Author: way0utwest

  • Capturing My Own Metrics: #SQLNewBlogger

    A customer was trying to compare two tables and capture a state as a performance metric. In this case, they were wanting to use Redgate Monitor and custom metrics, but since the tables were in a Memory-Optimized table, they couldn’t as Redgate Monitor runs inside a transaction.

    Note, there are workarounds, but they’re clunky.

    Fortunately, I had a quick solution, which involved SQL Server User Settable Objects. This post looks at how this works.

    The Scenario

    Let’s take the transaction out of the equation by using my own metric. SQL Server includes a few procedures that fall into a pattern. They are named with numerics as shown:

    • sp_user_counter1
    • sp_user_counter2
    • sp_user_counter3
    • sp_user_counter10

    Each of these corresponds to a value that is captured in a perfmon counter. The counters are in the objects called “SQLServer:User Settable”. The counter name is “query” and the instance is “User Counter n” where n is the number corresponding to the stored procedure.

    By default, these are 0, and you can see them here:

    2026-06_0191

    You can also see them in Perfmon

    2026-06_0192

    I’ll stick with SQL Server.

    If I want to alter a value, I call the appropriate procedure. I can do something like call the proc for 5 and set a value. I’ll then query the counters from T-SQL. You can see this below as the value for counter 5 is set to 3.

    2026-06_0193

    This value remains set. I’ll set counters 2 and 8 to 2, and then query again. Note that 5 is still set.

    2026-06_0194

    If I want the value set to 0, I need to set it. I’ll do that for counter 8.

    2026-06_0195

    These are metrics, so I need to pick an integer. I can’t set a decimal (or other type) and have it work. If the implicit conversion works, it works, but the value is a decimal.

    2026-06_0196

    If I make this a string, it fails with an error as ‘2.5’ doesn’t convert to an int. Same for a date. Using an int, like ‘5’, works.

    Solving the Issue

    In this case, for the customer, we solved the issue with a proc that performed their query The result of this query (and int) is sent to a counter value like this:

    CREATE or alter proc My_Checker AS BEGIN declare @i int select @i = count(*) from dbo.Customer a INNER JOIN dbo.Candidates b on a.CustomerName = b.PersonName select @i = @i + 1 EXEC dbo.sp_user_counter1 @1; END

    This can run from an Agent job on their schedule, updating the counter as appropriate.

    For their alerting, they can query this metric and set the boundaries that matter to them. In this case, whenever this is greater than 0, they want an alert.

    The user settable values aren’t that useful, especially as there are 10, but I’ve used them in a few places when I wanted to get instrumentation for an application. This allows me to easily capture values and watch them from any monitoring system.

    Worth knowing about and using if you need random things captured.

    SQL New Blogger

    This post took me around 15 minutes to write, though I spent about 10 minutes mocking this for our customer based on their system and then stripping out a few items that are specific to them.

    This is a post that shows how I can use features of SQL Server to solve a problem, which is something every employer wants. An AI would make this code easier, but I have to know to guide an AI in this direction and evalaute if this works. I’d certainly need to check if other apps were using these counters, which isn’t something an AI might think to do, especially as there could be lots of repos to scan.

    You can write something like this, showing how you’re use this features. Bonus points if you use an AI to help you (and disclose how).

  • T-SQL Tuesday #200: When I Look at a Query …

    This month is a milestone for T-SQL Tuesday. It’s number 200, which doesn’t sound big, but this is a monthly party (started by Adam Machanic). We have 12 blog party events a year. 200 means this has been running for almost 17 years (16 years and 8 months).

    I don’t care who you are, that’s impressive. Think about where you were, what you were doing, and what was happening in 2009.

    I haven’t been running it that long, but I am glad I took it over from Adam and have kept it going. Thanks to everyone who writes and reads the posts and especially the hosts.

    I tried to get Adam to host, but he declined. Fortunately Brent Ozar stepped up with a great initiation. My response below.

    At First Glance

    There are two things that immediately stand out to me when I see a query and create concern.

    1. cross joins
    2. functions in the where/on clause

    While there are other things I might see, these two stand out and usually I can guess there will be issues.

    For cross joins, I don’t see this as much when people use SQL Prompt or some other helper because they tend to use inner/left outer/right outer explicitly, or cross join. If you explicitly use a cross join, I might ask why, but these clauses require an ON clause, which means you’re deciding to join tables.

    Where I see people using old style joins, like this:

    select *
    
    from a, b
    
    where a.id > 23 and b.saledate > current_date
    
    or (a.id is null and b.saledate is null)

    I get worried. This happens in Oracle, and PostgreSQ, and it’s easy to forget to join a and b, especially when there are multiple tables. Usually cross joins happen with legacy join conditions.

    The other area is using functions in the WHERE clause. A common example is

    select *
    
    from customer
    
    where upper(customername) = ‘Steve’

    This function in the WHERE clause ruins the ability to see the data. The index is something like (‘Adam’, ‘bill’, ‘Steve’, ‘WILLIAM’). This can’t be used when the UPPER is applied. This often results in more reads, more scans than a system might otherwise take.

    There are plenty of other issues that can indicate performance issues, but these two are the ones I’ve often run into and the ones that would have helped 2004 Steve write and review better code.

  • Republish: Bad IT People

    I’m still on vacation in Napa Valley, at least for a few more hours. It’s a chance to get away for me, so you get to re-read Bad IT People.

    Hope you had a great weekend.

  • Independence Day 2026

    Tomorrow is the United States 250th Independence Day celebration. I’ve been hearing about this on the radio for weeks, and it’s struck me that I’ve been alive and lucky enough to experience both the 200th (1976) and the 250th (2026) milestones. I’m not sure either is more amazing than any other 4th of July, but the round numbers stand out.

    In 1976, I was an 8 year old, and I vaguely remember going to Mount Trashmore park in Virginia Beach. It was a fun day of a cook out with hot dogs and kool-aid, playing in playgrounds, and some few moments wading in the lake nearby. It felt like an amazing event to be a part of.

    As a 50+ year old, I’m less excited about the 4th of July, but I am excited this weekend. I’m flying this morning to San Francisco with my wife for a weekend in Napa Valley and a slightly better food and drink selection.

    While I’m away, I hope you have a quiet day, nothing goes wrong, and you ease into the weekend. Or if you’re in the US, I hope you also enjoy a day off.

    Steve Jones

    Listen to the podcast at Libsyn, Spotify, or iTunes.

    Note, podcasts are only available for a limited time online.