SQL Server with Mr. Denny


June 17, 2019  12:37 PM

Why I Travel So Much

Denny Cherry Denny Cherry Profile: Denny Cherry

I travel a lot. But occasionally I get asked “Why”. The answer to that is pretty simple and comes down to three simple things.

1. When I just out of high school (18 years old ish) I had no clue what I’d be doing for a living. My first real job out of high school was working Tech Support at an Internet Service Provider (ISP). I liked working with computers, and I’ve clearly stuck with it. Like most Americans, I had no idea that I’d have a chance to see much if any of the world outside of the US. I wasn’t sure what I’d do, or where I’d go. And I’ve been very fortunate that I’ve been able to see as much of the world as I’ve been able to so far. What this rambling reason really means is that because I can, I do. The world is a very cool place full of some exciting things, and I’ve been fortunate to be able to see as many of them as I can.

2. I’ve made a ton of friends around the world and many of them I wouldn’t get to see very often if I didn’t travel so much. Sometimes these folks see me in San Diego, sometimes they see me in other cities like Seattle, it just depends. Some perfect examples come to mind (I’m sitting in a hotel in Amsterdam as I’m writing this); tonight I’m having dinner with a friend from Holland since I happen to be in town; or when I saw a friend from Slovenia every month for about a year because we happened to go to the same conferences every month. It’s so cool to be able to see friends that live all over the world instead of just catching up with them in letters or online.

3. By going to some of the places that I’ve been, I’ve been able to meet people that I would never have met before and seen things that I would never have seen from one. A perfect example of that is when I went to Saudi Arabia to speak at a conference there. We had an extra day between arriving and the conference so a bunch of us when to visit the National Museum that was a few blocks away. We were able to tour the museum, which most people wouldn’t be able to do simply because of the cost of getting there and the fact that Saudi Arabia doesn’t have any way for tourists to visit. You either go there on business, or you don’t go there.

So as to why I travel so much really boils down to these three things. I definitely understand that I’m fortunate to be able to travel, especially as much as I do, especially considering my health issues from just a couple of years ago.  So because of all of this, I travel as much as I can, as often as I can (yes, I work when I travel) so I can see places and meet people because there’s a lot to see out there and I want to see as much of it as I can.

Denny

June 10, 2019  4:00 PM

Azure Database for MySQL server has a slow query log now

Denny Cherry Denny Cherry Profile: Denny Cherry

I’m not sure when this feature got introduced, but I just saw it for the first time recently (and I asked for it when Azure Database for MySQL server was in private preview); but Azure Database for MySQL server can show you the slow queries that are running against the MySQL database. In fact, it’ll log more than just the slow queries, queries that don’t have an index can be logged as well as long-running queries.

If you open the properties of a database in the Azure portal, there’s a couple of different places you can set these settings. The first is on the “Server parameters” blade where you’ll see all the various parameters that you can set for the MySQL deployment. The second is if you select the “Server Logs” option towards the bottom (it’s in the Monitoring section) you’ll see the log files that are being created (if there’s any). At the top of this page, you’ll see a blue bar which says, “Click here to enable logs and configure log parameters.” If you click that it’ll show you a shorter list of the parameters which only includes the settings that are related to the log file (shown in the screenshot).

Once you enable the settings that you need wait a few minutes for there to be load on the system, then go into the “Server Logs” setting of the Azure portal, and you’ll see a log file that you can download. Just open it in notepad (or your favorite text editor), and you’ll see the various log records that you enabled.

While I was playing around with this, I turned on the slow queries option, and I got a bunch of records in the log about queries that WordPress needs to do some tuning on. Since I’m guessing WordPress won’t be doing any database tuning, I’ll have to do it myself. But now that I have some data to work off of, I at least have a starting point.

Denny


June 3, 2019  4:00 PM

Free Performance Metrics from Traffic Manager

Denny Cherry Denny Cherry Profile: Denny Cherry

A couple of weeks ago DCAC moved our website from being hosted in a single data center to being a globally distributed web application with multiple Azure WebApps with one hosted in the US and one in Europe.  As part of having traffic manager configured and enabled for our site, we get some free reporting on performance of our webapp that’s included as part of the package.  As you can see from our performance chart which we can see on the left, we can see where in the world our readers are coming from, and depending on if the system is able to capture it or not, we can see the performance that those users are getting (all the colors in the chart).

For some users, we aren’t able to get back performance data, and those users are shown in White. The users who are shown in a color (blue/green is good, red is bad) the system was able to get performance data for.

There is a little change which helps gather real-time measurements of data.  Within the settings for Traffic Manager, there’s a setting for “Real Time measurements” which gives you a little bit of javascript to include in your website (I’ve removed our key from the picture). You’ll want this javascript on every page that you have on your website. If you’re using WordPress like we are, we were able to have this on every page by putting it in the default footer that’s part of our theme.  The code that’s in the Measurement JavaScript box in the Azure portal, I copies and put in our Theme’s footer between the <?php wp_footer(); ?> and the </body></html>. This put the code at the very bottom of the page, so even if there’s a delay in loading it, it won’t affect our website performance.

By using this basic reporting, we can see some pretty valuable information.

  1. We can see where our website viewers are coming from
  2. We can see what the performance for our viewers are

By showing us where in the world the folks reading our website, our coming from this gives us more insight into what kinds of things we should continue to write about. It also gives us insight into where we might need to deploy additional WebApps for better performance so that users in those areas get better performance.

By showing us the performance that users are getting around the world, we can see roughly what kind of performance improvement that users will see if we deploy another copy of our website to their region.  As we can see from the graph users from India, South East Asia, Australia, and Africa are getting 200ms or higher performance levels from browsing our website.  Based on this it might make sense to deploy another copy of the website into Singapore (which would improve performance in South East Asia, India, and Australia) and another copy of our website to the South African data center to improve performance for users in Africa.

Creating additional WebApps in these regions isn’t a give-in. 200ms isn’t a guarantee that there’s a problem. In our case we can talk to people in those various regions of the world and have them browse to the website and see if there’s a problem with viewing the website or not.  In this case, the data that’s available from Azure is just a piece of the puzzle to see if there is a performance problem or not.  This is because only the information that’s available for the portal isn’t everything.

Denny


May 27, 2019  4:00 PM

Cost Threshold for Paralellism

Denny Cherry Denny Cherry Profile: Denny Cherry

Cost Threshold for Parallelism is a setting in SQL Server that’s been around for as far back in the product that I can think of. It decides how expensive an operator needs to be because SQL Server will use multiple threads for that operator instead of a single thread. The Cost Threshold for Parallelism setting is set for a default of 5 (which is a horrible setting, but we’ll get into that later) which means that any operators which have a cost of below 5 will only use a single thread and any operators will have a cost of 5 or more will use multiple threads.

SQL Server has a default of 5 because it has to have something as the default. But that setting should be changed for most systems.  For OLTP systems, I typically recommend changing this setting to a value of 50.  The reason that I like to start with a setting of 50 for OLTP application is that any operators that have a cost of less than 50, typically aren’t going to see any improvement by using multiple threads. They will usually get slower because of the cost of using parallelism.  So we want to keep the quicker queries from using parallelism so that it’s only being used by the more expensive queries that actually need it.

Now 50 isn’t a perfect setting.  After making a change to the setting, the server needs to be monitored to make sure that to many queries aren’t going parallel.  The Cost Threshold for Parallelism might need to be adjusted again after the initial change, so keep in mind that this isn’t a one size fits all recommendation.

Another thing to keep in mind is that changing this setting will cause your plan cache to be expired and all the plans will need to be recompiled, so you should see CPU load on the server go up right after you change the setting.

Denny


May 20, 2019  6:00 AM

Making WordPress a globally distributed application on Azure WebApps

Denny Cherry Denny Cherry Profile: Denny Cherry

Before I start, I’m just going to say that if I can do this with WordPress, then it can be done with anything.

The goal that I was going for to have a WordPress website that was running on multiple contents (I only used two regions, but this can be scalled out as needed to other regions as needed) so that responce time on one site of the planet would be similar to users of the site that are on the same side of the planet as me.  I also wanted the failover for all of this if an Azure Region goes offline to be as quick and painless as possible.

Here’s what I had to get setup. I setup two Azure WebApps both of which were running Windows. One was setup in US Central while the other was setup in North Europe.

Website Failover

I tried doing this with Linux WebApps, but some of the underlying functions that I needed to make this all work.  Specifically I needed Site Extentions, which are only available on Windows WebApps today.

After setting up the two web apps, I uploaded the website content to the Azure WebApp.  Once the site was uploaded, and the custom DNS was working on the WebApp it was time to start making the replication of the web traffic work.

In a new web browser I opened the .scm.azurewebsites.net version of my site.  To get the SCM site it’s http://{Azure WebApp}.scm.azurewebsites.net.  Since my Azure WebApp is named www-dcac-com-central by SCM site is https://www-dcac-com-central.scm.azurewebsites.net/. (There’s going to have to go to this site a few times, so keep the URL handy.)

Be very careful of the order that you are doing the replication. If you setup the replication from a blank website to your website, then it’ll replicate the blank website. So before doing this, make sure that you have a proper backup on your WebApp BEFORE you configure the mirroring.

One you have the website open click on the Site Extenstions on the right of the menu at the top.

You’ll see a galary option.  Select the galary option and find the “Site Replicator” extention. Enable this extention by clicking the plus sign.  A window will popup to complete the install, click install.

Once the installation is complete, go back to the Azure portal. From the Azure Portal stop the WebApp and then Start it again (clicking restart will not work).

Again in the Azure Portal select the second site (in my case the second region is the North Europe region).  From here you need to download the publish profile for the WebApp.  To do this from the Overview option tab select the “Get publish profile” option from the top right.

Just download the file, we’re going to need it in a second.

Go back to the first sites SCM website (https://www-dcac-com-central.scm.azurewebsites.net/ in my case) and click the play button on the Site Replicator extension.

This is going to give you the configuration screen for the Site Replication (it may take a minute to open). The setting screen is pretty simple.  There’s a browse button on the bottom left of the screen, click that and navigate to the publish profile file that you downloaded earlier.

Give the site some time to replicate all the changes to the other region.  When it says “Succeeded” it should be just about done. The larger the website, the longer this will take.  I FTPed into both WebApps and watched the files appear until they were all there.  On a standard WordPress install, this took about 10 minutes.

Once this was all finished, I repeated the process in the other direction.  I downloaded the publish file from the US Central version and configured the Site Replicator on the North Europe region; then I uploaded the publish file to the North Europe region.  I then gave the process about 10-15 minutes to settle down and do any replication that needed to be completed.

Once this was all finished, I was able to upload files, pictures, WordPress updates, etc. from either site and the change would be replicated to the other region within a second or two.

Once the website replication was handled it was site to setup Traffic Manager. This would allow people to connect to their local version of our website depending on where in the world they are connecting from. Getting the endpoints setup was pretty simple. I used basic geographic load balancing and pointed North/Central/South America to the US Central version, and Asia/Europe/Middle East/Africa/Anti-Artica to the North Europe version.

The only hard part was that because WordPress is going to have a ton of redirects you can’t do normal HTTP monitoring. Normally you could have traffic manager pointing to “/” for the path to monitor, but WordPress didn’t like this. I changed the website to use “/license.txt” instead as the path as this would cause the traffic manager can come online correctly. It isn’t a perfect situation, but it works well enough.

Once everything is setup and traffic manager is happy and working, we can point public DNS to the site.  In our DNS configuration for www.dcac.com we added a CNAME record to DNS. A CNAME record in DNS redirects the request to another record.  In our case we pointed www.dcac.com to www-dcac-com.trafficmanager.net. This allows the Traffic Manager service to resolve www.dcac.com to the correct version of the website.

We can test that this is working as expected by looking at the output of the nslookup command.

By running nslookup on my laptop (which is currently sitting at my house in San Diego, CA), we can see that I’m resolving www.dcac.com to www-dcac-com-central.azurewebsites.net.

If we do the same nslookup command from an Azure VM that’s running in Singapore, from the VM in Singapore when I do a nslookup on www.dcac.com I get back www-dcac-com-northeurope.azurewebsites.net.

From these outputs, I can assume that I’m viewing the version of the website that’s closer to the user.

I’ve now got two duplicate copies of the website running in two different Azure Regions.

Database Failover

On the database side of things, need need to setup some replicate for that as well. “Azure Database for MySQL servers” now supports multi-region replicas but there’s no auto-failover available yet (hopefully it’ll be coming at some point soon).  For the database copy I did basically the same thing as I did for the websites (and that I’ve done tons of times for Azure SQL DB).

For the database side of things I setup WordPress to use the dcac-mysqlcentral copy. From there I clicked the Add Replica button, and that made a copy of the data to a new server called dcac-mysqlnortheurope that I setup in North Europe.

Since there’s no automatic failover at the database level today, if I need to do a failover I need to edit the wp-config.php file on the webserver, and that’ll kick the connection over to the other server.  I also need to setup the needed PowerShell to do the failover. My next step of this process is going to be to setup some Azure Automation to handle all the database level failover, but so far this is a pretty good step as there’s not website level failover.

The End Result

The end result of all of this is that our website is setup and running in two different places for better availabililty and better performance of our application.

Denny


May 13, 2019  4:00 PM

SQL Server Performance Monitor Objects

Denny Cherry Denny Cherry Profile: Denny Cherry

Monitoring SQL Server Performance Monitor objects (Perf Mon for those in the know) can be an important part of monitoring your SQL Server instance. Finding information about the performance monitor objects that SQL Server exposes can be tricky, even though the SQL Server Project Team can documented what these objects all mean.

You can find the documents about the SQL Server Perf Mon objects online on the Microsoft docs website.

If you haven’t had a chance to check it out, I’d very much recommend it. If you are looking for which objects you should be monitoring this can answer a lot of questions for you.

If you need more help past that, let us know, we can help you out.

Denny

 


May 6, 2019  4:00 PM

Changing the Time Zone of Azure VMs

Denny Cherry Denny Cherry Profile: Denny Cherry

When you create a VM in Azure it’s always set to the UTC timezone. There are some times when that doesn’t work and it needs to be set for a specific time zone. In a perfect world, the apps could be fixed so that they could deal with the fact that the servers are now in UTC instead of the local timezone. But this isn’t always possible, and the server’s time zone needs to be changed.

The “normal” process to change the timezone for a Windows server doesn’t work as expected. You can change the time zone by right-clicking on the clock and selecting “Adjust Date and Time”. If you change the time zone here, it doesn’t actually do anything (at least it didn’t when I did it).  It may also change for a short period of time and then revert back to UTC.

If you use PowerShell to change the timezone the change will stick, even if the VM is deallocated and reallocated.

First, we need to see what the options are for changing the timezone.  We can see that by running the Get-TimeZone cmdlet.
Get-TimeZone -ListAvailable
If a list of every time zone possible isn’t helpful (and it probably isn’t) you can filter the list down as you probably know the name of the time zone you’re looking for.  My specific client needed a server created in the Pacific Time Zone, So I filtered it down by the word Pacific.
Get-TimeZone -ListAvailable | where ({$_.Id -like "Pacific*"})
Second, we then use the cmdlet Set-TimeZone to change the time zone of the VM.  You’ll want to put the ID from the Get-TimeZone cmdlet into the ID parameter of the Set-TimeZone cmdlet.

Set-TimeZone -Id "Pacific Standard Time"

Denny


April 29, 2019  4:00 PM

Data Platform Summit 2019

Denny Cherry Denny Cherry Profile: Denny Cherry

I’m thrilled to be able to announce that I’m presenting two pre-cons at the Data Platform Summit 2019 in Bangalore, India. The first all-day session that I’m giving is “Azure Infrastructure” which I’ll be presenting on August 19th, 2019. The second that I’m presenting is “HADR – SQL Server & Azure” which I’ll be presenting on August 21st, 2019.

Azure Infrastructure

In this daylong session, we’ll review all the various infrastructure components that make up the Microsoft Azure platform. When it comes to moving SQL Server systems into the Azure platform having a solid understanding of the Azure infrastructure will make migrations successful and making support solutions much easier.

Designing your Azure infrastructure properly from the beginning is extremely important. An improperly designed and configured infrastructure will provide performance problems, manageability problems, and can be difficult to resolve without downtime.

With the introduction of multiple Azure data centers now available in India, many companies will begin moving services from data centers into the Azure, and a solid foundation is a key to successful migrations.

HADR – SQL Server & Azure

In this session, we’ll walk through the needs and process to set up a Hybrid Always On Availability Group using servers on premises for production and servers in Azure for Disaster Recovery.

We’ll be looking at high availability and disaster recovery tuning requirements, troubleshooting steps and various best practices for Always On Availability Groups in SQL Server 2017.

Why you want to be there

These two all-day sessions will both be a fun-filled day of learning that you won’t find anywhere else. Be sure to register now, as prices go up each month (the next price increase is in just a couple of days).

Denny


April 22, 2019  4:00 PM

Why your company doesn’t need block chain

Denny Cherry Denny Cherry Profile: Denny Cherry

Blockchain is the new hot thing in IT. Basically, every company out there is trying to figure out where Block Chain fits into their environment.  Here’s the big secret of blockchain; your company doesn’t need it.

Blockchain is simply a write one technology that allows you to change records, but it keeps track of every change that was made.  Most systems need some auditing to see when specific changes were made for example thing about an order system that your company may have. You probably have auditing of some sort so that you can see when the new order comes in (it’s probably the create date field on the table), and there’s probably some sort of auditing recorded when the shipment is sent out. If the customer fixes their name, you probably aren’t keeping a record of that, because odds are you don’t care.

Think about what systems you have at your company. Do you need to keep a record of every single change that happens to the data, or do you care about what happens to only some of the tables?  Blockchain is a great technology for the systems that need that sort of data recording. But that’s going to be a small number of systems, and we shouldn’t be fooling ourselves into believing that every company needs a system like this.

I’m not g0ing to argue that there are no systems that need this; there definitely are some systems that due. But those systems are going to be in the minority.

Executives are going to read about how blockchain is this great new thing, and they are going to want to implement it. The thing about blockchain is that there’s one major thing that building a system on blockchain requires, and that’s lots of drive space. If you want to purge data from the system after 5-6 years, that’s great; you’ll need more drive space as deleting data from a blockchain database just means that you need more space as you aren’t actually deleting those records.

A friend of mine described Blockchain as a database in full recovery mode, and you can’t ever back up (and purge) the transaction log. That’s how the database is going to grow.  Remember those lovely databases that were on the Blackberry Enterprise Server back in the day? The database would be 100 Megs and the transaction log would be 1 TB in size. That’s precisely what blockchain is going to look like, but it’s going to be a lot worse because all your customers and/or employees are going to be using the application.  If you have a database that’s 100 Gigs in size after a few years (which is a reasonable size for an application) the blockchain lot for this could easily be 15-20 TB in size, if not 100TB in size. And you’ll have to keep this amount of space online and available to the system.

So if you like buying hard drives (and the nice car that they get from their commissions) then blockchain is going to be great. If you don’t want to spend a fortune on storage for no reason, then blockchain is probably something you want to skip.

Denny


April 15, 2019  7:56 PM

Servers running too slow, just add all the cores!

Denny Cherry Denny Cherry Profile: Denny Cherry

Recently Intel announced some major upgrades to their Xeon CPU line. The long and short of the CPU announcement was that Intel was releasing their 56 Core CPUs for public release. That’s just a massive amount of CPU power that’s available in a very small package. A dual socket server, with two of these CPUs installed, would have 112 cores of CPU power, 224 with Hyper-Threading enabled. That’s a huge amount of CPU power.  And if 112 cores aren’t enough for you, these CPUs can scale up to an eight-socket server if needed.

With each one of the processors, you can install up to 4.5TB of RAM on the server, per socket.  So a dual socket server could have up to 9TB of RAM. (That’s 36TB of RAM for an eight-socket server if you’re keeping track.)

For something like a Hyper-V or a VMware host, these are going to be massive machines.

My guess is that we won’t see many of these machines are companies. Based on the companies that Intel had on stage at the keynote (Amazon, Microsoft, and Google) we’ll be seeing these chips showing up in the cloud platforms reasonably soon.  The reason that I’m thinking this way is two-fold; 1. the power behind these chips is massive, and it makes sense that these are for a cloud play; 2. the people who were on stage at the Intel launch were executives from AWS, Azure and GCP.  By using these chips in the cloud, the cloud providers will be able to get their cloud platforms probably twice as dense as they have them now. That leads to a lot of square feet being saved and reused for other servers.

As to how Intel was able to get 56 cores on a single CPU, is through the same technique that they’ve used in the past. They took two dies, each with 26 cores on them and made one socket out of that.  In the olden days, we’d say that they glued two 26 core CPUs together to make one 56 core CPU. The work that Intel had to do, to make this happen was definitely more complicated than this, but this thought exercise works for those of us not in the CPU industry.

These new CPUs use a shockingly small amount of power to run. The chips can use as little as 27 Watts of power, which is amazingly low, especially when you consider the number of CPU cores that we are talking about. Just a few years ago, these power numbers would be unheard of.

Denny


Forgot Password

No problem! Submit your e-mail address below. We'll send you an e-mail containing your password.

Your password has been sent to: