Carpe Datum

Help for SQL Server

Buck Woody — Tue, 09 Feb 2010 15:29:00 GMT

Sometimes issues pop up with your system. You need a way to find help, quickly. Here’s a few links that might be useful – feel free to post a reply to this post with other sources you might know, from web sources to your favorite consultant.

Note – this list is not exhaustive, and I’m happy to add or edit it if you post a reply. I know a lot of SQL Server professionals, and I don’t want to leave out anyone or commit them to something!

Web Help – Use these links to do a little research on your own

Main Troubleshooting Page:	http://technet.microsoft.com/en-us/sqlserver/bb895929.aspx
The main Microsoft page for learning resources is here:	http://www.microsoft.com/sqlserver/2008/en/us/learning.aspx
There are lots and lots of great community sites. Here are just a few:	http://www.sqlcommunity.com/ http://www.mvps.org/links.html#SqlServer http://www.sqlpass.org/ http://www.mssqltips.com/ http://sqlservercentral.com
serverfault.com (mind your manners in here and you’ll get some help!)	http://serverfault.com/
Post a feature request or a bug here:	https://connect.microsoft.com/site/sitehome.aspx/?siteid=68&wa=wsignin1.0

Phone Help – Use these Links to get someone to talk with

The main launching page for Microsoft Support is here – they can do most anything over the phone:

http://www.microsoft.com/microsoftservices/en/us/support.aspx

On-site Help – Use these links to get someone at your facility

Microsoft consulting services can either help you with a deep problem or help you	http://www.microsoft.com/microsoftservices/en/us/consulting.aspx
Many of the “Most Valuable Professionals” (MVP) for SQL Server either work at support and consulting firms or know people who do. They are a great resource:	http://www.mvps.org/links.html#SqlServer
Microsoft Partners (Use SQL Server as a search term and use the filter tool for services and your location):	http://www.microsoft.com/sqlserver/2008/en/us/find-partner.aspx
I use the web search of “SQL Server” and “Consulting” to locate other folks. You can add the name of your town (like “Seattle”) to narrow the range:	http://www.bing.com/search?q=SQL+Server+consulting&form=OSDSRC

Make it Easy for People to Help You

Buck Woody — Mon, 08 Feb 2010 15:17:59 GMT

OK, there are probably a dozen or more of these kinds of posts, but I’ll dive in anyway. From time to time, people send me e-mails or comments on this blog asking for help. Sometimes it’s on the topic at hand, and other times the topic just jogs their memory about something else.

Often I’m happy to help. If I know the answer without doing any research (or even if I have to do a little) I’ll interrupt what I’m doing and dash out a note with the answer. But of course I have a job (three, to be exact) and so any time I help with a question I’m lengthening my day, spending less time with my family, and so on. If you think about it, everyone that you ask (and everyone I ask) is in the same boat – when someone helps me, I am taking their most valuable asset: their time. So I learned a very important lesson very early on: Make it easy for people to help you.

So here’s the steps to follow to do that – it really isn’t that hard:

Request, don’t demand. I got an e-mail yesterday (on Sunday) where someone found a blog entry about one topic, and basically said this about another topic: “Tell me where I can find ‘x’ so that I can alter it.” Guess what I did? That’s right, I hit the delete key. If you are asking a question from a professional, you need to understand that they normally get paid – very well, sometimes – for their time. Make sure your question is a question, not a demand.
Be clear about the problem. Vague statements don’t help – and very few people have the time to dig the real question out of you. Be specific. Ask the single question you really need help with.
Keep the problem limited. “Rewrite my code for me” isn’t going to happen. “help me with this line” might. “Where do I go to find out more about the SELECT statement” is even better. If your problem takes more than a few minutes for someone to answer, then you should probably get someone on-site to help you.
Explain what you’ve already done. This, of course, means you’ve already actually done something. What have you looked up, what do you already understand, where have you looked, what have you tried?
Be polite. Please and Thank You are magic words, whether you get the answer you were looking for or not.

I’ll add one thing when you’re responding to a blog – not just mine, anyone’s. If you want to ask a question, ask it as a reply to a post, not an e-mail. The author wants to answer the question once, and it’s almost a guarantee that you’re not the only one with that question. Also, other readers might know the exact answer and help you even more. I know, you have to register, all that stuff. Just consider it the price of getting your answer.

Quote of the Day: On What Really Makes Something True

Buck Woody — Fri, 05 Feb 2010 15:02:02 GMT

“Whether you believe you can or you believe you can't, you're probably right.” - Henry Ford

Multi-Monitor Support in SQl Server Management Studio 2008

Buck Woody — Thu, 04 Feb 2010 15:33:00 GMT

Did you notice that there are a lot of "little" things that were improved in SQL Server 2008? One of those was done by one of my favorite developers and deals with multi-monitor support. In previous versions of SQL Server Management Studio you could open the tool in multi-monitor and arrange the panes onto mulitple monitors. But when you disconnected the second monitory (as is often done with a laptop) you would "lose" the panes that were moved - SSMS just wasn't aware of those multiple monitors. Vlad spent a lot of time making that not true.SSMS is much better about being aware of the other monitors.

Sure, there's more to do in this area. We'll keep improving that feature, but I wanted you to know that sometimes little things take a lot of work - and aren't always noticed.

Thanks, Vlad!

How Does Microsoft Do IT?

Buck Woody — Wed, 03 Feb 2010 15:17:24 GMT

Microsoft is a big company – and of course we have a lot of IT infrastructure that we have to manage. It might surprise you to learn that we have an IT group, just like at your company. We have a networking team, a server hardware team, software teams, DBA’s, the whole bit. In fact, we have more Mac computers than just about anyone (other than that company down south from here) and we write some of the best-selling Apple software. We have a Linux lab.

How do we do that? How do you manage 80,000+ seats, especially when most of your company are a bunch of tech-savvy geeks? It’s a tough job, but the neat thing is that we tell you how we’re doing it – everything – right here: http://technet.microsoft.com/en-us/library/bb687780.aspx. If you want to focus in on just SQL Server, just check here: http://technet.microsoft.com/en-us/library/bb687798.aspx.

(By the way - I *totally* should be doing our marketing – isn’t that title catchy? My catch-phrases and product names would be a lot better than what we normally come up with. I’m just sayin’.)

Data Design

Buck Woody — Tue, 02 Feb 2010 15:31:00 GMT

I give series of classes and presentations on Data Design. I say “data” design instead of “database” design because we should consider more than just the database. Data might actually be stored in non-relational stores, such as Excel or XML files, and it might also be located in remote data stores like “cloud” technologies.

Here are the links I use in that presentation. Although this isn't a comprehensive list of Data Design topics, I’ll visit this topic from time to time so you may want to bookmark this page in your favorites:

Evidence and Comments:

http://tinyurl.com/ygvbadc

http://tinyurl.com/ydeh9ut

http://tinyurl.com/ycu9apq

http://tinyurl.com/ya7a9xc

http://tinyurl.com/ydlumyp

Importance:

Yes, it matters: http://tinyurl.com/yb5ucvn

Classic software development mistakes: http://tinyurl.com/y9x9vn8

Necessity of Good Design: http://tinyurl.com/yac3um3

Importance of Database Design Ranked high for BA/Dev interaction: http://tinyurl.com/yex4sjf

Structured Methods:

Test-Driven Development: http://tinyurl.com/maus78

From the BA’s Perspective: http://tinyurl.com/yea6pbe

Data-Map Examples: http://tinyurl.com/2d2nhr

Universal Patterns for Data Modelling: http://tinyurl.com/ybj7tb3

Tools:

EF Framework Design Example: http://tinyurl.com/yeuda49

Using EF and LINQ can cause divisions: http://tinyurl.com/yaljmhn

Transparent Data Encryption and the Latest Data Breach

Buck Woody — Mon, 01 Feb 2010 15:06:47 GMT

Well, It’s happened again. Hundreds of thousands of private records were stolen from a database. This one, however, was different. No one stole any passwords, no one did any social engineering, nothing was captured in-line. No, this one was accomplished by stealing the actual hard drives themselves! When a thief breaks in and steals hard drives, you can be sure they know what they are after.

So even if the company had taken all of the regular authentication precautions and so on, the thieves could still take this step. But the data loss could still be prevented…

Enterprise SQL Server 2008 (and later) has a new feature called “Transparent Data Encryption”. When you turn this feature on, you don’t have to change anything in your applications – from then on, the data files, log files, tempdb and even the data on the backup files are encrypted. Without the certificate you use to encrypt them, the thief can’t get to the data.

I’m sometimes asked why a company should upgrade to the latest version of SQL Server, or why Enterprise Edition is important. A few hundred thousand private records sounds like a pretty good reason to me.

Quote of the Day: On Relaxing

Buck Woody — Fri, 29 Jan 2010 15:04:08 GMT

Love this:

“There is no pleasure in having nothing to do; the fun is in having lots to do and not doing it.” – Mary Little

FILESTREAM: Storing Binary Objects in a database – or not

Buck Woody — Thu, 28 Jan 2010 15:47:16 GMT

Many shops need to store binary large objects (sometimes called BLOBS) in a database. There are really only two ways to do this: store in them in a table structure in the database itself using a binary data type, or store them in the operating system in a file folder somewhere and point to the file using a text field in a table.

Both of these approaches have issues. Relational Databases aren’t really designed to hold that much data in a single field – not ours, not anybody’s. And pointing to a file is risky, since the file might change, it isn’t under database control for security and backups and so on.

Enter FILESTREAM. In SQL Server 2008 we introduced a feature that actually combines the two approaches into one. Using FILESTREAM, you enter the data into a “column”, but in fact SQL Server stores the data on the hard drive. It’s lightning fast, doesn’t lose the “pointers” to the files, and keeps the whole thing under database control. You can secure it with the same permissions as the database, and it gets picked up in backups and so on.

It does have some limitations and caveats, so be sure and check out the reference data in the following link. And then try it yourself – pretty easy to set up and manage.

http://technet.microsoft.com/en-us/library/bb933993.aspx

Data as Strategy

Buck Woody — Wed, 27 Jan 2010 13:48:00 GMT

I’ve created a presentation called “Data as Strategy”, and in it I cover multiple solutions for various issues that organizations face in dealing with the “new world” of data management. No longer are we just DBA’s or Developers, now we’re part of the broader business strategy of using data in strategic decisions.

That presentation covers a lot of ground, really quickly, so I thought I would post some of the links I mention during the talk here. I’ll actually come back and update this page as time goes on, so be sure to bookmark it. I’ll have a mix here of Microsoft sources, outside opinions, whitepapers, tutorials and videos wherever I can.

SQL Server 2008
New features on SQL Server 2008: http://tinyurl.com/ylks3xo
Policy Based Management Link List Page: http://tinyurl.com/yldm4co

SQL Server 2008R2 General
What's New: http://tinyurl.com/yzhdbod
Editions Overview: http://tinyurl.com/ygajpga and http://tinyurl.com/yg3a98z
Features supported by Edition: http://tinyurl.com/yf7bdzx
Unicode Compression: http://tinyurl.com/ycf892z
SysPrep and SQL Server Installs: http://tinyurl.com/yzh7t6j

SQL Azure
Azure Overview: http://tinyurl.com/y8s52vh
Windows Azure Training Kit: http://tinyurl.com/5vrt7q
Writing a Azure Program in 5 steps: http://tinyurl.com/ye6chog
Azure Data Sync: http://tinyurl.com/ylsykfb
Future of programming with Azure Video: http://tinyurl.com/ykqsbsf
Reference list for Windows Azure: http://tinyurl.com/yze9azr

SQL Server 2008R2 UCP and Management
UCP and DAC Overview: http://tinyurl.com/qd8n7m
More DAC information: http://tinyurl.com/yhvjmsb
Virtualizing the Database Layer: http://tinyurl.com/yhfl9hh

SQL Server 2008R2 Programming and Data Modeling
Data Programming and Modeling Video: http://tinyurl.com/nxuyda
More Data Modeling Videos: http://tinyurl.com/yj44zkv

SQL Server 2008R2 StreamInsight
Stream Insight Overview: http://tinyurl.com/l2jhgo
StreamInsight Whitepaper: http://tinyurl.com/mt3u6h
StreamInsight Video: http://tinyurl.com/y9llph3 and http://tinyurl.com/yg9r8bq

SQL Server 2008R2 PowerPivot
PowerPivot Resources: http://tinyurl.com/ygntujb
PowerPivot Server Install Instructions: http://tinyurl.com/yzegm5l
Self-service BI Overview: http://tinyurl.com/l3bapy
Virtual Lab for PowerPivot: http://tinyurl.com/yg9tdef
Practical overview of PowerPivot: http://tinyurl.com/yk422up

SQL Server 2008R2 Master Data Management
Master Data Services Overview: http://tinyurl.com/mc7qmx
Master Data Services Tutorial: http://tinyurl.com/yl8v37q
Master Data Management Whitepaper: http://tinyurl.com/abzpem

Attaching a Database

Buck Woody — Tue, 26 Jan 2010 16:03:39 GMT

I had someone ask me yesterday how they could get to a database used by a product that they owned, but that was installed using SQL Server Express. They didn’t have access to the database, and they didn’t know the password for the service that started Express, so they wanted to know if they could look at the data.

There are a few ways to do this, but the quickest, safest and easiest for me is to “attach” the database on another Instance of SQL Server. To understand how this works, let me quickly explain how SQL Server uses databases from a physical perspective.

SQL Server uses two types of files, sometimes lots of each type. The two basic types are Data files and Log files. SQL Server basically writes data to the Log file first, and then on to the Data files. Of course it’s more complicated than that; but this is the basic flow.

When SQL Server first starts up, it reads (among other things) the master system database. And that’s where the crux of this post comes into play. In that database are two things that we care about right now: the logical name of the databases SQL Server knows about and the physical locations for the files that database uses.

With that information in mind, here’s a simple way to transfer a database from an Instance you don’t have access to into one that you do.

There are a LOT of disclaimers here.
First: don’t do this on a database you don’t own, and own all the data for. That’s illegal.
Second, don’t do this with a large database – the possibility of multiple files in the filegroups on a complex database is high, and you might not get them all.
Third: you have to be running the same (or higher) version of SQL Server on the destination server, and the same (or higher) edition of SQL Server on the destination server.
Fourth: you’ll have to stop the source Instance. If you have users on that system, you don’t want to stop it without consulting everyone first.

OK – with all of that stated, you first need to stop the “source” instance. Let it shut down completely and normally. This will (theoretically) commit all of the transactions from the logs to the databases.

Now copy the .MDF and .LDF files for the database you want to a location on your test Instance. If you don’t know what these are, you can try and copy what you think are those files, but don’t copy any of the system databases from the source system to your test Instance!

Start the source Instance back up and let the users back into the application.

On the test system, write-down the location where you copied both of those files (assuming there was one of each – it get’s trickier when there are multiples).

Open SQL Server Management Studio.

Right-click the “Databases” node in Object Explorer and select “Attach…” from the menu that appears.

Click “Add” to find the MDF and LDF files, and then “OK” to name the database.

You have two other options as well. You can use the sp_attachdb stored procedure (detail here - deprecated) or the newer CREATE DATABASE…FOR ATTACH statement (article on that here).

In any case, you now have the database, and you now “own” it in that test Instance. The logins will all be incorrect, since the names in the database security don’t match the new Instance’s master database. But you do have that data.

Tracking SQL Server Time

Buck Woody — Mon, 25 Jan 2010 15:10:28 GMT

In the past few blog posts I’ve showed you how to use several methods to track things in SQL Server. You can use the “tags” to the right of this post here at this site to list things like PowerShell, Performance Tuning and so on. Now that you’re armed with these tools, what should you track?

Well, one of the items I track is time. I track the time it takes for lots of things, but they fall into three general buckets:

Queries – Normally I track the five longest running queries with their query plans.
Maintenance – From how long each backup takes to index reorgs and rebuilds, I want to know how long these things take.
Jobs – Most all of us have SQL Server Agent Jobs, and developing a schedule of how long they are running is very useful.

For each of these, I track the minimum, maximum and average times. I look for outliers – things that suddenly change and so on.

There are a lot of uses for this information. From performance tuning to developing a recovery plan, all of these actions need to be taken into account.

I think I’ll write up an article sometime on how I do this – it’s a little long for a blog post.

Quote of the Day: A Boyfriend Test (1)

Buck Woody — Fri, 22 Jan 2010 14:54:04 GMT

My daughter is in her teen years, and I told her early on that she should have a few “tests” for any boy she dates (when she starts dating, that is). This is is one of them:

“You can easily judge the character of a man by the way he treats those who can do nothing for him.” - Goethe

Hovering Over the Titles in Activity Monitor

Buck Woody — Thu, 21 Jan 2010 15:16:11 GMT

I show this little-known feature in just about every perf tuning class I give, and it’s often a surprise to most of the folks there. It’s not earth-shattering or anything, but I do find it helpful.

In SQL Server Management Studio 2008, start the new Activity Monitor. Open one of the “Bands”, such as “Processes”. Move your mouse cursor over one of the header titles and leave it there. If that column is getting information from a particular Dynamic Management View (DMV), we’ll tell you where we are getting it from.

Just thought I would share that.

Performance Tuning Methodology Preferences

Buck Woody — Wed, 20 Jan 2010 13:22:00 GMT

I’m holding Performance Tuning Workshops this week for my clients in Washington and Idaho. During that workshop, I explain that there are several ways to examine the SQL Server component of an application. There’s “Hardware Component Pressure Measurement”, “Code Optimization” and “Index Analysis” as starting points, and most folks start there and then move on to use all of them in a mix. But a newer methodology called ‘Wait State Analysis” helps you to evaluate what the system is waiting on, so that you can knock those down one at a time.

I mention in the presentations that there is a new whitepaper out on that, and I promised to share that link here. So without further ado, here it is: http://msdn.microsoft.com/en-us/library/cc966413.aspx (link is at the bottom).