, which also means there are fewer services listening to allow an intruder into your environment. It will take some time getting use to this new direction, but it is the right direction that we should have taken years ago.

I have been preparing to upgrade my Windows and SQL certifications for 2012. I was checking out an article on switching from Server GUI to Server core. This article was written for Windows Server 8 Beta and was called How to Turn the GUI Off and On in Windows Server 2012. It gave me the impression you could also start from server core and go to a GUI, but I see the menu was different from the one they showed.

Here is mine and note that option number 12 is to log off the user:

Here is one from that article, take note of option number 12 (Restore GUI):

So it made me wonder how it is done and why not update the information for Server 2012 in the process. (No Beta or RC versions) I started with an installation of Server 2012 with a GUI and decided to remove it. Let’s begin with the process of removing the GUI, which is pretty easy. Also you might be asking yourself, why would I want to have SQL server running as a VM and as Server Core? (Plus are there any advantages you would gain using Windows Server 2012 Hyper-V over VMWare?)

Here is a very brief overview and my opinion:

• Faster Failover Times for Unplanned Failovers- Unplanned failover in a VM often takes ~5min. AlwaysOn unplanned failovers take <30sec. (Also, remember VMotion doesn’t help with unplanned failovers)
  
• Reduced downtime for SQL and Windows patching and upgrades. (Often overlooked as a benefit)
  
• Faster DR, as selected workloads can be quickly brought online in a DR scenario
  
• Readable Secondary for offloading report traffic
  

Open Server Manager and then Click on Manage, and then select Remove Roles and Features from the menu.

Select your server from the server pool and click next.

Since the GUI is not a Role, you can just click next again to skip past the Roles section. On the Features page, you need to ensure that the box is unchecked next to the “User Interfaces and Infrastructure” option, and then click next.

Now click the “Restart Destination Server” box, then click remove.

Then you will see Windows make the new changes:

After the reboot, all you have now is a Server core installation with only a command prompt.

Type in “sconfig”

Notice the menu has changed since Server 8 Beta and there is no more revert back to GUI option:

I was reading the Windows Server 2012 Pocket consultant and it mentions that you have 3 types of installations now:

1. Server with a GUI
   
2. Server Core
   
3. Server with minimal interface
   

I also found this article on the Microsoft Premier Field Engineers blog called Converting a server with a GUI to or from Server Core. It also talks about the three installations of Windows Server 2012 as well. I will look at exploring Server with a minimal interface at a later time. For now, I am just going to focus on Server GUI and Server Core.

To add the server back to a GUI, you will need to use the new 3.0 PowerShell cmdlet Add-WindowsFeature Server-Gui-Shell to convert it back. Again, this is a very simple method to convert back to a GUI from Core. To get there, type in powershell at the command prompt and then type Add-WindowsFeature Server-Gui-Shell as seen below:

It will start running the job and you will see a status menu, like this one below:

When it is done, you will need to reboot the server and the easiest method is to type shutdown –r –f –t 0 (restart, force, time 0 seconds)

Again, it will make the changes for you while the restart is happening:

After the restart, you are back to the GUI as seen here:

I am really glad to see that Microsoft made this change and I can see a lot of good here. I am a big fan of flexibility and this allows for it, plus I really like the direction PowerShell is headed with this release of Windows.

I think it is great for people who start off with a GUI and then decide that it causes too much of a performance hit or they want to move a server to a regional office where there is no support staff. This allows you to save on resources and also lock down the exposed surface area. More importantly, if you ever ran into a weird issue, you could convert back to GUI mode for easily troubleshooting and you get to do all of this without having to re-install the box. Plus you get to make your boss think you are incredible as well, by doing the task in a very fast manner.

Good news came out earlier this month when Microsoft released Cumulative update package 2 (CU2) for SQL Server 2012 Service Pack 1. It can be downloaded here and if you are new to CU packages, there are not like a regular Service pack or Windows update. These CU fixes are not as tested as regular like either a service pack or Windows updates, plus in most cases the CU is only recommended if you have the issue they fixed in the release notes.

In addition to all of that, you also do not get a link to download. Instead you will need to sign up via email to get the download and it will look like this one:

Another thing to note is that this patch is only relevant for builds with SP1 installed. (11.0.3000 -> 11.0.3338) Do not attempt to install on SQL Server 2012 RTM (any build < 11.0.3000) or any previous version. To find your current version, you can type the following:

SELECT SERVERPROPERTY(‘productversion’) AS VersionNumber,

SERVERPROPERTY (‘productlevel’) AS ProductLevel,

SERVERPROPERTY (‘edition’) AS Edition

You could also just use this TSQL code: SELECT @@VERSION;

As you can tell, my SQL instance is ready for Cumulative update package 2 (CU2) for SQL Server 2012 Service Pack 1.

If you are like me, you enjoyed the benefits of sysprep with Windows servers or desktops. It saves a lot of time deploying new machines or reimaging machines that had an issue, like corruption or malware. Also, you could even customize your image to put in your own company’s logo and support telephone numbers. It gives you a sense of personalization to your company and allows them to go to the right support person. As for SQL server, I have not always been a big fan of sysprep, because quite frankly it was very limited. Let me show you what I mean.

First off, it is recommended that you try this image on a system without SQL installed. That way you do not need to worry about what components may or may not be installed. If you try it on a system that has an unsupported component of SQL server to use with sysprep, then you will see this error message below:

Launch the 2012 SQL installer and then go to Advanced, then select Image Preparation of a stand-alone instance of SQL Server. (As seen below)

Then click next a few times to install the setup files and eventually you will come to the menu and see this:

Please note how you do not see Analysis or Integration services, which is what I meant earlier by it is limited in the number of features it is supported. Now, I am going to go ahead and update
my SQL server instance to Cumulative update package 2 (CU2) for SQL Server 2012 Service Pack 1. Here is my instance version now:

Notice now, how many more features that I have with the menu with Cumulative update package 2 (CU2) for SQL Server 2012 Service Pack 1.

These new features, now allow anyone to use sysprep more often with SQL Server 2012 and I am very glad to see these changes. It is great when you have choices to make getting tasks done and I love having flexibility.

For more information on this topic, please see the following articles:

Install SQL Server 2012 Using SysPrep

Expanded SysPrep Support in SQL Server 2012 SP1 CU2

Description of new features in Cumulative Update 2 for SQL Server 2012 SP1

Cumulative update package 2 for SQL Server 2012 Service Pack 1 Download

Sysprep Overview

Last week, I had someone reach out to me asking for help with a new SCVMM 2012 server. It was a new installation and for the most part it worked well for them during the past month. However, recently they started seeing the error below, which prevent them access to the Virtual Machine Manager:

At first they ignored this error, since rebooting the SQL server seemed to fix the issue. As time went on, the error become more common and they found themselves rebooting more frequently. Sadly, they did not share that information with me that they had to reboot so often. With the information they had shared with me listed above, I recommended that they investigate the following areas:

1. Check for any auto services that are not started
   
2. Check to ensure SQL, VMM and WS-Management services (etc.) are started
   
3. Check Application and System logs for any service complainants
   
4. Check Application and System logs for anything wrong with SCVMM and/or SQL Server
   

For the most part, all services were started and not having any issues. They did not see any credential errors with any service in the logs. At first they suspected the SCVMM application had an issue based on the error logs below:

Application fault again:

At this point, I asked them if they had any experience with SCVMM and were aware of the relationship it has with SQL server. They stated that they had some, but were not sure of the SQL details, so at that point I explained that all of the information they see in the SCVMM console is stored in a database on the SQL server and the application depends on it. At that point they understood why I was asking them about SQL server.

Then they shared with me this gem of information that I was looking for to help them:

I also asked them to log into SQL and check the error logs to confirm what we were seeing from this error message on the SCVMM server. (Above)

They first looked at the installation logs and saw an underlying error in their initial setup logs were coming from SQLServer :

InnerException.Type: System.Data.SqlClient.SqlException, InnerException.Message: A transport-level error has occurred when receiving results from the server. (provider: TCP Provider, error: 0 – The semaphore timeout period has expired.)

When I was able to finally persuade them to look at the SQL server, there are no other databases on that server. However, we did see this error message was all over the place:

A transport-level error has occurred when receiving results from the server. (provider: TCP Provider, error: 0 – The semaphore timeout period has expired.)

At this point, I let them know that “The semaphore timeout period has expired” is a Winsock error that usually indicates the TCP peer did not ACK some data. This typically means that some part of the network layer croaked. (i.e. a NIC or switch is malfunctioning) It could also probably happen if the server machine was unplugged or hard powered off, but SQL Server itself or the client process wouldn’t cause this type of error. Even if you forcibly terminated the SQL Server process on the server, you wouldn’t see this behavior. It would at least send a TCP Reset. (RST)

I asked them if they knew how TCP works and then explained the 3 way handshake that occurs with these diagrams:

Basic overview:

Then I went further and explained some more details with this one:

I recommended that they either use Netmon or Wireshark  to perform a network captures to look at where the failure is actually occurring. However, they saw this error message below and just asked the data center people to look at the SQL server NIC and switches. (It appears it’s either a NIC issue or a networking issue on the SQL server side. I found these 2 logs after digging around for a while and they seem to be consistent throughout)

Personally, I would have gone a step further and pin-pointed it to the actual device. I would have done a quick TCP based trace route, like tcptrace. (It will require you to install wincap) A lot of people are not aware that ICMP traffic, like ping and tracert, is often either dropped or deprioritized a lot in busy networks and it can lead to false alarms. If your network traffic is leaving a building, then I would recommend using a TCP connection to test, so that way I am guaranteed to get a response back. After that, then I would have confirmed my findings with the packet capture by looking for a TCP 3 way handshake.

Moving forward, I hope that people were able to see the key here is to look at basic networking and event viewer (plus other application) logs to get a true sense of what is going on. Sometimes, your perception of things can lead you down the wrong path as you allow it to become a reality. In this case, most of the early troubleshooting was focused on SCVMM and not a lot of attention was placed on the SQL server.

This is an article that I have been meaning to write for months. It was very exciting for me when I was trying to solve this issue and I was even more pleased with performing this task when I found a few DBA’s had never attempted this operation. I want to share that knowledge with you today.

Little background about this configuration:We had a production 2008 R2 SQL Server that was configured as Synchronous Database Mirroring (High-Safety Mode) which means we were writing to both nodes at the same time and had a witness server for automatic fail-over. Here is a great description from Books Online about our mirroring:

To achieve synchronous operation for a session, the mirror server must synchronize the mirror database with the principal database. When the session begins, the principal server begins sending its active log to the mirror server. The mirror server writes all of the incoming log records to disk as quickly as possible. As soon as all of the received log records have been written to disk, the databases are synchronized. As long as the partners remain in communication, the databases remain synchronized.

After synchronization finishes, every transaction committed on the principal database is also committed on the mirror server, guaranteeing protection of the data. This is achieved by waiting to commit a transaction on the principal database, until the principal server receives a message from the mirror server stating that it has hardened the transaction’s log to disk. Note the wait for this message increases the latency of the transaction.

The time required for synchronization depends essentially on how far the mirror database was behind the principal database at the start of the session (measured by the number of log records initially received from the principal server), the work load on the principal database, and the speed of the mirror system. After a session is synchronized, the hardened log that has yet to be redone on the mirror database remains in the redo queue. For more information, see Database Mirroring Sessions.

We were limited on space (isn’t that always the case?) and were only doing mirroring on key databases. However, as the application started to mature, another database was created to capture additional monitoring and a decision was made to store this database on the witness server. At the time of this decision, there was plenty of disk space.

As we know, the more time passes, the more things grow and features get added, which means the database was growing at an alarming rate. I did a lot of monitoring on the database and created jobs to purge data that was no longer required due to our retention period. For 6 months, I made sure we had enough space to prevent auto growth and that we kept the database with at least 10% of free disk space. Sadly, it just got to a point where we literally had less than a week left before we would fill up the drive.

I had to come up with a solution and it needed to be tested, plus done quickly. To make matters worse, I had a lot on my plate that was just as important and I could not engage anyone else for assistance. I need to perform 2 critical operations that same week and get this database taken care of ASAP.

My initial thoughts were that I could either do a restore or attempt to mirror the database to the new server that was built earlier. I first tried the restore method, but I had reservations about it, since I knew the backup took a very long time. The problem I ran into was the backup was timing out due to various issues. I was more intrigued by the mirroring option, so I decided to test that approach and ended up learning a lot.

More importantly, I was successful in getting this database migrated to the new server that has plenty of disk space. I would like to show you how I proceeded.

The current servers in the mirror were the following:

SQLM101 – Principal

SQLM102 – Mirror

SQLW101 – Witness server and additional DB added for monitor logs

SQLW102 – New Witness server with additional DB added for monitor logs

At first, when I attempted this process, I was not very knowledgeable about SQL Server end points and I did not understand there were certain roles for each type of end point. Luckily for me I was able to reach out to Robert Davis (Blog | Twitter) who helped me understand this topic better and then I found this great article in Books Online sys.database_mirroring_endpoints (Transact-SQL)

I was then able to come up with this script to do a test for active end points and also see how they are configured:

–Check for SQL Endpoints

select * from sys.database_mirroring_endpoints

* Not all results are shown here (please note role_desc field)

The screen shot below shows you the end points in SQL Server Management Studio (SSMS) and the results of the script.

When I ran this script, I was very interested in was the role_desc field which shows me how this end point is configured, which is only for the Witness Role. It can’t be used for any other type, such as None, Partner or All. Then I ran some more queries to get more details on the end points.

–Check for end point connections

select * from sys.dm_db_mirroring_connections

SELECT e.name, e.protocol_desc, e.type_desc, e.role_desc, e.state_desc, t.port, e.is_encryption_enabled, e.encryption_algorithm_desc, e.connection_auth_desc

FROM sys.database_mirroring_endpoints e JOIN sys.tcp_endpoints t ON e.endpoint_id = t.endpoint_id

–Check for SQL Endpoints (simple view)

select * from sys.endpoints

–Check for SQL Endpoints (more options to see which ones are in the mirroring state)

select * from sys.database_mirroring where mirroring_state is not null

–to see who has granted Mirroring and the grantee

SELECT EP.name, SP.STATE, CONVERT(nvarchar(38), suser_name(SP.grantor_principal_id))
AS GRANTOR, SP.TYPE AS PERMISSION, CONVERT(nvarchar(46),suser_name(SP.grantee_principal_id))
AS GRANTEE

FROM sys.server_permissions SP , sys.endpoints EP

WHERE SP.major_id = EP.endpoint_id

ORDER BY Permission,grantor, grantee;

I ran this script (above) on all of the current SQL Server Mirror pairs and it let me know a couple of key things: the role of the end point and the encryption method. This would allow me to create a new end point on the new SQL server. I did also check that there was no current endpoints, as seen below:

Here is the script that I ran to create the new endpoint to match the others, please note the role is all:

/****** Object: Endpoint [Mirroring] Script Date: 08/21/2012 15:17:17 ******/

CREATE
ENDPOINT [Mirroring]

    AUTHORIZATION [SMATTIE\stangsct]

    STATE=STARTED

    AS TCP (LISTENER_PORT = 5022, LISTENER_IP = ALL)

    FOR DATA_MIRRORING (ROLE = ALL, AUTHENTICATION = WINDOWS NEGOTIATE, ENCRYPTION = REQUIRED ALGORITHM RC4)

GO

This would allow me to mirror on top of the existing mirror from the stand point of the new server. I still needed to modify the existing Witness server to move the role from “witness” to “all.” Again, Books Online was very valuable and helped me get the information I needed via this article: ALTER ENDPOINT (Transact-SQL)

I ran this script one section at a time so I could see a before and after:

–Check for SQL Endpoints

select * from sys.database_mirroring_endpoints

– Change Endpoint role

Alter Endpoint [Mirroring] For Database_Mirroring (Role = All);

–Check for SQL Endpoints

select * from sys.database_mirroring_endpoints

At this point, I was ready to proceed with copying the database backups and then start the mirror. (SQLW101 and SQLW102)

Then I restored the database on the new Witness server and made sure that it was set to NORECOVERY.

It should look like this when you restore the database:

Now, I went ahead and setup the mirror (using the wizard for this demo) Click Configure Security

It will now bring up the security wizard:

We are migrating the current witness; there is no need for a new witness server

Add the Principal (primary) server

Add the Mirror (secondary) server, by selecting Connect

Then log into the server and click Next

Choose your service account:

Click Finish

Click Close, if no errors are mentioned:

Now click Start Mirroring

If all was successful, then you will see this status
listed (below) next to the database:

After the database is caught up and mirrored, then you can fail over the database to the new server and then break the mirror between the 2 witness servers. The next step is to add the new server to the mirror pair by removing the old witness server and then added the new witness server. All that is left at that point is to modify your database connection strings for your application(s).

That is how you can build a mirror on top of an existing mirror.

As stated on Monday, I have started a series on different database types for various data programs. (Microsoft SQL Server, MY SQL and Access) My goal here is to document the different data types and categories that reflect basic information about them.

This is the Microsoft Access Data Types (for old school reasons)

As stated yesterday, I have started a series on different database types for various data programs. (Microsoft SQL Server, MY SQL and Access) My goal here is to document the different data types and categories that reflect basic information about them.

Here are the MySQL data types and the three main types are Text, Number, and Date/Time types:

*The integer types have an extra option called UNSIGNED. Normally, the integer goes from a negative to positive value. Adding the UNSIGNED attribute will move that range up so it starts at zero instead of a negative number.

*Even if DATETIME and TIMESTAMP return the same format, they work very differently. In an INSERT or UPDATE query, the TIMESTAMP automatically set itself to the current date and time. TIMESTAMP also accepts various formats, like YYYYMMDDHHMMSS, YYMMDDHHMMSS, YYYYMMDD, or YYMMDD.

I am going to start a series on different database types for various data programs. (Microsoft SQL Server, MY SQL and Access) My goal here is to document the different data types and categories that reflect basic information about them.

For more detailed information, please see this article Data Types (Transact-SQL)  and also Paul S. Randal did a great job as well with this article Inside the Storage Engine: Anatomy of a record

I am going to start with Microsoft SQL Server:

Please see red below

Please see red below

Please note: ntext , text, and image data types
will be removed in a future version of Microsoft SQL Server. Avoid using these data types in new development work, and plan to modify applications that currently use them. Use nvarchar(max), varchar(max), and varbinary(max) instead.

Source: SQL Server 2012 Books Online [Data Types (Transact-SQL)]

Added for clarity: 

Each database has a counter that is incremented for each insert or update operation that is performed on a table that contains a timestamp column within the database. This counter is the database timestamp. This tracks a relative time within a database, not an actual time that can be associated with a clock

SQL Server 2005 Books Online timestamp (Transact-SQL) 

Please note:

The timestamp syntax is deprecated. (It has been since SQL Server 2008 BOL) This feature will be removed in a future version of Microsoft SQL Server. Avoid using this feature in new development work, and plan to modify applications that currently use this feature.

SQL Server 2012 Books Online (rowversion (Transact-SQL))

Tonight at 8:30 PM EST, the number one ranked Fighting Irish of Notre Dame take on the defending champion and number two tanked Crimson Tide of Alabama. My Irish are the underdogs by 10 points, but I believe they will win tonight.

BCS Title Game Odds Favor Tide by 10.

As usual, the views and opinions expressed in this article are mine and do not necessary state or reflect those of Microsoft.

This is part two of this series and you can look at part one here to get caught up.

Today, we will talk about getting ready for a hiring loop, which is basically a three-step process consisting of a resume submission, a tech screen and interview.

Step 1 (Resume)

This is advice was originally intended for any company, but to allow me to be more creative, I am going to approach it as it you wanted to apply at Microsoft since that is a pretty recent experience in mind.

If you are ready to apply, I strongly suggest that you consider using a Microsoft email address, such as from Outlook or Hotmail. Most companies like to see loyalty for their products, for example think of how GM would feel if you should up in a Ford or Volvo wanting to be a GM mechanic. It may not be the message that you want to send sub-consciously. Again, as noted yesterday,  all teams are different and some could care less, but I would play it safe so it does not even become an issue down the road to determine a tie breaker between candidates. Let’s be honest here, this whole process is about winning points for a team to make a decision on deciding the best candidate. I am sure that all of us want to be delivering a great message to help us obtain our goal.

Before you submit your resume, take a moment to consider what you want it to say to a person evaluating where you would make a good candidate. If you are going for a contract or vendor role, then a traditional resume will work in most cases. This type of resume is where you list your jobs and a few bullet points that describe your functions. In fact most recruiters can only operate with that type of resume.

However, if you seek a career at Microsoft, then I recommend that you look at doing a Situation, Action and Results (SAR) type of resume that focuses more on your abilities with quantified results. In my opinion, this format allows you to quickly point out to a hiring decision maker that you are more than capable of their requests. A great place to start for some great interview tips is looking at this video from Brent Ozar called Resume Tune Up. They look at resumes from a manager’s perspective as they glance at a resume to determine if a candidate should be in the hiring discussion. I like the approach that Brent and his team take to show you a before and after effect with resumes. They start off with the resume as it was submitted and then show you how they would submit it after making some changes.

Resume Do’s and Don’ts

Do…

• Focus on writing about your accomplishments in quantifiable terms. For example, “worked as test lead on three major product launches” or “increased sales in my division by 125% during FY98.”
• Spend more time highlighting your relevant experience, as opposed to all previous work experience.
• If you include an objective, make it straightforward, concise, and related to the job you are seeking.
• Include your name, phone number, and e-mail name at the top of your resume. Managers or HR contacts may not have access to the Address Book when they need to reach you.
• Provide a chronological view of your background, including dates and positions held. This does not necessarily have to be the primary focus, but most hiring managers prefer that this information be easily accessible.
• List your educational background. This should be near the bottom for an experienced candidate.
• Be concise. Keep your resume to one or two pages.
• Use bullet points to describe your key skills, projects, and accomplishments. Ideally, keep each bullet point to one line.
• Provide sufficient white space so that key skills, projects and accomplishments can be easily identified without having to read the entire resume.
• Be clear as to how you used a technology if you list it. For example, “knowing VBScript” is not the same as “created a Web site using VBScript” and providing the URL.
• Spell out acronyms and explain code names. Not everyone is aware of all projects and their components.
• Print it out and have someone else proofread your hard-copy resume. The Spelling checker and grammar checker do not catch everything, and you see typos more readily on paper than you do on-screen.
• Have your resume available both as a Word document and in ASCII text.
• Please be sure to scan your resume for viruses.
• If you are changing job functions or trying to stretch into a position that is quite different from your current role, include a summary of accomplishments at the top of your resume.  This summary should outline 5 to 10 reasons why you should be interviewed for the position. This takes time, but is an essential part of a targeted job search.
• Indicate whether or not you are open to relocation.

Don’t…

• Send out a resume with typos, misspellings, or grammatical errors.
• Include salary or level information.
• Indulge in long, detailed paragraphs.
• Spend too much time on graphics or cool fonts.
• Include extraneous information such as, “references provided upon request” or personal information.
• Laundry lists of job duties, skills, or classes are not the best means to present your skills and background. Most hiring managers are not going to take the time to read through a long list of information.

SAR Resume

The SAR resume highlights your greatest accomplishments in relation to the position you are targeting.

Advantages

• Immediately highlights your strengths.
• Extremely flexible.
• Maximizes your chances of catching a reader’s interest.
• Easy to adapt to the targeted position without sacrificing quality.
• Permits you to display originality in your ideas and manner of presentation.
• Enables you to lead the reader in the direction in terms of specific skills and accomplishments.
• Permits you to describe yourself in better marketing terms.

Disadvantages

• A good SAR resume takes skill to put together.

This is an example of a traditional resume:

This is an example of a resume in the SAR format :

Step 2 (Tech screen)

Look at the job description and see what skills they are looking for as an applicant. Make note of those and before you beginning preparing, look back on your career. You really do not want to come into a Microsoft interview prepared to play trivial pursuit, instead you should come in with the mindset that you are a leader who a good deal of experience. For example, if someone asks you how you does DNS work, instead of just saying it does this and it does that… take a moment to say to yourself, “It depends, what am I being asked to solve?” Then as you remember real life experiences, then you could say, that you can explain how it works in a perfect world and how it works in a real world keeping in mind how TTL’s can impact your changes. (ELB traffic for the wrong host)

The point that I am making is that you answer should not only reflect what you know, but it should also highlight what you have learned from previous mistakes or over sights (like the different TTL’s example listed above) that will demonstrate both knowledge and experience. Another good thing to do is after you answer the question, take the extra step of talking about where you got that knowledge and point to it on your resume. For example, when I was at Amazon, I saw the TTL impact with ELB’s on a daily basis (before they made their current changes with the IP pool) and I will call that out and also some issues that I saw with this at Rackspace. This gives the interviewer a sense of where you been and what you learned, plus it gives you a both a chance to have a normal conversation. That is a key point in an interview that I feel many people miss when answering questions.

An interview to me should feel like a good first date. I am comfortable, plus feel like I am talking and learning about the other person. I also want to know if we want the same results and would want to be together. Most of all, I just want to enjoy the moment and be entertained with the conversation. I do not want to feel intimidated, abused or any other negative feeling. I want to leave there with some desire to talk with them again.

Step 3 (In person interview)

Your interview should be the same message as your tech screen, but here you have more of a chance to elaborate on concepts. It is real important to pay attention to the discussion and find ways to share knowledge of past experience and exhibit your passion. I would find ways to look expand a conversation to help the interview process get a good handle on me. For example, if they asked me what tools would I use to troubleshoot network connectivity… I would not just list the tools and be done with it. I would start off with a few tools, like ping/tracert, and then explain that I use them and what information it tells me. Then I explain a few scenarios where I used them in a previous position (or two) from my resume and where I ran into issues, like a network that drops ICMP traffic. (Where I had to use a TCP based version of those tools) Then I continue doing that with each tool. It tells the interviewer that I have knowledge, plus I know what to expect and that I also have regular experience with these tools. It opens the door to allow them to see how you think and what you have been doing.

Again, the goal of any conversation to keep it interesting so both parties want to contribute, plus it takes the awkwardness off both people. It gives that opportunity to talk naturally and not focus on words on a piece of paper or stay in question mode. A normal conversation tells people so much more than a multiple choice test.

I have been in interviews at many different places, where it did not go so well and I ended up spending a lot of time playing trivial pursuit or brain teaser questions. Needless to say, that did not leave either one of us in a position where we really wanted to talk with each other. It is much better at that point where both you and the hiring team are looking forward to their next interaction with you. Remember the 3 dates from my example yesterday? It is much better to leave knowing you have something to look forward too and be waiting to talk with them again.

As usual, I hope this information helps you wherever you decide to pursue a career.

Have you ever wanted to work at Microsoft Corporation, but was a little taken back by the process? One question that I have received a lot is what is it like to interview with Microsoft or what the environment is like or the sheer amount of contractors, etc. Some people have expressed interest in these opportunities and asked me on the best way to proceed.

Please note that the opinions and views expressed in this blog are those of the author and do not necessarily state or reflect those of Microsoft.

As usual, I pause before answering most questions (saying to myself “it depends“) by looking at what are you looking to achieve here? What is your background? Basically, I need to know what I am solving for you before answering that question. Dealing with those questions a few times with some friends, I thought I would give some of my experience, in case others may have a question or two. I will do a two-part series on this topic, which again is based on my personal experience and my own opinion. The first article os on what opportunities can be available, i.e. roles and the second blog will be more toward the “loop” process.

There are 4 types of roles offered:

FTE: Full Time Employee (aka Blue badge) is a vested employee who receives benefits and stock options from Microsoft. If you are seeking this avenue, you should have the focus of wanting a career and not trying to fulfill a position.

Intern: Normally for college students to allow them to get some experience and a feel for a career at Microsoft. They are not vested, but they enjoy a good number of perks like a FTE.

A dash (aka A-): Contractor who can work at Microsoft for a year (365 days) and then is terminated. They will need to take 100 days off before trying another contract position. This is a shorter window for you to leave a good impression. (Part of a measure to prevent future law suits)

Vendor (aka V-): Somewhat like the contractor, (mentioned above) except you could work in this role for many years. It is seen as an outside company that has obtained a statement of work to perform a role or task for Microsoft.

Both the contractor and vendor roles are also known as Contingent staff, which means that there is a risk that employment can end rather suddenly for various reasons. (Budget, fit, etc.) It should be noted that most times, Microsoft will only pay for contracts one year at a time to allow them flexibility in changing things to fit their needs. This is a very dynamic environment and you need to be able to deal with change easily to succeed here. Microsoft is not really a place where things are allowed to be static and never change. Things move at a good pace and there is tremendous opportunity to be a part of that.

Now that we have the roles covered, let’s take briefly about the environments. I want to stress that Microsoft is very complex and has a lot of moving parts. Keep in mind that this is the company that designed Active Directory  so it will be a very mature design. (Personally, I have yet to see another environment as massive)

Since this is Microsoft and they have a lot of partners, you will see most, if not all, hardware manufactures represented at Microsoft. (i.e. HP, Dell, Lenovo, EMC, NetAPP, etc.) Be prepared to see a lot of servers and network devices in action in a lot of different environments. You will see production, pre-production environments (called stage in most other companies that I have been in) and many test (or dev) environments. There is a lot of things to manage and the best way I can sum it up best is to be prepared to take a few months to get to know all the moving pieces. I have been at a lot of big companies in my career. From my memory, you could take all the infrastructure of Amazon, Dell and HP/Compaq combined and it would still be small in comparison to Microsoft. It can be a bit overwhelming as you get started to develop a feel for how big environments are configured. No worries, just remember that Rome was not built in a day, so allow yourself a few weeks to learn things.

So, what is the point of why I called out the hardware environment?
You should consider your background and how would your resume tell a hiring manager that you can handle these complex environments.

Let’s talk about the other (and more important) environment at Microsoft… the teams and people. A quick search on the web will show you that there are about 94,000 employees. It is a large number and as expected, there will be a very diverse background and culture of people. With that being said, this is the first company that I have joined where each team has the potential to be radically different. You could be in the same building and have everyone in the same department, but one side of the building will have much different personalities. Most companies of this size that I have been in (Amazon, Dell, HP, etc.) you may see different personalities, but for the most part you could generalize the type of person. For example, support personnel could be very friendly and follow very similar troubleshooting methods. Granted I am not saying everyone did the exact thing, but you could easily follow the logic. At Microsoft, it is hard to describe, but people are really different and you may be caught off guard by the way people may think about resolving issues.

The point that I am making is that every time is very unique and has different needs, so a title of Systems Engineer will not mean the same job responsibilities for teams. Some teams may look at a Systems engineer like a Windows Admin, where another team sees it as a Project Manager or Developer Ops. This can make it a challenge when you interview, since most job postings for System Engineer will have the same words for what they are seeking. Do not be alarmed or worried, just do your best to find a position that interests you and if things change if you make it to an interview, oh well. Not all interviews will mean offerings and keep your head up that things happen for a reason, plus things have a way of straighten themselves out over time.

A bit of advice, if you want to work at Microsoft and are having trouble getting a group’s attention or having difficulty with the tech screen or interview loop. My suggestion would be to look into going after an orange badge as a contractor. Yes, it is a risk and you have to respect the fact that you will be Contingent staff, which means that there is a risk that employment can end rather suddenly for various reasons. However, it will give you an opportunity to see what the environment is like and a chance to see if you would enjoy the team. This option will also allow you a chance to determine if your skill sets are enough to do the job or if you need to grow in other areas. Most importantly, it gives you a try before you buy approach and a lot of the contractor companies do also offer regular benefits.

If you are not able to take that route, then you should be prepared for some tough screening that will focus on both knowledge and experience. In most cases, you should count on a full day interviewing and have at least 5 different people talk with you to determine different aspects of the hiring process. During this process, they will look at basic things like fit, skill set, desire, etc. Keep in mind that all teams are different, so they will drill you in different ways, Some teams like using brain teasers and some like asking real tough questions to help them evaluate you.

Do not look at this process as having to be right or wrong in your answers, but instead pay attention to what they are asking you. A good answer in most situations is to ask a question to confirm what they are asking you to solve. You want to be able to demonstrate that you can comprehend and you get bonus points when you do the following:

• Think on your feet to solve problems on the fly that are new to you or ambiguous in nature
  
• Provide answers to simple question, but also elaborate more details and point to your resume where you performed that function
  
• How engaged are you in the conversion and do you look for opportunities to insert yourself in the discussion
  

If you go to an interview like it is a test, that is seeking right and wrong answers, then you may want to reconsider your approach. Taking the lead in answering questions is always good, but also adding other details like what you would use certain tools and in previous experiences is huge. Ask yourself this question, which candidate would you hire?

1. Answers all questions correctly in a swift manner, but we have 10 minutes at the end where we do not speak and leave the process earlier
   
2. Answer most questions correctly, but misses a few and we spend 10 minutes talking about other interests
   
3. Answers every question and relates it to previous experience and takes the time to take me over his career as we discuss certain technical skills that helps me paint a picture of their growth.
   

If you have a hard time choosing candidate C, then look at it from a dating perspective as seen below:

1. Answers all questions about themselves, but does not try to extend the conversation and seem to be more interested in the TV or menu than you
   
2. Answers your question and says a few things that do not seem to add up, but does make an effort to keep talking with you
   
3. Answers and stays highly engaged in your conversation and looks for a chance to start-up new topics with you
   

In the above examples, we have the first person, who is listening to you and just reacting to you, but is no making no effort to connect with you. Even though they answered all your questions does not mean they provided you with a good vibe. It would not surprise me if you had to leave early that they would not be more than willing to take that opportunity to leave and not have any plans to get back in touch with you. The second person was engaged in the conversation and tried to add to it, but it is clear that there is not a lot of common ground to keep the conversation slowing. It seems a bit awkward and not as comfortable. The last person was on the same page as us. They wanted to keep talking and had a lot of interests in us. If we had limited time to spend with only one person, most of us would pick the last person, because we felt more comfortable and we could not get enough of them. They left us wanting more, because we both had a desire to keep sharing the conversation.

If you want the job (or the date) the best thing you can do is leave the people involved wanting more from you as you walk out the door. You want people to look forward to hearing from you and the sooner they can fulfill that desire, the better it looks for you. It is always nice when we are being actively pursued instead of waiting for them to contact us back.

As usual, I hope this information helps you.

Hard to imagine that a web site (this one) that I decided to put together is now 1-year-old.

On December 6, 2011, I wrote my first article, which was called a Brief Introduction to Windows Azure and I was exposed to the world. After that first blog post, I continued to write many more articles with topics ranging from SQL to Windows to Cloud computing and even an article about the loss of life while celebrate the impact of a great man named Tom Phillips.

As I read that first article today, I am amazed at how much has changed this year. Please keep in mind that things have not changed only for me, but a lot has also happened in the computing world.

This time last year, I was only one year into the greater Seattle area and was an employee of Amazon Web Services. 4 months after my first article, I would take a risk by becoming a vendor (contractor) at Microsoft Corporation, instead of playing it safe by staying as a full-time employee with benefits. Today, I am at Microsoft as a blue badge (FTE or full-time employee) and am grateful for my past growth that has helped me get where I am today.

Looking at my first article, I see that some of that information is out of date now, i.e. prices, offerings, etc. That is the challenge with the “cloud” or IT in general, it is always in constant change and motion. One way of doing, installing or administrating a server has now been replaced with a better and more efficient manner.

Looking back this past year at technology, it is not hard to notice that Steve Jobs is no longer with us. I still remember some very powerful words of wisdom that he shared with Stanford graduates in 2005. It hit me when he passed away and again reminded me how precious this journey of life is meant to be fully embraced.

Here are his words that have been burned into my brain:

No one wants to die. Even people who want to go to heaven don’t want to die to get there. And yet death is the destination we all share. No one has ever escaped it. And that is as it should be, because Death is very likely the single best invention of Life. It is Life’s change agent. It clears out the old to make way for the new. Right now the new is you, but someday not too long from now, you will gradually become the old and be cleared away. Sorry to be so dramatic, but it is quite true.

Your time is limited, so don’t waste it living someone else’s life. Don’t be trapped by dogma — which is living with the results of other people’s thinking. Don’t let the noise of others’ opinions drown out your own inner voice. And most important, have the courage to follow your heart and intuition. They somehow already know what you truly want to become. Everything else is secondary.

Stay Hungry. Stay Foolish. And I have always wished that for myself. And now, as you graduate to begin anew, I wish that for you.

For more information on the full speech, please read the following article.

Most of the things in life that I have paid attention to this past year… has been all about change. Amazon’s offering in the cloud has changed to include dropping prices to help companies make a better investment to offering many more products or improving products to have more functionality. Again, nothing remain the way it was a year ago.

It is really encouraging to see things get better and be a part of this change to make a difference. It is amazing to see how no job or product is really ever done. I used to think that we could get one thing done and keep it that way to improve it, but that is not the case today. Today, we are seeing more people replace hardware out in 3-4 years, instead of 5-6. Software upgrades are way more common. Past 4 years has seen 3 versions of SQL Server, from 2008 to 2008 R2 to 2012. Database mirroring, which was introduced in SQL Server 2005 SP1, is now on the deprecation feature list to be replaced by a new feature in SQL Server 2012 called AlwaysOn!

So what in the world am I talking about it? Nothing really, I am just rambling and reflecting on how fast this past year went by, while also trying to figure out where the time went!

As I wind down my thoughts, I want to say how grateful for all of you that are either following my blog or used it to fix an issue. I want to say thank you for allowing me to give back to you as many others have given to me.

As the years come to a close, I wish all of us a great year in 2013!

Hello everyone and I hope all is going well with you. Some of you may have been wondering where I went or trying to figure out the reason for my MIA status during the past month and a half.  This article will solve some of that mystery.

A month ago I took a new job with Microsoft. I went from a vendor role that was more hands on with servers and SQL to a Microsoft blue badge or Full Time Employee (FTE) role as a Crisis Manager dealing with service outages. There were various reasons for doing so, but most of all it was a good fit and an exciting opportunity with an excellent team to be a Service Engineer.

I have really enjoyed being a part of the SQL community and contributing my little part. I still plan to blog about SQL Server and virtualization.

On another note… Go Irish! (College football BCS championship is 7 Jan 2013)

If you’ve been working with the Release Candidate (RC) version of SQL Server 2012 (formerly code-named “Denali“), you are likely aware that the final release is now available as an evaluation download here, or from MSDN with in the Standard, Business Intelligence, and Enterprise editions. If you want to perform a straight upgrade from the evaluation edition, it’s a relatively simple process. From the installation screen, select the Upgrade from a previous version of SQL Server option (even though it doesn’t list RC0, it is the same process as from say SQL 2008).

The reason, which I bring up this issue, is that I had a set of VM’s that were configured with AlwaysOn running the RC version and the version was expired and gave the error message below:

The reason for that error is that per the license agreement that you agreed to as part of the installation, the beta and RC time is limited to 180 days.

The good news is that you can get around it pretty easily. Run setup from your RTM version of SQL and then click the installation tab on the left and select the Upgrade option as seen below:

Select all of the appropriate options for your installation, and your server will be upgraded pretty much like you would say during a service pack.

Please note: However, if you want to move from the release candidate or beta version (RC) to one of the licensed editions, you could run into trouble. In my testing, what has happened is that parts of the installation will upgrade properly, but others will fail. This can be seen from the instances screen when upgrade is run for a second time.

Unfortunately, the “left behind” instance will not be upgraded with the other components, and now that this has been run, the Edition upgrade function will fail to work as well. In one of my cases, I needed to do a complete SQL uninstall and then perform a reinstall. (Which is always fun) The really good news is that the reinstall did work and the way to avoid this is to do the Edition upgrade first. If you’re unfamiliar with this, you can find it on the Maintenance tab of the SQL Server installation Center (as seen below)

Once you complete the edition upgrade, you can now go ahead and perform the standard upgrade as you would with the evaluation version.

Notes for VPC-to-VPC-with-OpenSWAN:
yum install openswan
chkconfig ipsec on

***********************************************************

Singapore Setup

Singapore OpenSWAN config:

# /etc/ipsec.conf – Openswan IPsec configuration file
#
# Manual: ipsec.conf.5
#
# Please place your own config files in /etc/ipsec.d/ ending in .conf

version 2.0 # conforms to second version of ipsec.conf specification

config setup
nat_traversal=yes
# we should exclude ourselves, but that’s dynamic.
# The other end should not be behind NAT anyway. If it is via port forward, avoid 10/8 that Amazon uses
virtual_private=%v4:10.0.0.0/8,%v4:192.168.0.0/16,%v4:172.16.0.0/16
# amazon kernels have no KLIPS support
protostack=netkey

conn singapore-tokyo
authby=secret
auto=start
type=tunnel
left=172.16.0.100
leftid=x.x.x.x(EIP singapore)
leftsubnet=172.16.0.0/16
right=y.y.y.y (EIP Tokyo)
rightsubnet=10.0.0.0/16
ike=aes256-sha1;modp2048
phase2=esp
phase2alg=aes256-sha1;modp2048

————-/etc/ipsec.secrets

#include /etc/ipsec.d/*.secrets
x.x.x.x y.y.y.y: PSK “mysecret488″

*********************************************

Tokyo Setup

# /etc/ipsec.conf – Openswan IPsec configuration file
#
# Manual: ipsec.conf.5
#
# Please place your own config files in /etc/ipsec.d/ ending in .conf

version 2.0 # conforms to second version of ipsec.conf specification

config setup
nat_traversal=yes
# we should exclude ourselves, but that’s dynamic.
# The other end should not be behind NAT anyway. If it is via port forward, avoid 10/8 that Amazon uses
virtual_private=%v4:10.0.0.0/8,%v4:192.168.0.0/16,%v4:172.16.0.0/16
# amazon kernels have no KLIPS support
protostack=netkey

conn singapore-tokyo
authby=secret
auto=start
type=tunnel
left=10.0.0.100
leftid=y.y.y.y (EIP Tokyo)
leftsubnet=10.0.0.0/16
right=x.x.x.x (EIP Singapore)
rightsubnet=172.16.0.0/16
ike=aes256-sha1;modp2048
phase2=esp
phase2alg=aes256-sha1;modp2048

—- /etc/ipsec.secrets

#include /etc/ipsec.d/*.secrets
y.y.y.y x.x.x.x: PSK “mysecret488″

****************************************************

service ipsec start

echo 1 > /proc/sys/net/ipv4/ip_forward (add to /etc/rc.local)
iptables –table nat –append POSTROUTING -s 172.16.0.0/16 –out-interface eth0 -j MASQUERADE (add to /etc/rc.local)
(For Singapore 172.16.0.100 is you want this to be a NAT instance for access to the internet.)

iptables –table nat –append POSTROUTING -s 10.0.0.0/16 –out-interface eth0 -j MASQUERADE
(For Tokyo 10.16.0.100 is you want this to be a NAT instance for access to the internet.)

route add -net 10.0.0.0 netmask 255.255.0.0 gw 172.16.0.100 (For Tokyo instance in the same subnet as OpenSWAN instance)

Be sure to set SRC/DST to disabled on OpenSWAN instance. otherwise no forwarding to backend instances. To do this, right click on instance and select “Change Source/Dest Check”.

https://smattie-download.s3.amazonaws.com/OpenSWAN-Notes.txt

https://smattie-download.s3.amazonaws.com/VPC-to-VPC-via-OpenSWAN.png

If you are like me, you might be testing new machines in a lab environment or making videos or blogs about new features. Virtualization has helped in numerous ways to facilitate this new learning. It helps to spin up new machines to test or validate these many different configurations. You also might be like me in that you are limited by the amount of software keys that you have access to activate Windows. Well, this post should help you moving forward.

By default, Windows Server 2008 R2 does not ask for a product key during setup and it will allow you to run 30 days for free before needing to be activated. However, you can continue to use it after a 30 day trial, but past that time, the system will eventually start to reboot every 2 hours or run with reduced functionalities the longer it runs without being activated. If you want to extend that trial period, there are currently two workarounds that will not require you to implement a valid product key.

“Rearm” Windows

This first method will allow you to extend the “evaluation” period by. The evaluation period is also known as the “activation grace” period. Type in the following from an elevated
command prompt:

slmgr.vbs –rearm

This will reset the activation expiration and the machine is good for another 30 days. You can do this up to 3 times and get 120 days free trial in total. For more information about this command, please see this article: Slmgr.vbs Options

NOTE: If you are using sysprep a lot with a base image and want to avoid this issue, please see this blog: Sysprep, SkipRearm, and Image Build Best Practices

Windows Server 2008 R2 Editions 180 day Evaluation Keys

The last option, which allows you with more flexibility, which I believe is the best way to proceed, is to enter a product key that will allow you to “evaluate” Windows 2008 R2 for 180 days. Here are the keys needed to allow you to run for free for 6 months:

Windows Server 2008 R2 Web = KBV3Q-DJ8W7-VPB64-V88KG-82C49

Windows Server 2008 R2 Standard = 4GGC4-9947F-FWFP3-78P6F-J9HDR

Windows Server 2008 R2 Enterprise = 7PJBC-63K3J-62TTK-XF46D-W3WMD

Windows Server 2008 R2 Datacenter = QX7TD-2CMJR-D7WWY-KVCYC-6D2YT 

When you are done, it will look like the following:

Hey everyone, sorry for the delay in writing new content, but a lot of stuff has been going on for me. For the meantime, I will just say that the change of everyday life and the challenges (or opportunities depending on your point of view) that it will present you as we do our best to get through it. Sometimes we are not prepared for them and it may overwhelm us at first.

With that being said, it is good to be back and I am excited to provide an update about this great script.

Today, I want to do 2 things… I would like to say thank you to Ola Hallengren for reaching out to me shortly after I wrote this article. He was very pleasant and polite as he asked a very simple question that started our dialogue:

Could I just ask one thing. Why is it that you prefer not to verify the backup files?

 Once he said that I kind of already knew where he was going and over the course of a few emails he got me back on track. Which now leads me to my second point and I am going to correct that earlier information and share with you what I learned.

First and foremost, it is highly recommended that you do a checksum on your backup as pointed out here. It is also recommended that you do a verify as well and that is what I am going to correct in this article. With that being said I am going to create a do over with this topic and will highlight the new recommendations.

As with the same with the last time, I want to talk a little bit more about configuring his script in hopes that you don’t just download it and click execute without making some changes.

Before even think about executing Ola’s maintenance script, I like to configure SQL server to turn on backup compression by default. This step is done to save disk space and I also turn on the advanced options of sp_configure: (Steps below)

– Change backup compression to be on

sp_configure ‘backup compression default’, 1 reconfigure

–Show advanced options 

EXEC sp_configure ‘show advanced option’, ’1′

RECONFIGURE

GO 

– Check settings

sp_configure

It will return these results (backup compression default) when you are done: (1 is on and 0 is off)

Before you begin modifying Ola’s scripts, I would suggest that you the SQL Server backup and Frequently Asked Questions sections. It will help you determine what to configure for your business needs. It is important to point out that the recommended order of the jobs is to do the index maintenance, then integrity check, and finally the backups. (I will point these out again)

So right off the bat, let’s look at the first area where we need to change in the script. The location of the backups and here is the default location: (Click on the image for a larger view)

I changed mine to reflect where I wanted to back up to a Z drive and you will need to ensure the modified path exists: (Click on the image for a larger view)

Then the next thing, I looked for was the retention period of backups and make sure that you meet your requirements. I noticed this comment in the SQL Server backup section “Specify the time, in hours, after which the backup files are deleted. If no time is specified, then no backup files are deleted. Backup files are deleted after each database is backed up and verified. Backup files are deleted only if the backup and verification of the database were successful.”

Here are the default settings: (Just search for checksum)

These settings are left alone and no changes are needed. (Different from original post)

The last thing I do is order the steps. Stated earlier, the recommended order of the jobs is to do the index maintenance, then integrity check, and finally the backups.

The default settings are seen here:

To change them follow these steps:

• Search (Control –F) for DatabaseBackup – SYSTEM_DATABASES – FULL
  

  
  

• Label the Index Job step 1
  
• Label your Integrity checks as the next 2 steps
  
• Label the other jobs accordingly
  

When you are done following his recommendations then it should look like this:

NOTE: The above 3 steps are done to help you configure the scheduling easier in SQL Agent jobs.

Please note that Ola makes these recommendations for frequency of running the jobs:

User databases:

• Full backup one day per week
• Differential backup all other days of the week
• Transaction log backup every hour
• Integrity check one day per week
• Index maintenance one day per week

System databases:

• Full backup every day
• Integrity check one day a week

Cleanup:

• sp_delete_backuphistory one day a week
• sp_purge_jobhistory one day a week
• CommandLog Cleanup one day a week
• Output File Cleanup one day a week

When you execute the script, it will create the jobs and as you will see later on, the numbering did in fact help us when we schedule the jobs, since it will be in the correct order:

Please note that all of these jobs are configured with no schedules, so you will need to create them. Here is how the jobs look:

The last step is to configure the jobs to run with either his recommendations or modifying those recommendations to fit your needs. I have swapped out all of my old scripts in favor of his scripts and I have been really impressed with the results.

The last thing to do is configure your backup retention for your current polices. Here is the default setting:

sqlcmd -E -S $(ESCAPE_SQUOTE(SRVR)) -d master -Q “EXECUTE [dbo].[DatabaseBackup] @Databases = ‘SYSTEM_DATABASES’, @Directory = N’Z:\Backups’, @BackupType = ‘FULL’, @Verify = ‘Y’, @CleanupTime = 24, @CheckSum = ‘Y’, @LogToTable = ‘Y’” -b

For a one week retention period, you will need to use 168, which is 7 *24.

sqlcmd -E -S $(ESCAPE_SQUOTE(SRVR)) -d master -Q “EXECUTE [dbo].[DatabaseBackup] @Databases = ‘SYSTEM_DATABASES’, @Directory = N’Z:\Backups’, @BackupType = ‘FULL’, @Verify = ‘Y’, @CleanupTime = 168, @CheckSum = ‘Y’, @LogToTable = ‘Y’” -b

That is all you need to do here to keep you 1 week worth of data and now you also get a verify of your backup as well. The mistake that I made earlier was reading this information and coming away with a different understanding: (Click on the image for a larger view)

I thought that meant that as long as you are not verifying the backup then it will not remove files, since it needed to do both a backup and verification. It was much simpler than that… all you need to do was modify the job and then set the cleanup time as desired. Lucky for me, Ola caught this and shared this bit of wisdom.

In general you should modify/add parameters in the jobs. You should not change the default values in the stored procedures.- Ola Hallengren

Backup and more importantly Restore operations are essential database maintenance tasks. I think Buck Woody said it best; we do not have a backup strategy as much as a restore strategy that defines our worth to an organization. You can watch his comment here Session1: Blitz! SQL Server Takeovers

Having a good backup and restore plan in place helps Database professionals to restore from any failure in a very efficient way. In order to recover from a failure, you will need to restore a set of backups in a logically correct and meaningful sequence. (I will also state an old wise tale… practice makes perfect, so it is recommended that you test your restore strategy as much as possible) SQL server provides a bunch of backup types (Full, Differential, Copy only, etc.), creating a correct recovery sequence for a given point in time can be a tricky task depending on your level of alertness. (i.e. Being woke up at 2 or 3 AM with the database on fire is not a very pleasant experience) Add to this the complexity of having multiple fork scenarios or the fact that there can be multiple correct recovery plans with vastly different runtime costs. The task of creating an optimal and correct recovery sequence can get very complicated real fast.

In order to address this problem, SQL Server Management Studio (SSMS) in SQL 2012 has designed a tool called the “Recovery Advisor“. Recovery Advisor helps in creating a correct and optimal restore sequence. I will highlight the features of recovery advisor that are going to make life easier during database restores.

In order to play around with recovery advisor, let us first create a database and then take some backups of this new database. The following script creates a database and then takes a full back up, then a differential backup and then a transaction log backup. It then restores the database to differential backup and then takes a transaction log back up again.

–Create Database

create database test_recovery_fork 

GO

–Create a Table in the Database

use test_recovery_fork 

GO

create table t (c int) 

GO

–Back up Database (dif)

BACKUP DATABASE [test_recovery_fork] 

TO DISK = N’C:\Backup\Test_recovery_fork\test_recovery_fork.bak’

WITH NOFORMAT, NAME = N’test_recovery_fork-Full Database Backup’, 
SKIP, NOREWIND, NOUNLOAD, STATS = 10 

GO

–Insert changes

insert into t values (1) 

–Back up Database 

BACKUP DATABASE [test_recovery_fork] 

TO DISK = N’C:\Backup\Test_recovery_fork\test_recovery_fork.dif’

WITH DIFFERENTIAL, NOFORMAT, NAME = N’test_recovery_fork-Differential Database Backup’,
SKIP, NOREWIND, NOUNLOAD, STATS = 10 

GO 

–Insert more changes 

insert into t values (2) 

–Wait for a minute and then back up the log  

BACKUP LOG [test_recovery_fork] 

TO DISK = N’C:\Backup\Test_recovery_fork\test_recovery_fork.trn’

WITH NOFORMAT, NAME = N’backup_ test_recovery_fork -Transaction Log Backup’,
SKIP, NOREWIND, NOUNLOAD, STATS = 10

GO

–Recover to the differential backup.

USE [master]  

ALTER DATABASE [test_recovery_fork] SET SINGLE_USER WITH ROLLBACK IMMEDIATE 

GO

RESTORE DATABASE [test_recovery_fork] FROM DISK = N’C:\Backup\Test_recovery_fork\test_recovery_fork.bak’

WITH FILE = 1, NORECOVERY, NOUNLOAD, STATS = 5 

GO

RESTORE DATABASE [test_recovery_fork] 

FROM DISK = N’C:\Backup\Test_recovery_fork\test_recovery_fork_dif.bak’

WITH FILE = 1, NOUNLOAD, STATS = 5  

GO

ALTER DATABASE [test_recovery_fork] SET MULTI_USER

GO 

–Insert more changes

use test_recovery_fork  

GO

insert into t values (3)

–Backup Log with new changes

BACKUP LOG [test_recovery_fork] 

TO DISK = N’C:\Backup\Test_recovery_fork\test_recovery_fork2.trn’ 

WITH NOFORMAT, NOINIT, NAME = N’test_recovery_fork-Transaction Log Backup’,

SKIP, NOREWIND, NOUNLOAD, STATS = 10 

GO

At this point, we now have 4 backups (1 Full, 1 Differential and 2 Log backups) for “fork scenario.” These backups form a basic fork configuration (where we have 2 possible point in time restores) as noted with the 2 different .trn transaction log backups.

To open recovery advisor, go to the database node “test_recovery_fork” in the Object Explorer, you can hit F7 or go to View and Select Object Explorer Details (as Seen Below)

Now in Object Explorer, you can right click, then go to tasks, go to restore and click on database. (See below)

This will open the Restore Database dialog:

By default, the recovery advisor tries to restore the database to the last backup taken. In this case it’s the second log backup. Look closely to the restore plan generated, which is to restore the full backup then the differential backup and finally a log backup.

Observe that the time of log backup (as seen below) is 1:41, i.e. this is the second log backup as the first log backup occurred at 1:39. (Seen above with the 4 backups taken) Thus, when we want to restore ‘test_recovery_fork‘ recovery advisor is (by default) giving us a correct restore sequence to the last backup taken (taking into consideration the restore that we had performed earlier).

In most scenarios, that might be the correct plan of action, but what if we do not want to restore to the last backup, but want to do a point in time restore? Well that’s exactly what the new ‘Timeline‘ feature will allow you to perform.

For example, let’s say that you want to restore to a specific point in time, like just after the differential backup, but before any of the log backups.

Click the timeline button of the restore dialog and it will open up the backup timeline dialog. The time line will show the different backups taken so far. In order to, give a specific time for database restore, check the ‘Specific date and time‘ radio button.

This will enable the date and time input boxes, along with the slider given below the timeline.

We can either directly choose the date/time that we want or we can use the slider to choose a location relative to the backups taken. Drag the slider to a point just after the differential backup (but before the first log backup).

Click OK.

Now we have a different restore plan that only involves the differential backup plan and nothing else. In this case, the transaction log backups were already included as a part of differential backup recovery and were ignored.

All in all, I found that the recovery advisor is a handy tool that is able to create optimal recovery plans for a given point in time. It makes the task of restoring a database less painful.

In an earlier article, I spoke about Ola Hallengren’s highly recommended database maintenance scripts. Today, I want to talk a little bit more about configuring his script in hopes that you don’t just download it and click execute without making some changes.

Before I even think about executing Ola’s maintenance script, I configure SQL server to turn on backup compression by default. This step is done to save disk space and I also turn on the advanced options of sp_configure: (Steps below)

– Change backup compression to be on

sp_configure ‘backup compression default’, 1 reconfigure

–Show advanced options 

EXEC sp_configure ‘show advanced option’, ’1′

RECONFIGURE

GO

– Check settings

sp_configure

It will return these results when you are done: (1 is on and 0 is off)

Before you begin modifying Ola’s scripts, I would suggest that you read the SQL Server backup and Frequently Asked Questions sections. It will help you determine what to configure for your business needs. It is important to point out that the recommended order of the jobs is to do the index maintenance, then integrity check, and finally the backups. (I will point these out again)

So right off the bat, let’s look at the first area where we need to make changes in the script. The location of the backups and here is the default location: (Click on image for larger view)

I changed mine to reflect where I wanted to back up to a Z drive and you will need to ensure the modified path exists: (Click on image for larger view)

Then the next thing, I looked for was the retention period of backups and make sure that you meet your requirements. I noticed this comment in the SQL Server backup section

“Specify the time, in hours, after which the backup files are deleted. If no time is specified, then no backup files are deleted. Backup files are deleted after each database is backed up and verified. Backup files are deleted only if the backup and verification of the database were successful.“

Here are the default settings: (Just search for “checksum” using CONTROL-F)

I made changes to the backup cleanup time and I turned off the verify command: (Keep in mind what was said earlier which can impact your retention period: “Backup files are deleted after each database is backed up and verified.“)

In this scenario, we can allow for a backup to be alive for a certain amount of time in hours and as long as we are not verifying the backup. For example, we can specify that we want a week for our retention period by selecting the Cleanup Time to be 168 hours (24 x 7) and then leaving Verify set to N. (As seen below)

The last thing I do is order the steps. As stated earlier, the recommended order of the jobs is to do the index maintenance, then integrity check, and finally the backups.

The default settings are seen here:

To change them follow these steps:

• Search (Control –F) for DatabaseBackup – SYSTEM_DATABASES – FULL
  

  
  

• Label the Index Job step 1
  
• Label your Integrity checks as the next 2 steps
  
• Label the other jobs accordingly
  

NOTE: The above 3 steps are done to help you configure the scheduling easier in SQL Agent jobs.

Please note that Ola makes these recommendations for frequency of running the jobs:

User databases:

• Full backup one day per week
• Differential backup all other days of the week
• Transaction log backup every hour
• Integrity check one day per week
• Index maintenance one day per week

System databases:

• Full backup every day
• Integrity check one day a week

Cleanup:

• sp_delete_backuphistory one day a week
• sp_purge_jobhistory one day a week
• CommandLog Cleanup one day a week
• Output File Cleanup one day a week

When you are done following his recommendations and my suggestions to number the steps, then it should look like this:

When you execute the script, it will create the jobs and as you will see later on, the numbering will help us in scheduling the jobs:

Please note that all of these jobs are configured with no schedules, so you will need to create one for them. Here is how the jobs look by default under schedules:

The last step is to configure the jobs to run with either his recommendations or modifying those recommendations to fit your needs. I have swapped out all of my old scripts in favor of his scripts and I have been really impressed with the results.

As usual, I hope this information helps you.

Recently a friend of mine asked me for help when he was not able to get a database out of restoring mode. When he tried to restore the database to the latest backup, he got this error message below:

Msg 3104, Level 16, State 1, Line 1

RESTORE cannot operate on database ‘VirtualManagerDB’ because it is configured for database mirroring. Use ALTER DATABASE to remove mirroring if you intend to restore the database.

Msg 3013, Level 16, State 1, Line 1

RESTORE DATABASE is terminating abnormally.

He even took a screenshot of this error when he tried using the GUI to recover from this error message. (This is the same error message as the text above)

As you can tell here, he was in a bit of a trouble here. He had a database that was not sync and therefore was not available to fulfill requests for data, since it is offline. To make matters worse, he decided to break the mirror and re-setup mirror, but that did not help either. Again, this issue remains, because the database was in need of recovering data from a backup and you need to resolve that issue before moving on to the next problem.

Side note: I recently learned this lesson and wanted to pass it along.

Before we start, I want to digress for a little vocabulary lesson. “Backup” is a noun. “Back up” is a verb. You don’t “backup a database.” You “back up a database”. When you “back up” a database, it creates a “backup” of the database -Robert L. Davis (Blog | Twitter)

So, I advised my co-worker to perform the following steps to resolve this issue on the principal partner:

• Break the mirror by typing:
  

  ALTER DATABASE [VirtualManagerDB] SET PARTNER OFF

• Restore your Database from your latest full backup by typing: 

RESTORE DATABASE [VirtualManagerDB] FROM
DISK = N’E:\MSSQL10.MSSQLSERVER\MSSQL\BAK\virtualmanagerdb.bak’

WITH FILE = 1, NORECOVERY, NOUNLOAD, STATS = 10

GO

• Restore your Database from your latest differential backup by typing:
  

  RESTORE DATABASE [VirtualManagerDB] FROM DISK = N’E:\MSSQL10.MSSQLSERVER\MSSQL\BAK\virtualmanagerdb.dif’

  WITH FILE = 1, NORECOVERY, NOUNLOAD, STATS = 10
  

  GO 

• Restore your transaction logs from your latest backups using norecovery by typing: (More than likely you will be playing back a multiple set of logs)
  

  RESTORE LOG [VirtualManagerDB] FROM DISK = N’E:\MSSQL10.MSSQLSERVER\MSSQL\BAK\virtman_2012-08-28-23TO70001.trn’

  WITH FILE = 1, NORECOVERY, NOUNLOAD, STATS = 10
  

  GO 

• Recover the database to bring it online by typing: 

  RESTORE DATABASE [VirtualManagerDB] WITH RECOVERY;
  

  GO

At this point, the database should be fully restored and back online, but you will still need to re-create the mirror to get your high availability back. You will need to perform these tasks on the mirror partner:

• Copy all of the backups (you just restored, full, dif, logs, etc) from the principal server to the mirror server
  
• Open SQL Server Management Studio (SSMS) and you will see mirror database is disconnected.
  
• Verify your current backup does restore your database properly by testing it on another system. (Very important to NOT assume it is working)
  
• Before proceeding to the next step, I would take a minute to review this blog from Robert L. Davis (Blog | Twitter) on SQLU DBA Week – Recovering Lost Data
  
• Robert talks about the ability to verify that your database restores properly and how it can be queried before you enable it for mirroring.
  
• Restore your Database from your latest full backup by typing:

RESTORE DATABASE [VirtualManagerDB] FROM DISK = N’E:\MSSQL10.MSSQLSERVER\MSSQL\BAK\virtualmanagerdb.bak’

WITH FILE = 1, NORECOVERY, NOUNLOAD, STATS = 10

GO

• Restore your Database from your latest differential backup by typing: 
  
  

  RESTORE DATABASE [VirtualManagerDB] FROM DISK = N’E:\MSSQL10.MSSQLSERVER\MSSQL\BAK\virtualmanagerdb.dif’
  

  WITH FILE = 1, NORECOVERY, NOUNLOAD, STATS = 10
  

  GO 

• Restore your transaction logs from your latest backups using norecovery by typing: (More than likely you will be playing back a multiple set of logs) 
  
  

  RESTORE LOG [VirtualManagerDB] FROM DISK = N’E:\MSSQL10.MSSQLSERVER\MSSQL\BAK\virtman_2012-08-28-23TO70001.trn’
  

  WITH FILE = 1, NORECOVERY, NOUNLOAD, STATS = 10
  

  GO
  

• Restore the final log file in “STANDBY mode” instead of specifying RECOVERY or NORECOVERY. (Example below)
  
  

  RESTORE LOG [VirtualManagerDB] FROM DISK = N’E:\MSSQL10.MSSQLSERVER\MSSQL\BAK\virtman_2012-08-28-23TO70001.trn’
  

  WITH FILE = 1, STANDBY, NOUNLOAD, STATS = 10
  

  GO

• This makes the database go into a read-only mode and you can query against it for testing.
  
• If all went well with your testing, then re-run that final log restore again with the NORECOVERY flag instead of the STANDBY flag to configure it for mirroring. 
  
  

  RESTORE LOG [VirtualManagerDB] FROM DISK = N’E:\MSSQL10.MSSQLSERVER\MSSQL\BAK\virtman_2012-08-28-23TO70001.trn’
  

  WITH FILE = 1, NORECOVERY, NOUNLOAD, STATS = 10
  

  GO

Now you are ready to re-create the mirror and get your system back online.

Every so often I see an article talking about TSQL Database backups and they give an example like the one below:

BACKUP DATABASE [AdventureWorks] 

TO DISK = N’C:\MSSQL11.MSSQLSERVER\MSSQL\Backup\AdventureWorks.bak’

WITH NOFORMAT, NOINIT, NAME = N’AdventureWorks-Full Database Backup’,

NOSKIP, REWIND, UNLOAD, STATS = 10 

GO

Or I may run into a co-worker or friend who passes along a SQL backup job and I see the exact same syntax. Heck, even I was doing it in my older scripts as well, but luckily for me, Robert Davis (Blog | Twitter) pointed the default syntax that SQL Backup uses. Of course he asked me why I was wasting lines of code, which made me scratch my head. I have confirmed this information with Books Online and today I would like to share those with you.

First off before we get too far into this topic, let me answer the question… What are the default TSQL backup syntax? 

They are listed here:

NOINIT, NOSKIP, NOFORMAT, NO_CHECKSUM, STOP_ON_ERROR, REWIND, UNLOAD

As you can see from the example with backup above, I just put in all the default values. So really, I am just typing to practice and wasting time out of my day. :p

So how can you confirm this information that I am sharing with you today? Good question and an excellent practice to make a habit.

You can open SQL Server Books Online (BOL) by going to the following:

• Start
  
• All Programs
  
• Documentation and Tutorials
  
• SQL Server Books Online
  

When then looks like this and click on Search:

For this search, just type the word backup and select the BACKUP (Transact-SQL)

Now Scroll down to Media Set Options and take note of the underlined syntax (as highlighted in purple boxes) These are your default syntax levels.

Want more proof? Click (at the top of the article in BOL) on Transact-SQL Syntax Conventions

Take note of this legend:

So, please pass along the word and let others know to stop typing a syntax that are already configured to be a default level. I am sure they would appreciate not having to double their efforts. J

I been meaning to blog about this great feature for some time. It was released earlier this year, like way back in March of 2012. It is a report of your SQL server’s health in a very quick method and it is very simple, plus easy to install.

Overview

The SQL Server 2012 Performance Dashboard Reports are Reporting Services report files designed to be used with the Custom Reports feature of SQL Server Management Studio. The reports allow a database administrator to quickly identify whether there is a current bottleneck on their system, and if a bottleneck is present, capture additional diagnostic data that may be necessary to resolve the problem.

Common performance problems that the dashboard reports may help to resolve include:

• CPU bottlenecks (and what queries are consuming the most CPU)
• IO bottlenecks (and what queries are performing the most IO)
• Index recommendations generated by the query optimizer (missing indexes)
• Blocking
• Latch contention

The information captured in the reports is retrieved from SQL Server’s dynamic management views. There is no additional tracing or data capture required, which means the information is always available and this is a very inexpensive means of monitoring your server.

Reporting Services is not required to be installed to use the Performance Dashboard Reports.

Installation

The SQL Server 2012 Performance Dashboard Report can be obtained here. This is a very easy install to perform and can be handled in these steps:

Save the file

Click Next

Accept the License

Fill out the Registration Information

Click Next

Click Install

It will provide a status

Click Finish

Configure the SQL Server Management Studio to browse this custom report

You are not quiet done yet with this installation. Please read the readme text file

Open the file location of %ProgramFiles(x86)%\Microsoft SQL Server\110\Tools\Performance Dashboard

Click on Setup.sql

Execute the query to build the report

Wait for the message Script completed!

Right-click on Databases, Select Reports, then Custom Reports

Browse to %ProgramFiles(x86)%\Microsoft SQL Server\110\Tools\Performance Dashboard and then Select performance_dashboard_main.rdl

Run Custom Report (Take note of the warning)

Here is your new dashboard

Here is where you should click to see a quick overview of your SQL Server

For those who are new to SQL and are not quite ready to query DMV or DMF, looking at wait stats, or gather perfmon counters then this tool can help you get a snapshot of your SQL server. It is a pretty helpful tool and can help most people who are responsible for managing a SQL server.

In an earlier article, I talked about Finding and then Creating the Start Menu for both Windows Server 2012 and Windows 8. In those earlier articles, I talked about Microsoft wanting to rid themselves of the Start Menu and clearly pointed out this one from CNET news.

Well, today I want to confirm that it appears that is not the case where Microsoft is trying to prevent people from hacking the Start Menu back in. I am using the final version of Windows 8 (from my TechNet subscription) and I was successfully able to install the third-party software from Stardock called Start 8 to give me the Start Menu. As you can see here is the new logon screen with a Seattle background:

This is the build of Windows 8 that I am using:

And here is my Start Menu (well technically Windows Orb):

Before leaving you, I do want to point out that it is very easy to navigate without installing Start8 using the new Metro UI in Windows 8. If you are using a Windows Keyboard and are attached directly to the computer (not remote) then you can just hit the Windows Logo (next to the space bar and by the ALT key) to pop up the new Start Menu. They have moved things around a bit, like shutting down and logging out, so I recommend that you take a look at my earlier article here that walks you through the new changes.

NOTE: The only time it may be a trouble moving around the new start menu is when you need to connect to the machine remotely and you need to depend on dragging your mouse to the top right or bottom right corner to expose the new menu.

If you are like me, you get annoyed at having to do the same thing over and over… especially, if it is tedious and goes on for hours, I am sorry, but I get bored very easily in that situation.

One thing I get to do a lot on a daily basis is log into many different machines. I do this to check on things, verify backups/jobs, plus if time permits to be proactive and look at the following good advice:

• Look for performance improvements because it’s the right thing to do even if there are not any performance issues

As you know, each machine is different and as time goes on there are certain tools you get used to having at your disposal. One of the minor things I like to use is the command prompt and I really like to be able to cut and paste out of it. By default the command line prompt is not configured to allow the QuickEdit feature.  To confirm if you can copy and paste from it, just open a command prompt and then right click at the top and select properties (Seen Below)

You will notice that QuickEdit Mode is unchecked by default and this will prevent you from copying from the command prompt. (Seen Below)

For a one-off scenario or for a few machines, it can be ok, to just right-click the command prompt each time and enable the QuickEdit mode. However, for more than 10 servers, it gets old fast and you may catch yourself swearing lightly under your breath… WTF, why couldn’t they not just make this the default setting.

I have been searching for a few months to find an easier way to do this and in the past found a few methods, but there were not as clean as this one. Lucky for me, today I came across a method to set this feature using PowerShell (Which by default does have QuickEdit Mode enabled) by typing this command:

• Set-ItemProperty -path “HKCU:\Console” -name QuickEdit -value 1

That is all you need to do and now you have QuickEdit Mode enabled for your command prompt and no more double clicking to get it.

I do want to point out that this PowerShell command will make the following changes to your registry, so that means you will need to be an Admin of the box if you want to execute this command. Here are the changes it will make:

Now, you can go about your business without being delayed and have your QuickEdit mode for your command prompt. Enjoy!

Before we begin, I want to provide a little background on this tool.

Database Engine Tuning Advisor (DTA) is a component of SQL Server Management Studio, which is an integral part of Microsoft SQL Server, the database management application. This tool is used for improving query performances by examining the way queries are processed and recommending enhancements by specifying indexes, views and partitioning on the database. It replaced the Index Tuning Advisor in Microsoft SQL Server 2000. The Database Engine Tuning Advisor is installed along with other components of Management Studio when you install SQL Server. In case it does not get installed, you can choose to add it after running the setup again. You will need to modify the Management Tools – Complete to get this feature installed.

Basically, it is used to configure a database so that queries will process more efficiently. It does this task by modifying indexed views, indexes and partitions, which can make queries perform faster because you are fine-tuning the physical design of the database.

This tool was introduced by Microsoft with SQL 2005. DTA has made significant advancements over the predecessor and the main purpose is much the same as the Index Tuning Wizard. DTA does seem to produce more reliable results and appears to be more efficient at doing it. Unfortunately, there is still the possibility of getting some bad recommendations from this tool, so please proceed with caution. And yes, the bigger issue is that there was blanket acceptance of all indexes. You can learn more about this tool here. This tool is available in Microsoft SQL Server 2005 and newer releases.

So, how do you use this tool? Good question.

Instructions:

Navigate to the Database Engine Tuning Advisor by clicking “Start,” “All Programs,” “Microsoft SQL Server 2008,” “Performance Tools,” then “Database Engine Tuning Advisor.”

Click “Connect” in the “Connect to Server” window.

1. Open SQL Server Management Studio (SSMS) and save a query. I grabbed one from a sample off Microsoft TechNet. (As seen below)

USE AdventureWorks2008R2;

GO

SELECT pp.FirstName, pp.LastName, e.NationalIDNumber

FROM HumanResources.Employee AS e WITH (INDEX(AK_Employee_NationalIDNumber))

JOIN Person.Person AS pp on e.BusinessEntityID = pp.BusinessEntityID

WHERE LastName =
‘Johnson’;

GO 

– Force a table scan by using INDEX = 0.

USE AdventureWorks2008R2;

GO

SELECT pp.LastName, pp.FirstName, e.JobTitle

FROM HumanResources.Employee AS e WITH (INDEX = 0)
JOIN Person.Person AS pp

ON e.BusinessEntityID = pp.BusinessEntityID

WHERE LastName =
‘Johnson’;

GO 

NOTE: You can download the Adventureworks Sample databases from here.

Follow these simple steps to do get your query analyzed:

• Click “Start”
• Go to “All Programs”
• Select “Microsoft SQL Server 2008 R2″
• Choose “SQL Server Management Studio”
• Click “New Query” to open the Query Editor.
• Paste the SQL script copied from Microsoft TechNet into the Query Editor, then save as “Test_DTA.sql” in a familiar directory.

There is also a short cut to this process. You can open your query in SSMS and right-click in the Window to select “Analyze Query in Database Engine Tuning Advisor” (seen below)

Then it will open in DTA as seen below:

Please see screenshot listed below for assistance with these steps:

• In the “Session name” field enter a unique name, such as “Session1.”
• Choose “File” as the type of Workload and then navigate to the “Test_DTA.sql” script you saved earlier.
• Choose “AdventureWorks 2008R2” as the “Database for Workload Analysis,
• Select it again in “Select Databases and tables to tune.”
• Check “Save tuning log.”
• Click the “Tuning Options” tab to review the defaults for future reference. (This will not be used at this time.)
• Navigate to the toolbar and click “Start Analysis.”

Wait for the analysis to complete, which you can observe on the “Progress” tab.

Review the results in “Recommendations” tab to view the “Index Recommendations” and “Partition Recommendations” columns. Carefully examine those indexes or partitions that DTA recommends that should be added or deleted.

Deselect any recommendations that you don’t want to keep. Click “Apply Recommendations” or “Save Recommendations” under “Actions” for changes that you want to implement.

Now, run the newly saved script in a test environment to verify results before running the new script in production.

There is also a good article on MSDN about this subject called Tutorial: Database Engine Tuning Advisor

As stated in an early article of mine, there is a great free book from Microsoft Press that talks in good depth about Windows Server 2012. Today, I wanted to share some of the information that I learned from the book. (Limits and new features)

Here is a simple comparison of the different limits between Windows Server 2008 R2, Windows Server 8 Beta and Windows Server 2012 Release Candidate 1:

NOTE: Please click on image for a larger view.

Another great bit of information is the Technet article called What’s New in Hyper-V which outlines the new changes. Cool things like the following:

Hyper-V Replica
Resource metering
Improved Storage migration (No more downtime)
Virtual Fibre Channel
Virtual NUMA

I have been meaning to talk about this topic for a while now and tonight I made sure that I put out the good news. Today, I want to introduce you to the SQL Server Maintenance Solution by Ola Hallengren which is a script that will create a series of scripts that will perform SQL agent jobs to run backups for transaction logs, full and differential for both system and user databases. If that was not cool enough, it also optimize indexes and perform integrity checks on all of your databases. These are highly polished scripts that in the US Army that would be considered high-speed, low-drag. (Definition #1: A high quality, efficient way to do something, performed with exceptional speed, but still simple and streamlined)

Designing or even maintaining scripts that you have acquired over the years can take a lot of effort. This may lead to you to have longer days as you have to update them to work with new features. I had a series of scripts and I was very happy with them, but as I grew more in my knowledge of SQL, I wanted more abilities added in my scripts. Then I started seeing more and more people I follow on blogs, tell me the same thing… use OLA’s scripts.

Many other people (only listing a few people here) have mentioned these scripts and also have highly recommend them:

SimpleTalk
Brad McGehee
Brent Ozar (Few times here)
Greg Low
SQL Authority

I am also going to recommend the saw the same thing… Please do yourself a favor, take a minute and go to his website, then download his script and test them out your self. I think you will be very impressed with how fast it installs and how fast these jobs run. He even looks for other back up programs besides using the default SQL server tool.

Another cool thing, these scripts support SQL 2012 and earlier versions as he documented  here.

Provisioned IOPS are a new EBS volume type designed to deliver predictable, high performance for I/O intensive workloads, such as database applications, that rely on consistent and fast response times.

One of the biggest complaints about Amazon Web Services has been disk performance. In order to get fast performance it has been necessary to use RAID 0 and spread writes across multiple EBS volumes. However, there is a problem with that approach, if one of those disks is having a negative performance impact; then the whole RAID group will perform poorly. If you lose a single disk in that RAID group, then there is a good chance that you have lost the entire logical volume. Many AWS customers have balked at the idea of risking instability for performance.

High I/O Instances

On 31 July 2012, AWS introduced High I/O Instances – these instances have two terabytes of local storage in the form of a solid state drive. (2 x SSD-based volumes each with 1024 GB of instance storage) Needless to say, these instances are fast and they have been benchmarked at up to 120,000 IOPS. However, they don’t provide quite the CPU throughput of other instance types currently provide, but the primary goal with this product is to address the perception that EBS storage isn’t fast enough in AWS.

There is one considerable downside, though – High I/O instances use instance store storage that’s local to each instance. This is commonly referred to as ephemeral storage and it is important to note that the storage disappears after an instance is stopped and started. There is no guarantee that your instance will start on the same server. After the instance starts back up, the local storage must be re-initialized to prevent your data being visible to another customer. High I/O instances are great for customers who need consistent, high random throughput storage access for databases that have resiliency across multiple servers. In short; high I/O instances are great for distributed databases.

The High I/O instances could be used with SQL Server by running an AlwaysOn Availability Group and creating asynchronous replicas on more durable storage using EBS volumes or by backing data up to S3 volumes. This approach carries the risk of data loss, however if you need high I/O throughput in AWS, then it’s one approach that you can take to achieve this goal.

Provisioned IOPS and EBS-optimized instances

Amazon released Provisioned IOPS EBS volumes, which is a more durable approach to high performance I/O storage. Regular EBS volumes can be expected to perform like a typical 7200 RPM drive: ~100 IOPS for random data access. In the past this is where RAID comes in – enterprise administrators created RAID volumes to spread I/O across multiple drives to achieve higher throughput. Unfortunately, this carries some risk of volume failure and again, one slow disk will make the entire RAID volume perform poorly.

Instead of resorting to RAID for performance, you can now provision drives and request to get a set performance level. Using the Provisioned IOPS feature, you can request space in AWS just like current EBS volumes, but now you can also specify guaranteed performance between 100 and up to 1000 IOPS. This volume is still over a 1Gbps ethernet connection and should be seen like network attached storage, so the total data throughput is limited, but actually you are able to get higher random I/O throughput with fewer drives. You can use this new feature on bigger instance types (m1.large, m1.xlarge, and m2.4xlarge) instances on any EBS volume, even the root volume (aka boot drives). You can launch selected Amazon EC2 instance types as “EBS-Optimized” instances. EBS-optimized instances deliver dedicated throughput between Amazon EC2 and Amazon EBS, with options between 500 Mbps and 1000 Mbps depending on the instance type used.

Things to Consider

You will want to know how fast your current storage performs and how much this might cost you to implement in AWS. There are a few things to measure, the first to look at is the theoretical maximum performance of your storage using SQLIO This will give you a fairly accurate idea about the performance of your storage in both MB/s and IOPS. You can combine this knowledge with information from Performance Monitor to gather actual usage metrics. This will help you determine if it is feasible to move into AWS.

The second and an even important question for most people is how much will it cost you. Provisioned IOPS cost $0.10 per Provisioned IOP per month. Keeping that in mind that 1,000 IOPS will cost $100 per month on top of a slightly higher cost per MB. At 100GB and 1,000 IOPS, an EBS volume will cost $112.5 per month. It’s expensive, but if you need guaranteed performance, it’s worth it.

You can get a more reliable estimate of your bill by using the AWS Simple Monthly Calculator.

Options

Customers looking for fast storage now have multiple options for deploying SQL server. The less risk averse can use provisioned IOPS instances for guaranteed performance on highly durable storage. DBAs who are more open to risk can use the High I/O instances to get high throughput using local storage. As stated before, there is more risk involved, but there are solutions to help you alleviate that risk.

To learn more about EBS Provisioned IOPS, see the Amazon EBS detail page.

I got some great news for you and it is straight from a Microsoft’s blog Windows Server 2012 final release timing and that news is that they have provided release dates to Windows Server 2012.

Release To Manufacturing (RTM): First week of August 2012

Release to General Public: 4 September 2012

These dates were also confirmed by another Microsoft TechNet blog Dell and Hyper-V for the smaller business

I am looking forward to this new release and it appears that many others are as well, since this product was downloaded so many times more than other release candidates in the past

With Virtualization gaining more traction in all phases of business, competitors are getting better with their products. Most Virtualization vendors offer some sort of free Virtualization software today, that is capable of handling a small to medium load.

With that being said, the free hypervisor space is a very important segment for IT pros getting started with virtualization. It allows organizations to give virtualization a try without having to invest with an incredible financial obligation. Furthermore, some IT shops are just fine with only running free virtualization tools in their production environments.

Last month at the Microsoft TechEd North America, details were laid out for Hyper-V Server 2012. Please note that this is the free hypervisor and it should not be confused with Windows Server 2012 with the Hyper-V role; they are officially different products and should be seen that way. Hyper-V Server 2012 will bring incredible scale to the free virtualization space; and, in fact, no other free offering comes close.

This is absolutely an incredible new free version of Hyper-V and it will support up to 4 TB of RAM on the host, individual virtual machines with 1 TB of RAM, and guest vCPU assignments up to 64 cores. Amazingly, that this free hypervisor really exceeds with what many revenue offerings is mind-boggling. The release candidate for Hyper-V Server 2012 is available under certain access levels from Microsoft from resources such as a TechNet subscription. The release candidate installs just like any version of Windows or Hyper-V Server that is available today, quick and painlessly.

After the release candidate is installed; the menu on the host console is materially unchanged from the previous versions of the free 2008R2 Hyper-V as shown below:

The capabilities and expectations of this free hypervisor are very high. Mix in those feelings with the free migration technologies that do not require shared storage, and it’s pretty clear that Microsoft is serious about its legacy in the virtualization space. My only hope is that more IT professionals take some time out of their schedules to check out these free release candidates to evaluate their future use. I am very impressed with the changes.

In my last post, I talked about finding the Start Menu in Windows Server 2012 (formerly code name Windows Server 8) since it was a very quick visual difference between previous versions of Windows. It takes some time adjusting to the new direction with the Metro UI, but it can be done.

I highly recommend that you get used to the new Metro UI Start Menu, as I am hearing in numerous circles that Microsoft is determined to get people used to the new interface. The feeling is that Microsoft is currently working on removing old legacy code so that no one will even be able to program or hack the Start Menu back into the product. I am not certain if this is true, but I did want to give you ample warning that these feature may be gone forever.

As I previously pointed out, you can read more about this decision here.

Here is what the new Start Menu looks like today in Windows 8/Server 2012:

Here is what the new Desktop looks like today in Windows 8/Server 2012:

Notice that the infamous Windows Orb (that was first introduced in Windows Vista) is missing! No worries, you can still get it back using a third party software from Stardock called Start8. When you click the link, you will see the page listed below:

Click on Download and it will ask you for an email address so it can email you the link:

This is what your Email should look like and you will see the download link:

Then you  will have around a 2 MB file to install:

If you have User Access Control turned off (which it is by default) then say yes

Then it will launch into the installation with the Welcome screen as seen below:

Accept the license agreement to continue with the installation:

You can accept the default or change the installation folder:

Before you can hit finish, it will ask you if you want to login and automatically go to the Desktop. (Choose accordingly to your desires)

Notice the familiar Windows Orb is back!

Right click on the Orb and you will bring up the options page, to include the ability to shutdown or launch the Run command.

As stated earlier in this article and in yesterday’s as well, I would not count on this feature staying in and recommend that you get used to the new Metro UI instead. I really believe that Microsoft will ensure that this option will not exist.

As I stated in my last post, I am going to do a series of the new features in Server 2012. Today, I am going to start with the most obvious change… the start menu that has died and is no longer present. Rumor mills have it that Microsoft is determined to leave it out, by removing all old legacy code that would allow you to even hack it back.

However for today, you can download a new tool called Start8 from Stardock that will put the Windows orb back in and allow you to have your start button back.

Soon after the installation of Windows Server 2012 RC1, one thing you definitely note is the absence of the good old Start Menu that we have been using since 1995. (Release of Windows 95)

To find the Start Menu with 2012, just press the Windows Key on your keyboard or move (actually hover and it might take a few tries to catch) your mouse to the lower left corner of the screen. (As seen below)

A small windows pops up and when you click that small window the Start Menu in tiles format (Metro UI) appears. The new Start Menu looks like you see below.

In the developer preview it was possible to switch back to the good old start menu layout by changing a group policy. Unfortunately that does not work anymore. I have not found the way yet, to get the old Start Menu back the way it appears in Windows Server 2008 R2 without installing software. For more information on why this was changed, please see this article: Why Microsoft murdered the Start button in Windows 8

To log off, simply click your account and you will be able to either log off or lock your session as seen below.

Although the new menu above looks fancy, I am not a big fan to have this feature on my servers and to be honest, I still prefer to have the old Start Menu, especially when going to Administrator tools.

Now that you have found the Start Menu, try to find the Shutdown/Restart buttons! Go ahead I dare you to find them on your own and without scratching your head. Just kidding, to find the Shutdown or Restart button move your mouse to the top right corner of the screen (or the lower right corner of the screen). Three icons on the right side of the screen pop-up as you see below.

The top button is the Search that you can use to find applications (PowerShell, SQL, Windows Explorer, etc.)

The middle button also provides access to the new Metro UI Start Menu

The bottom button is the Settings icon and that’s the one you need to do any shutdown or restarts! (You can also use command line tools)

Click on Power and it becomes clear what you need to do to either shutdown or restart. (Plus disconnect)

As I stated in an earlier post here, Windows 2012 Release candidate one was released on 31 May 2012. This product is focused on being designed with cloud computing in mind and a lot of features were built with that focus. This will not be an easy upgrade where you just slip in a disc and click next until you click finish. Now, this install will make you focus on your infrastructure and what are your business needs.

Hyper-V gets a steroid bump in performance and functionality. It now makes for a legitimate argument that can be won to deploy Windows Server 2012 as your choice for hypervisor. A lot of really well thought out features were added to help make management of the VM more obtainable to include resource management of network, memory and CPU.

To help you get up to speed, I highly recommend that you take a few hours to review these videos as they give you an overview of the new Server OS and some demonstrations. I will be doing a series of articles on Windows Server 2012 (formerly codenamed Windows Server 8) to highlight my experience.

In the meantime, enjoy these videos from Microsoft.

This article will show you how to create a new primary partition using Disk Management and Diskpart (preferred) in Windows from an empty unallocated partition or volume. This is in line with Disk Partition Alignment Best Practices for SQL Server article.

Partition Alignment in Windows Operating Systems

The way partition alignment works depends on the version of Windows being used and the version in which the partition alignment was created. The following sections describe how partition alignment works in Windows Server 2008, the Windows Vista® operating system, and Windows Server 2003 and earlier.

Windows Server 2008 and Windows Vista: New Partitions

In Windows Vista as well as Windows Server 2008, partition alignment is usually performed by default. The default for disks larger than 4 GB is 1 MB; the setting is configurable and is found in the registry at the following location:

HKLM\SYSTEM\CurrentControlSet\Services\VDS\Alignment

However, if OEM setups are delivered (for example, with recovery partitions), even fresh installations of Windows Server 2008 having partitions with undesirable partition starting offsets have been observed.

Whatever the operating system, confirm that new partitions are properly aligned.

Windows Server 2008: Pre-existing Partitions

New partitions on Windows Server 2008 are likely to be aligned. Yet partitions created on earlier versions of Windows and become associated with Windows Server 2008 maintain the properties under which they were created. That is, in the absence of partition alignment being explicitly performed, these partitions are not aligned.

Windows Server 2003 and Earlier

Partitions created on versions of Windows up to and including Windows Server 2003 by default are not aligned. Partition alignment must be explicitly performed.

With that being said let’s get back on track to creating new volumes.

A primary partition is a type of partition created on a hard drive that can have an operating system installed on it. A primary partition functions as though it were a physically separate hard drive. By default on a basic MBR disk, you can create a maximum of 4 Primary partitions, or 3 Primary partitions and 1 Extended partition with up to 128 logical volumes in the extended partition on a single hard disk.

NOTE: You must be logged on in an administrator account for these commands to work.

If you disable the Disk Defragmenter service, then you will get the error below when you try to do anything in Disk Management. If you get this error, then make sure that the Disk Defragmenter service is set to only Manual.

To fix this issue, open services.msc and right-click on the Disk Defragmenter service to go to properties. Change the startup type and seen below:

Open Server Manager, in the left pane under Storage, click on Disk Management. As you can see I have 2 disks that are not formatted or assigned a drive letter. The first step to creating a volume, you will need to initialize your disk as seen below:

In the middle pane, right-click on the empty unallocated partition or volume and click on New Simple Volume.

This will launch the Wizard. Click on the Next button

Type in how many MB (1 GB = 1024 MB) you want to use from the unallocated partition to create the new partition with, then click on the Next button.
NOTE: If you want to use all of the unallocated partition to create the new partition with, then type all of the maximum disk space shown for the unallocated partition. If you do not use all of the maximum disk space, then you can use the left over space from the unallocated partition to create another partition.

Select an available drive letter that you want to be assigned to the new partition in Computer, then click on the Next button.

Type in a volume label name that you want for the new partition, check the Perform a quick format box, and click on the Next button.

Click on the Finish button.

The new primary partition or volume has now been created with the assigned unallocated partition space.

If you are having been using the new ability to manage servers remotely using command line tools, then I am sure you have been using PowerShell. If you are new to command line tools, then this may seem strange to you, but trust me, Microsoft has been doing a great job making it easier to use and more powerful with each release. If you ever get stuck, then please try out the cmdlet (pronounced command let) Get-Help and in most cases it will clear things up for you.

One of the first things you may have run into is trying to execute an existing PowerShell script on a newly installed Windows server and getting an error, like this one:

scriptname.ps1 cannot be loaded because the execution of scripts is disabled on this system. Please see “get-help about_signing” for more details.

Here is the list of available PowerShell execution policies (please note the Default) with a brief description of each:

AllSigned: All scripts as well as configuration files must be signed by a trusted publisher. This includes script on the local as well as remote machines.

Default: Uses the default Restricted policy.

RemoteSigned: All scripts as well as configuration files downloaded via the Internet must be signed by a trusted publisher.

Restricted: As mentioned this is the PowerShell default and will not allow you to execute scripts or load configuration files.

Unrestricted: This allows you to execute all scripts as well as load configuration files however if you download a script from the internet you are prompted to approve it before it can execute.

To change the PowerShell Execution Policy which you can execute any of the following above from within the PowerShell command window:

In my case, I tend to go with the Set-ExecutionPolicy RemoteSigned

You can also change the Execution Policy via the registry by navigating to the following:

HKLM\Software\Microsoft\PowerShell\1\ShellIds\Microsoft.PowerShell

Then locate the string value ExecutionPolicy and add any value from the list below:

Restricted
AllSigned
RemoteSigned
Unrestricted

For example if you want to set your PowerShell execution policy for RemoteSigned add the following:

HKLM\Software\Microsoft\PowerShell\1\ShellIds\Microsoft.PowerShell
“ExecutionPolicy”=”RemoteSigned”

You know, it is pretty amazing that over 50% of the servers in the world today are running as virtual machines. This means that they run on top of hypervisors like VMware’s vSphere, Microsoft’s Hyper-V, and Citrix’s XenServer. However, the need to protect data on those virtual machines or to recover individual files hasn’t changed from physical hosts. We will need the ability to recover for various and numerous reasons.

For the most part, the world of virtualization is still new to most Windows Admins and the pursuit for the best way to back up your virtual machines is still being debated in the community. Even though Microsoft has some official documentation on this topic, most people still scratch their head and say “Isn’t there an easier way?” As usual, the answer is.. “It depends!” Service level agreements, data retention requirements and customer expectations normally drive our success in this area.

Hyper-V Backup Tools Types

Anyone who has a backup tool in place today, will probably agree with me that most backup vendors will say, “You don’t need another tool, we can backup everything including Microsoft Hyper-V.” While that might be true for some vendors, I still recommend that you will first need to consider the various types of backup tools that are available to you and test which one works best for you. Currently you have 3 main types:

• Physical Server Backup Tools — Most traditional physical server backup tools (like Symantec Backup Exec, NetBackup, etc.) can back up virtual machines running in Hyper-V only with agents. This will act just as if they were physical servers. Recently, many of these backup tools have been adapted to communicate directly with the vSphere or Hyper-V servers to back up your virtual machines. While these tools weren’t originally designed as a virtualization backup strategy, they have proven to be a good option, especially for environments with a hybrid (mix of physical and virtual) approach and they give you more flexibility.
• Multi-Hypervisor Backup Tools — There are a number of tools for a virtualization-specific backup, which was originally designed to back up only the vSphere hypervisor, but they have now been adapted to back up multiple hypervisors platforms, such as Hyper-V and Citrix. For mixed environments, these have also been proven to be excellent options.
• New Hyper-V Only Backup Tools — Finally, there are a few select tools that were created and designed solely for the purpose of backing up Microsoft Hyper-V virtual machines. If you have one or more Hyper-V servers in your environment or if you just want to evaluate some new tools designed just for Hyper-V, these are what you should consider below.

Please note: As always, you must select the best tool for your particular backup needs and data retention policies! So test your ideas before going to production.

Microsoft Hyper-V Only Backup Tools – New and Improved

The last category, we discussed above is directed toward a Microsoft Hyper-V host and here is the list:

5nine Cloud Backup for Hyper-V – is available in both a commercial and a free version, is a backup program for Hyper-V. It was designed with service providers in mind and it has a fully scriptable PowerShell API interface to make it more ideal for them. (It also makes it a good choice for large enterprises as well) Like these other Hyper-V backup tools, Cloud Backup works at the host-level (not the guest) so that image-based backups can be done and will not impact the guest VMs. Microsoft VSS is used to ensure that backups have application integrity and so that these backups can be performed anytime without the requirement for a backup window, and “bare-metal” virtual machine restores can be performed as well.

Altaro Hyper-V Backup – which claims to back up a Microsoft Hyper-V host in just “5 clicks,” this tool will also use Microsoft VSS, plus it will offer you the ability to boot a VM from the backup, file level backup, and also allows for a live backup of Linux VMs. Here’s what it looks like:

 Hyperoo Hyper-V Backup – unlike the other two mentioned Hyper-V backup tools above, Hyperoo is a bit different. Basically it will back up your Hyper-V virtual machines to the Hyperoo cloud service. Instead of buying a backup program, you are buying a monthly license for each of your Hyper-V hosts.

Hey Everyone, I hope you all had a safe and wonderful 4th of July.

As you may or may not be aware, Microsoft has released SQL Server 2012 and on 31 May 2012 released Windows Server 2012 (RC1) Release Candidate One (which should mean Release To Manufacturing or RTM should be soon) I highly encourage all IT professionals to get up to speed on these products are big steps forward for both Microsoft and the community. These products were designed in mind and geared to be deployed with Cloud Computing in mind. Not just from a private cloud, where you deployed Virtual Machines for your own use, but also a public cloud, like Windows Azure or Amazon Web Services where external customers can use your infrastructure to deploy their own environments for “on-demand computing.”

SQL Server 2012 has a new feature called Always On and I noticed there are numerous conversations on forums, blogs, etc about it. However, from reading these free books, I see that a lot of people are not fully aware of what this new buzz word means and a lot of gems are not being discussed about it. There are a few different configurations for this feature and you should really look into them in a test environment. Here is a sample of the information from this free Microsoft Press books on SQL 2012:

Availability groups are built on top of Windows Failover Clustering and support both shared and nonshared storage. Depending on an organization’s RPO and RTO requirements, availability groups can use either an asynchronous-commit availability mode or a synchronous-commit availability mode to move data between primary and secondary replicas. Availability groups include built-in compression and encryption as well as support for file-stream replication and auto page repair. Failover between replicas is either automatic or manual.

When deploying AlwaysOn Availability Groups, your first step is to deploy a Windows Failover Cluster. This is completed by using the Failover Cluster Manager Snap-in within Windows Server 2008 R2. Once the Windows Failover Cluster is formed, the remainder of the Availability Groups configurations is completed in SQL Server Management Studio. When you use the Availability Groups wizards to configure your availability groups, SQL Server Management Studio automatically creates the appropriate services, applications, and resources in Failover Cluster Manager; hence, the deployment is much easier for database administrators who are not familiar with failover clustering.

Now that the fundamentals of the AlwaysOn Availability Groups have been laid down, the most natural question that follows is how an organization’s operations are enhanced with this feature. Unlike database mirroring, which supports only one secondary, AlwaysOn Availability Groups supports one primary replica and up to four secondary replicas. Availability groups can also contain more than one availability database. Equally appealing is the fact you can host more than one availability group within an implementation.

Windows Server 2012 really enhances it network stack, IIS and Hyper-V to make it possible to make this your operating system of choice for Cloud Computing. Here is a sample of the information from this free Microsoft Press books on Server 2012:

Virtualization can bring many benefits for businesses, including increased agility, greater flexibility, and improved cost efficiency. Combining virtualization with the infrastructure and tools needed to provision cloud applications and services brings even greater benefits for organizations that need to adapt and scale their infrastructure to meet the changing demands of today’s business environment. With its numerous improvements, Hyper-V in Windows Server 2012 provide the foundation for building private clouds that can use the benefits of cloud computing across the business units and geographical locations that typically make up today’s enterprises. By using Windows Server 2012, you can begin transitioning your organization’s datacenter environment toward an infrastructure as a service (IaaS) private cloud that can provide your business units with the “server instances on demand” capability that they need to be able to grow and respond to changing market conditions.

Hosting providers also can use Windows Server 2012 to build multi-tenant cloud infrastructures (both public and shared private clouds) that they can use to deliver cloud-based applications and services to customers. Features and tools included in Windows Server 2012 enable hosting providers to fully isolate customer networks from one another, deliver support for service level agreements

Without further delays and to help you get up to speed, Microsoft has released the following books:

Introducing Windows Server 2012 and Introducing SQL Server 2012

You can also find Kindle versions of these books at Amazon as well. (Links: SQL and Windows.)

Last but not least, here is a link for all free Microsoft Press Ebooks here.

Notes to go along with the video:
Add in html and apache (set apache to run on launch) index.html

HTML
HEAD
TITLE
A Small Hello from AWS
TITLE
HEAD
BODY
This is very minimal “Hello World” HTML document for the AutoScaling, ELB, and RDS in VPC.

BODY

chkconfig –level 345 httpd on
service httpd start
netstat -a (see if httpd is listing)

1. Do a Create Image (make an AMI)

2. Create Load Balancer production-lb in VPC

3. Commands used

as-create-launch-config MyLC –image-id ami-XXXXXXXX –instance-type m1.small –group sg-xxxxxxx –region us-west-2

as-create-auto-scaling-group MyGroup –launch-configuration MyLC –availability-zones “us-west-2b,us-west-2a” –min-size 2 –max-size 10 –vpc-zone-identifier
“subnet-XXXXXX,subnet-XXXXXXX” –default-cooldown 300 –load-balancers production-lb –region us-west-2

as-put-scaling-policy policy-scale-UP-on-CPU –auto-scaling-group MyGroup –type ChangeInCapacity –adjustment 2 –cooldown 300 –region us-west-2

mon-put-metric-alarm alarm-server-scale-UP –alarm-actions arn:aws:???????????????????? –metric-name CPUUtilization –unit Percent –namespace AWS/EC2 –statistic Average –dimensions=”AutoScalingGroupName=MyGroup” –period 60 –evaluation-periods 2 –threshold 60 –comparison-operator GreaterThanThreshold –region us-west-2

as-put-scaling-policy policy-scale-DOWN-on-CPU –auto-scaling-group MyGroup –type ChangeInCapacity “–adjustment=-2″ –region us-west-2

mon-put-metric-alarm alarm-server-scale-DOWN –alarm-actions arn:aws:??????????????????? –metric-name CPUUtilization –unit Percent –namespace AWS/EC2 –statistic Average –dimensions=”AutoScalingGroupName=MyGroup” –period 60 –evaluation-periods 5 –threshold 20 –comparison-operator LessThanThreshold –region us-west-2

as-update-auto-scaling-group MyGroup –min-size 0 –max-size 0 –region us-west-2

mon-describe-alarms –region us-west-2
mon-delete-alarms alarm-server-scale-UP –region us-west-2
mon-delete-alarms alarm-server-scale-DOWN –region us-west-2

as-describe-policies –region us-west-2
as-delete-policy policy-scale-UP-on-CPU –auto-scaling-group MyGroup –region us-west-2
as-delete-policy policy-scale-DOWN-on-CPU –auto-scaling-group MyGroup –region us-west-2

as-describe-auto-scaling-groups MyGroup –headers –region us-west-2
as-delete-auto-scaling-group MyGroup –region us-west-2
as-delete-launch-config MyLC –region us-west-2

In an earlier, post I talked about doing backups and also making sure they are being done with data that has been verified as healthy. We want to avoid the scenario of garbage in leads to garbage out. As stated in that earlier article:

Robert Davis (Blog | Twitter) provided me with some good advice:

You should be running DBCC CheckDB regularly. Daily if your system can handle it, but at least weekly. You don’t need to check for the last good DBCC CheckDB if you are running regular CheckDB. For backups, you should be using the CHECKSUM option of the backup command.

So how do we get there and what tool can we use? The answer is simple we can use the built in tool DBCC CHECKDB.

DBCC CHECKDB checks the logical and physical integrity of all the objects in the specified database. If DBCC CHECKDB is ran on a database please note that you do not need to run DBCC CHECKALLOC, DBCC CHECKTABLE, and DBCC CHECKCATALOG since DBCC CHECKDB includes all the three command. Usage of these built-in DBCC commands is listed below.

DBCC CHECKALLOC – Checks the consistency of disk space allocation structures for a specified database.

DBCC CHECKTABLE – Checks the integrity of all the pages and structures that make up the table or indexed view.

DBCC CHECKCATALOG – Checks for catalog consistency within the specified database. The database must be online.

Along with above three DBCC commands it also runs following two tasks to check the validity of the database both physical as well logical. If the DBCC check has returned any errors on a production database, the best solution is to RESTORE your DATABASE from BACKUP. If you’re unable to schedule a full DBCC CHECKDB in a maintenance window, you can add WITH CHECKSUM to your backup routines, and do full CHECKDB’s at a different time (SQL 2005 and later).

Note that a BACKUP [...] WITH CHECKSUM does not replace DBCC CHECKDB. Paul Randal has more details here.

How many times have we had to perform an inventory of either our equipment or software assets and then get asked to import that information into a database of some sort?

Recently a friend at work asked me for some assistance on creating such a database to inventory server hardware. Lucky for me they had most of the data fields input taken care of and to add to that luck, they were scanning bar codes and dumping the data into Microsoft Excel. I made a suggestion at that point to save the data as CSV files and then we can quickly input the data into Microsoft SQL Server.

In this article, I am going to walk you through the process of creating the database, importing the data from the CSV file and then adding more information from an additional CSV file. In this scenario, I am going to create a database called Inventory using TSQL command: CREATE DATABASE [Inventory]

Then I am going to locate my CSV files that I will use to demo this import:

Now, we can right-click on the newly created database, go to Tasks, then Import data:

It will generate this Wizard to walk you through the import:

Now, we can select the data source we wish to use, which in this case is a Flat File Source:

We will browse to the file

Then select file (ensure you either choose all files or CSV files)

You can also preview how the import will look as well before you actually do it:

Configure how you want to import the database, server and connections:

You can also change the table name as well, other wise it will take on the name of the file it is importing from:

Then you can run the package now to import:

Confirm your selections and then select finish and you are on your way:

Then you can verify the import in SQL Server, by expanding your database, then tables and then Select Top 1000 rows

As you can see the fields are the same as what we saw in the CSV file in Microsoft Excel. The last step is updating the inventory database from another scan, say we have another person using this method at another location and we want to update their data into our database. It is the same process as listed above, but you will need to check mark what is highlighted below:

Basically it will tell the import to skip the top column, which is used mostly as a header:

 And there you have it, an easy import of data into a new user database for you to keep track of your assets. As usual, I hope this information helps.

This file contents from this video can be download here. This zip file will have the latest AWS tools and other files needed that are shown in this video to successfully install the command line tools for Amazon Web Services.

Well, the answer to this question begins with every experienced DBA’s favorite answer… It depends…

A great place to start is with this MSDN article on Database Mirroring and Log Shipping (SQL Server). It is kind of like a questionnaire to determine what would be best and it explains this topic in fairly simple terms.

Before I begin, I do want to point out that you can do both mirroring and log shipping. Another point about mirroring is that the official recommendation by Microsoft is to only mirror a maximum of 10 databases per server because each mirroring session consumes 2 threads. Log shipping does not have this limitation. You can find more details/best practices with this KB Article and TechNet blog.

Database mirroring is a functionality in the SQL Server engine that will read from the transaction log and then copies those transactions from the principal server instance to the mirror server instance. Database mirroring can operate synchronously or asynchronously. If configured to operate synchronously, the transaction on the principal will not be committed until it is hardened to disk on the mirror. Database mirroring supports only one mirror for each principal database. Database mirroring also supports automatic failover if the principal database becomes unavailable. The mirror database is always offline in a recovering state, but you can create snapshots of the mirror database to provide read access for reporting, etc.

Log shipping is based on SQL Server Agent jobs that periodically take log backups of the primary database, copy the backup files to one or more secondary server instances, and restore the backups into the secondary database(s). Log shipping supports an unlimited number of secondaries for each primary database. It also can provide you with a more flexible solution to use a reporting server instance with older data.

In my experience, database mirroring is preferable to log shipping in most cases, although log shipping does have the following advantages:

1. Provides backup files as part of the process
2. Multiple secondaries are supported
3. Possible to introduce a fixed delay when applying logs to allow the secondary to be used for recovering from user error or data corruption

More information about both technologies is available in SQL Server Books Online in the topics “Log Shipping Overview” and “Overview of Database Mirroring“.

I have created a simple spreadsheet for both technologies to help you determine which one might be more suited for your needs:

Database Mirroring Chart:

Log Shipping Chart:

Last week, I was building out a 5 Node Windows 2008R2 cluster for use as a Hyper-V server.One of my nodes was giving me trouble. I was able to successfully install, configure and add it to the cluster, but when I started placing Virtual Machines it would lose access to the Cluster Shared Volumes. (CSV)

After that loss to the storage, it would put the VM in a stopped state with an invalid configuration. Due to a bad HBA driver install, I had to rebuild the node, but somehow one of the NIC interfaces got plugged into the wrong port on the switch and it was no longer on the correct VLAN, which happened to be the cluster heartbeat connection. At the time, I was not aware that this had happen. I just saw that I was having issues with a network connectivity when I reviewed the cluster validation report.

This time around, when I tried to verify the NIC settings, I was running into a challenge. I could not see the network switch to verify and was not able to see if we connected to the proper networks.

Lucky for me a quick search on the internet lead me to here. BGPing is a great PowerShellscript that will allow you to test numerous devices all at once and then output the results to a .csv file. (Comma-separated values)

It is pretty simple to run. First you need to configure the script to use a host file and an output file:

Then you need to either add IP addresses or Server names. In my case, I wanted to ping all 5 networks to all 5 nodes, so I entered in all of the IP addresses:

Now, you are ready to run the script:

And now you will see the output file as seen below:

With these contents:

Of course, this timed out when I was at home since I am on a different network scheme. However, at work I was able to see that the one node I was having issues with was unable to ping any of the nodes on the cluster heartbeat. Plus all of the other nodes could not ping this problem node on the cluster network, so it became real clear that this node had a NIC connected into the wrong port on the switch. I then was able to advise the network team what actions to take to resolve the issue.

In my last article, I talked about COPY_Only backups and I very briefly mentioned Recovery Models for SQL Server. I will one do do a more detailed post when I find some time, but I did want to share this little tip with you about the impact the Model database can have on your databases.

This week I was setting up some database mirroring for a few environments using some setup scripts. It was failing and at first, I was puzzled by the result. I checked the script and I could see the databases get backed up, then get copied over to the secondary server to get properly restored. However, it would never get mirrored and I knew on most of other servers how we were configured, so I made a an oversight. As I was talking to myself trying to figure out what was happening, my co-worker asked me what was troubling me. I showed him the process I was doing  and he was also surprised that it would not mirror.

Then I was talking with Argenis Fernandez (Blog | Twitter) over Lync who reminded me of the impact the Model (system) database can cause. So let me show the impact

Lets check the Model database, which the default setting for all system databases is simple recovery model as you can see here:

As you might be aware, this setting will prevent you from doing any SQL high availability options like log shipping or database mirroring, plus you will lose the ability to restore to a specific point in time. So what happens if we create a database with the Model database setup like this one?

As you can see we created a new database called testwith_simple so lets check the properties:

As you can see it took the same setting as the Model database. So what happens if we change Model’s database on the fly without any restarts!

Now that we set the Model database to Full, let’s create a new database:

As you can see we created a new database called testwith_full so lets check the properties:

As you can see it took the same setting as the Model database again, which is the Full Recovery model. As usual, I hope this article helps.

One of the great books, I have read this year and have really enjoyed has been Rod Colledge’s (Blog | Twitter) SQL Server 2008 Administration in Action. I have really enjoyed this book and for being technical it was really well laid out, plus easy to follow. My copy has been thoroughly worn and marked up with notes. If you are  wanting to be a SQL Server DBA or you want to improve your skills, I highly recommend this book.

Today I am going to cover one of his Chapter 10 topics about backups, specifically Copy_Only.

First off we need to briefly cover that there are three different recovery models for SQL Server: Simple, Full and bulk logged.

As you maybe aware, each database has a transaction log and it is unique for each database. The head of a log chain is a full database backup and each transaction log backup forms a part of this chain. To restore a database to a point of time, you need to have a database in Full Recovery mode (or Bulk logged) the restore is based upon a requirement to have an unbroken chain of transaction log backups. In this recovery model, you will need the correct log sequence numbers and have them “played back” in order. If you tried to do them out of the LSN order, you will get an error like the message below:

Msg 4305, Level 16, State 1, Line 1
The log in this backup set begins at LSN 24100000002283000001, which is too recent to apply to the database.
An earlier log backup that includes LSN 24100000002236000001 can be restored.
Msg 3013, Level 16, State 1, Line 1
RESTORE LOG is terminating abnormally.

So what do you do in the situation where you have  current database backup rotation but you want to back up a database to restore on another server? For example, let’s say that a developer is designing a solution and wants to take the current database and restore it to his workstation. He does this without telling anyone, but when the next SQL Server restore happens from a differential it will fail due to this backup change. Sadly, that is when you find out what happened. Why did this happen? Well, a SQL Server differential backup uses a Differential Changed Map (DCM) to track which extents have changed since the last full backup.

You might be familiar with tape backup types: Differential backs up any data that has changed since the last full backup and leaves the archive bit alone.  Incremental backs up any data that has changed since the last full or incremental backup and resets the archive bit. SQL Server for the most part, the backups act like a Incremental backup where it resets the archive bit. With the Copy_Only backup, it kind of acts like a Differential backup.

In my example above with the developer when they did that backup, it reset the DCM to relate to the developer’s backup instead of what we have configured as either a maintenance plan or a SQL Agent job.What makes it ugly, is if we ask the developer for the backup he took and sadly find out that he deleted it to save space on his workstation. So that means we will have a broken transaction log with missing LSN’s which will not allow us to do a point in time restore during this time period. (We also pray that no inserts bad code or that we need to restore to this point in time)

To use this feature and allow others to take a copy of the bases for testing, all you need to do is type this command: (Click on the image below to see it in full size)

Here is the command as well in text:

–Perform a COPY ONLY Database backup
BACKUP DATABASE [AdventureWorks] TO DISK = N’Z:\Backups\SQLServer\DB\AdventureWorks\AdventureWorks_20120601.BAK’
WITH COPY_ONLY, NOFORMAT, NOINIT, NAME = N’AdventureWorks-Full Database Backup’, SKIP, NOREWIND, NOUNLOAD, COMPRESSION, STATS = 10, CHECKSUM
GO

In an earlier article, I wrote about a few new tools for SQL Admins that were released by Microsoft.

As I have done in the past I will correct myself to ensure that I am passing along helpful information. This article is another example and I would like to explain my thoughts. When I wrote that article a few weeks ago, I did a few quick simple tests using that calculator and it seemed to be simple to use with good results. What I did not explore was testing for larger and more complex configurations, like Jonathan Kehayias (Blog | Twitter) did with his test. As you will see in his article, using the calculator made some really bad recommendations. I highly recommend three things moving forward here:

- Read Jonathan Kehayias‘ book. (Troubleshooting SQL Server: A Guide for the Accidental DBA)
- Read Jonathan Kehayias‘ blogs on SQL Server Memory. He has two excellent ones. (Here and more importantly this one as well)
- Do not use this calculator to configure your SQL Server’s memory configuration.

I agree with this blog (Wow… An online calculator to misconfigure your SQL Server memory!) in that this is an employee of Microsoft and he made a bad recommendation that can really hurt customers. We all make mistakes and I have made a few recently as well. However, the best thing to do to own up to your mistake and take corrective action. In my opinion I do not think Sakthivel Chidambaram is trying to correct the information with his tool here. He has to remember that he has identified himself as an employee of Microsoft and is also using a MSDN blog to present information. Most people, to include me, will take his word at face value and make a bad assumption that this tool has been thoroughly tested. The truth is that this information was not validated before being published. (Left out a lot of key points to consider, like installed drivers, like host bus adapters (HBAs) or FusionIO storage, monitoring agents that require memory, etc.)

Sure he made a disclaimer (later on after Brent Ozar and Jonathan Kehayias pointed out his errors on 24 May 2012) and changed his stance when more people found out about his tool (Excel spreadsheet) and pointed out the errors with his information. The problem is that it took me a few weeks to learn that information was incorrect and these disclaimers were not listed in his original article:

Few things to take care before you use this calculator:

IMPORTANT: This calculator is designed to help Novice SQL Database Administrators or someone who manages SQL Server but is really not an expert in SQL Server. So to put it simple, this calculator is meant for SQL Server professionals with Level 100 knowledge in SQL Server

Based on my support experience, lot of performance & memory related problems are because customer’s don’t set a Max. Server Memory and they don’t do that because they don’t know what to set. This is my personal interest to help those people who are in need but confused because there are so many recommendations around how to calculate & set this value.

I am sorry for my earlier recommendation and I stand corrected, plus this reminds of a trait that I have learned… test everything thoroughly.

It’s time dissect the two main types of flash chips in order to understand why not all SSDs are created equal. What is after all the physical difference between SLC and MLC?

SLC stands for Single Level Cell and just like the name suggests can store one bit per NAND gate hence SLC cell has two states: (0 or 1) based on the charge of the NAND gate.

MLC on the other hand stands for Multi-Level Cell and uses multiple voltage threshold levels in order to store 2 or even 3 bits (also called TLC – Triple-Level Cell) in the same NAND gate. This is done by coding 4 or even 8 states (in the case of 3 bit TLC) on the same gate so the MLC will typically one of the following states : (11, 10, 01, 00). The benefit over SLC is the increased capacity per chip (2 or 3 times more) but at the same time the voltage reference levels are a lot tighter which leads to more rapid degradation of the cell after a lot of P/E (Program/Erase) Cycles. Once the MLC NAND gate has degraded the reads are no longer predictable because the stored value  overlaps reference  levels. In this case the memory will report an error or if the controller supports it will retire the cell and replace it with one from the reserve capacity.

Typical number of write cycles is pretty solid around 100K for SLC and floats around 10K for MLC (different dies can have very different quality and will wear differently). This number is still high enough for a consumer life-cycle in the case of  MLC if the entire memory is programmed 5 times  daily for 5 years and  runs up to 50 years for SLC under the same usage.

In the case of MLC the program cycle take 2 or 3 times more than for SLC since the  programming signal has to be a lot more precise to code 4 states in the space of 2. This leads to higher speed and increased number of IOPS (IO Operations Per Second) for SLC type of memory compared to MLC.

Within the past 2 months, there has been a couple of key tools introduced for SQL Server. Recently on 20 April 2012, Microsoft released the Reporting Services Migration Tool and on 18 May 2012, they released Max. Server Memory calculator.

The Reporting Services Migration Tool (Beta) was mentioned here. I do want to point out that the source and target server must be at least SQL Server Reporting Services 2008 R2 or later. In the most simple term, this allows you to have a backup and restore tool for SQL Server Reporting Services. (SSRS) It does allow you to restore or even migrate your existing native report server via a PowerShell script.

You can download this tool here. The downloaded file MigrationTool.zip has the following components:

The tool comes in two flavors

1. Command-line tool (RSMigrationTool.exe):
RSMigrationTool.exe [-n ] -i [-h] -f [-p ]
-n Native report server url (e.g. http:///reportserver). Optional if native instance name is specified. You should specify this option if WMI does not return the correct url.
-i Name of the native report server instance. Migration tool will use WMI to connect to the remote instance. If the connection cannot be made, native report server url must be specified, and passwords and history snapshots will not be backed up.
-h Report history will be exported. Please note that migrating report history requires direct modification to data in report server catalog and is not publicly supported. Backing up the catalog before migration and verifying correctness of the reports after migration are strongly recommended.
-f A folder where the files are to be stored. The folder will be created if it doesn’t exist.
-p Password that will be used to encrypt password in data source and subscription definitions. If this option is not specified, passwords will not be backed up.

2. GUI tool (RSMigrationUI.exe):
You can either type in the parameters in the textboxes, or define your input in an xml file and click “Load Configuration”. A sample configuration xml file is provided (MigrationSettingsSample.xml).

3. Either tool will generate a PowerShell script (Migration.ps1) and a set of files that represents a backup of your server.   Usage of the restore script Migration.ps1:
PS E:\tmp\DenaliMigration> .\Migration.ps1 -TargetSiteUrl http://SharePointSite -Password -TargetFolderUrl “http://SharePointSite/Shared Documents” -TargetCatalogConnectionString “Data Source=;Initial Catalog=;Integrated Security=True”

4. Log files:

• Logs generated by the tool is located in the “Logs” folder under the same folder where the tool is located.
• Logs generated by the PowerShell script is located in the “Logs” folder under the folder for migration files.

I do not recommend using this calculator listed below and you can find out more here. (Updated 1 June 2012)

For the next tool, Sakthivel Chidambaram wrote a calculator after he noticed a lot of people were referencing his blog. (Importance of setting Max Server Memory in SQL Server and How to Set it) You can download this Calculator: here. It is essentially an Excel spreadsheet that allows you to enter values, as seen below:

Step 2 in the spreadsheet will require you to gather the results of a SQL Query:

Please note: This calculator will not help on server having more than 1 instance of SQL Server because the memory required for OS, SSIS, Antivirus etc., are shared so the calculation becomes little complex.

You can monitor the health of SQL Server Memory using these Performance monitor counters:

1. Page Life Expectancy Counter under SQL Server Buffer Manager which is the average number of seconds a page stays in cache. It should be atleast = 300 * (Total RAM in GB)/4. Though our recommendation earlier was PLE should be 300, if a page is staying just for 5 minutes on a server with 128 GB RAM is not a right value to say that there is no memory pressure.
2. Memory Grants Pending, in the Perfmon object SQL Server Memory Manager (Close to 0 or lower is better)
3. Memory grant queue waits in the Perfmon object SQL Server Wait Statistics Object (Close to 0 or lower is better)

In my earlier post, I talked about SSD drives and how they can use an algorithm called Static Data rotation to help prevent premature wear on the drive. Today, I am going to briefly cover the write operations of SSD/Flash Memory.

One of the limitations of flash memory (Which preceded SSD technology) is that while it can be read or programmed a byte or a word at a time in a random access fashion just like regular memory. (RAM)  It can only be erased one “block” at a time. This will set all bits in the block to 1 which is the default state for NAND memory.

Writing a byte to flash memory involves two steps: Program and Erase. (or also known as “P/E”) The block is written to a new cell and the old block needs to be erased.

The programing can be done at cell level (setting it to the “0″ state)  via a process called tunneling while the floating gate is being flooded with high voltage using the on-chip charge pumps.

Erasing can be done only on an entire block (resetting it to the “1″ state), through high negative voltage that is pulling the electrons off the floating gate via process called quantum tunneling. Flash memory is divided in erase segments (often called blocks or sectors).

As usual, I hope this helps…

For many years, RAM and Disk was and still is today, the main culprit for bottlenecks in performance. With the lower cost of Memory, the shift has fallen more on disks to be the slow point in systems. Large Enterprise companies would have numerous external storage devices to get around this issue. (SAN, DAS, NAS, etc.) For smaller companies with a considerable lighter budget, they had a tougher issue trying to resolve this challenge.

Recently (late 2009)  there has been a change to how hard drives works that has generated a lot of positive buzz. Solid-state drives (SSD) have been making a lot of headlines with their performance and throughput.

However, one of the main drawbacks of SSD  has been reliability. This leads to a common perception is that you should only use these type of drives for static content that will not have a lot of writes to the drive. This is more true in consumer products like home computers, but in enterprise  devices that is not always true. Most people are aware that every NAND cell has a certain prescribed number of Program/Erase (P/E) cycles and as data is being written to disk, chances are it will remain unchanged for weeks or months. This means that the cells that are being used to store that data will have the same wear level (used P/E cycles) for the weeks or months that data was unchanged. This can become a problem for the remaining free cells that are going to be taxed even more and could reach their end of life which will make the entire drive read only or even fail it completely.

Little did I know that SSD drive that I was testing was based on a chip-set that implements an intelligent Static Data Rotation algorithm as part of a set of technologies to increase the reliability of the drive. This means that the SSD controller  would actively rotate the static data  from cells intensively used to other cells that were least used during  idle periods  to allow the drives wear leveling to work  at it’s best. But what happens when you stress test a disk and you place about  3 times the size of the drive worth of data in a couple of hours while the drive is half full. This algorithm will kick in and start moving data around even when the drive is not idle and you will see a decrease in performance until the wear level is stabilized.

I discovered this technology while running a bunch of tests using SQLIO based on Jonathan Kehayias (Blog|Twitter) article about Parsing SQLIO Output to Excel Charts using Regex in PowerShell with a 6GB file and I got some good results. I started using the drive and installed a few Virtual Machines until the drive was half full. I kept running SQLIO and Crystal Disk Mark test sadly to see that the performance getting worse.

Essentially Static Data Rotation is a feature to make sure that you can use the drive for the Mean time between failures (MTTF) prescribed by the manufacturer and prevents premature wear on the cells that store hot data. Last but not least, a common question that comes up with this topic, “Is this algorithm (Static Data Rotation) a common feature across all SSD manufacturers?”

The answer is no, this is one of the strong selling points for the newer SSD controllers that implement Duraclass, which is trademarked to Sandforce. I would also like to point out that newer Sandforce controllers (which btw many manufacturers use this controller) have this technology as well but older ones do not have it. It would be nice to think that any Enterprise class controller would have its own implementation of  a Static Data Rotation algorithm, but sadly that is not always the case. As long as you get a Sandforce based or OCZ Octane or Vertex 2 and up SSD you are ok.

In yesterday’s post, I talk about using PowerShell to configure many NIC’s with NETSH. Today, I am going to cover doing it with PowerShell and make it simple where it will prompt for the information in a GUI.

So first off, you will need to ensure that you have Executaion Policy changed to allow for signed scripts to be run. (set-executionpolicy remotesigned)

Here is the script:

And now lets see it in action, but let’s check out our NIC settings before we begin:

As you can see I am just using DHCP, so now lets run the script:

And lets check on our settings for the NIC after the script completes:

As you can see that was pretty simple and got the job done nicely. I really like PowerShell and think Microsoft did an awesome job with it.

I had a task recently that required me to spin up a new environment for testing. I had about 20 new virtual machines that I needed to deploy and they all had static entries. As you may or may not know, the SCVMM console does not allow you to copy and paste like you can with RDP or Remote Desktop connections. Hyper-V does allow you to “text” a paste but it can take a while, which means I had a problem. I did not want to waste my time typing out each and every NIC setting for all 20 machines. I needed to deploy this solution faster and really find a way to limit the typing.

I decided that I would use netsh and PowerShell to accomplish this task. I am only using PowerShell to allow me to add multiple lines and to be upfront, I am not actually using any PowerShell  commands. With that being said, lets see what I did to help limit this time being consumed.

So, here you can see that I have a NIC named the default “Local Area Connection”

I am going to rename it to LAN (as you would rename a NIC when you have multiple NICs so that you can easily identify it) I just right click on it and select rename:

Just for sanity I am going to show you that I do not have any settings configured for this NIC:

No DNS is set either:

No WINS is set as well:

I know, WINS is being deprecated, but I know some people that still rely on WINS for some older applications.

Now, I will launch PowerShell and enter in these NETSH commands:

netsh interface ipv4 set address name=LAN source=static address=192.168.1.23 mask=255.255.255.0 gateway=192.168.1.1;netsh interface ip set dnsserver “LAN” static 192.168.1.6 primary;netsh interface ip add dnsserver name=LAN 192.168.1.34 index=2;netsh interface ip set winsserver “LAN” static 192.168.10.55;netsh interface ip add winsserver “LAN” 192.168.10.56 index=2

Notice there are multiple entries with a “;” separate them so that they all get run at the same time. I use the “Type Clipboard Text”

Here is the command that I used to change my NIC settings:

Now lets check the NIC settings after we ran this command to see what happened.

Let’s check the WINS entries:

As you can see, we are golden. The nice thing about this method is that I only typed this command once (in Notepad) and just pasted one time without having to type much.

As usual, I hope this helps you and if you have any other questions, please look at the new NETSH commands.

This is a well-known and often mentioned best practice that I have recently found out that is not implemented much. I was even surprised to see that most installer packages to even include Microsoft’s SQL Server install do not set this feature up. Instant Initialization has been around since Windows XP and SQL Server has taken advantage of it since SQL Server 2005, but it is not built-in or even turned on by default. In fact, the installer even with SQL Server 2012 does not even mention it. Which I am hoping by now, you are wondering.. what is he taking about?

I am taking about granting the account for the SQL Server service to “Perform volume maintenance tasks” with the local security policy. Allowing this access will allow SQL Server to take full advantage of the Instant Initialization feature of NTFS. This will reduce the time taken when initializing files during database creation, restores and auto-growth events. For example, if you have a new 20 GB Database file on 15k RPM drives and it will take the following:

Performance Test with Zero Initialization
Hardware: Dell Precision 670 Dual Proc (x64) with Dual Core, 4 GB Memory, RAID 1+0 array w/4-142 GB, 15000rpm disks
CREATE DATABASE with 20 GB Data file = 14:02 minutes
ALTER DATABASE BY 10 GB = 7:01 minutes
RESTORE 30 GB DATABASE (EMPTY Backup) = 21:07 minutes
RESTORE 30 GB DATABASE (11GB Backup) = 38:28 minutes

Performance Test with Instant Initialization
Hardware: Dell Precision 670 Dual Proc (x64) with Dual Core, 4 GB Memory, RAID 1+0 array w/4-142 GB, 15000rpm disks
CREATE DATABASE with 20 GB Data file = 1.3 seconds
ALTER DATABASE BY 10 GB = .4 seconds
RESTORE 30 GB DATABASE (EMPTY Backup) = 5 seconds
RESTORE 30 GB DATABASE (11GB Backup) = 19:42 minutes

For more detailed information, please see the following article on Kimberly Tripp’s blog. But as you can clearly see you are saving a considerable amount of time.

I am going to show you how simple it is to set up and more importantly, show you the low risk associated in allowing this access.

First you need to see the service that starts your SQL Server as seen below from either the services applet in the Control Panel or the preferred method SQL Server Configuration Manager:

SIDE NOTE: Managing the service account password is a lot easier in SQL Server 2005 with the SQL Server Configuration Manager (SQL-CM). The SQL-CM allows you to change the password to a service without an active connection (meaning even if the service isn’t started) and it invalidates the login token so that password changes don’t require a restart of the service. SQL-CM also has a command-line interface and is scriptable with WMI. The WMI Provider allows server settings, client and server network protocols, and aliases to be scripted through the WMI Provider by means of simple VBScript code (or by using the command-line tool). What you could end up doing is creating a script that changes the password of your services on all of your servers (for example, when a password policy is enforced that requires that the passwords of service accounts be changed). 

Quote is taken from same blog mentioned above on sqlskills.com.

To use instant initialization, your SQL Server service must be running with an account that has the required privilege. You can modify it by typing gpedit.msc in the run button:

Then you will need to navigate to Computer Configuration, Windows Settings, Security Settings, User rights and then Perform volume maintenace tasks as seen below:

As you can see, the default setting is only allowing this task for Administrators of the box:

Click on the Add User or Group radio button and then add your user:

After click ok, you now notice your account is listed and once granted, SQL Server automatically uses instant initialization:

Recently, I followed the steps in this TechNet article (Install Reporting Services SharePoint Mode as a Single Server Farm) to create a Reporting Services 2012 Service Application. However, when I try to create the SSRS service application, I get the following error: “The timer job completed, but failed on one or more machines in the farm.”

The farm includes only one SharePoint server, so I went looking through the ULS logs and noticed this:

Digging some more through the logs, I noticed a more interesting error, the server was not able to access a registry key:

So then I took a look at the permission of the key and notice that the account did not have write permissions, so I changed it to the following:

Sadly, it still did not work, so I looked at the install components of SQL Server (specifically Reporting services) and that is when I knew what the problem was causing me. I installed the reporting service add-in, but “Reporting Services – SharePoint” was missing. With the add-in only still you can install the service application, but it will throw the error as seen above. I just installed “Reporting Services – SharePoint” and checked the registry, it has the expected setting now.

As a suggestion, it should not be possible to install the service application when the “Reporting Services – SharePoint” is missing.