2021-03-22: Ten Years of Blitzing

Ten years ago, I unveiled sp_Blitz at the 2011 PASS Summit, in front of a live audience.

Remember the PASS Summit?

Remember live audiences?

<sigh>

Anyway, to celebrate the 20th anniversary of BrentOzar.com, I've thinking back about the last couple of decades to reminisce about what I've learned, what's changed, and so forth.

Open source kinda sorta existed when I got started with computers as a kid way back in the 1980s. I would buy a programming magazine, prop it up in front of the computer, and actually type programs in by hand on my Commodore 64. Lemme just tell you, that sucked. I've been bad at debugging for four decades.

Today, anybody in the world can install open source utilities in a matter of seconds, make their job easier, increase their value as a professional, and give back to those same open source projects. It's not fair to say that your Github profile is the new resume - most folks just don't have the time to contribute to open source, nor are they allowed to at your day job. However, I think it's fair to say that your Github profile is one of many possible advantages when competing for a job. When someone asks, "Do you have experience with X?" it's amazing to be able to answer, "Yes, and I've even contributed code and documentation to X. Here's my Github profile, and here are some of my interactions with the team."

So with that said, here's this month's new version of the First Responder Kit. It's free - free as in speech - and it thrives thanks to the community members mentioned in the changes below. They're real people just like you with day jobs, and they decided that they wanted to contribute or fix something. They're the heroes that make this whole thing work.

When you find a bug or want something changed, read the contributing.md file.

To get the new version:

• Download the updated FirstResponderKit.zip
• Azure Data Studio users with the First Responder Kit extension: ctrl/command+shift+p, First Responder Kit: Import.
• PowerShell users: run Install-DbaFirstResponderKit from dbatools
• Download the updated Consultant Toolkit in your account
• EU customers: check your email for the updated version from Gumroad, our European distributor

Consultant Toolkit Changes

I updated it to this month's First Responder Kit, but no changes to querymanifest.json or the spreadsheet. If you've customized those, no changes are necessary this month: just copy your spreadsheet and querymanifest.json into the new release's folder.

sp_Blitz Changes

• Improvement: we ignore databases named DBAdmin, DBAtools, and DBMaintenance. (#2787, thanks Erik Darling.)
• Fix: made all of the scripts consistently refer to https://www.brentozar.com rather than variations without https, the www domain, etc. (#2808, thanks John D. Ballentine III.)
• Fix: removed the checks for active & inactive heaps in user databases. This is covered really well by sp_BlitzIndex now, and it's slow in sp_Blitz. (#2818, thanks Erik Darling.)

sp_BlitzAnalysis Changes

• This is the second release that includes sp_BlitzAnalysis by Adrian Buckman. If you want to install this script, install sp_BlitzAnalysis.sql - it's not included in the normal installation scripts yet just to limit the blast radius in case something goes wrong. For tips on how to use it, check out the documentation. For questions or problems, see the "For Support" section of this doc.
• Fix: expanded one of the columns from the money datatype up to decimal(38,2) to handle big data. (#2807, thanks Ian Menton.)

sp_BlitzCache Changes

• Improvement: when @IgnoreSystemDBs = 1, we ignore queries with a context of databases named DBAdmin, DBAtools, and DBMaintenance. (#2787, thanks Iomckee.)
• Improvement: @OutputType parameter now accepts 'NONE'. (#2817, thanks Jefferson Elias.)

sp_BlitzFirst Changes

As we mention in the documentation, Azure SQL DB isn't technically supported by any of the First Responder Kit scripts because Azure SQL DB doesn't have all of the cool diagnostics stuff that we get on-premises, like sp_MSforeachdb. However, at some point in the future, we'll officially support a lot of them up there, and this month's changes inch us towards that.

• Improvement: in Azure SQL DB, we now sys.dm_db_wait_stats instead of sys.dm_os_wait_stats. (#2821)
• Improvement: pass @OutputType parameter to sp_BlitzCache so if you specify 'NONE', there really is no output. (#2817, thanks Jefferson Elias.)
• Fix: insert of changed statistics would sometimes fail. (#2738, thanks Jefferson Elias.)
• Fix: Azure SQL DB was erroring out due to addition of sp_MSforeachdb in the recently changed statistics check. (#2819)

sp_BlitzIndex Changes

• Improvement: we ignore databases named DBAdmin, DBAtools, and DBMaintenance. (#2787, thanks Erik Darling.)

sp_BlitzLock Changes

• Fix: was throwing an error in some cases when @ExportToExcel = 1. (#2824, thanks Pierre van de Moosdijk and Erik Darling.)
• Fix: was throwing an error when SQLAgent - Initial Boot Probe had a logged deadlock. (#2800, thanks Roger S Allen and Erik Darling.)
• Fix: errors with time zone conversion, parallel deadlocks, gathering object names. (#2825, thanks Erik Darling.)

sp_BlitzWho Changes

• Fix: in the deltas view, expanded one of the columns from the money datatype up to decimal(38,2) to handle big data. (#2813, thanks Ian Menton.)

sp_DatabaseRestore Changes

• Improvement: you can now move the full text catalogs with the new @MoveFullTextCatalogDrive parameter. (#2805, thanks Frederik Vanderhaegen.)

For Support

When you have questions about how the tools work, talk with the community in the #FirstResponderKit Slack channel. If you need a free invite, hit SQLslack.com. Be patient - it's staffed with volunteers who have day jobs.

When you find a bug or want something changed, read the contributing.md file.

When you have a question about what the scripts found, first make sure you read the "More Details" URL for any warning you find. We put a lot of work into documentation, and we wouldn't want someone...

2021-02-22: New sp_BlitzAnalysis Script

Along with the usual bug fixes & improvements, this month's release includes a brand new sp_BlitzAnalysis script from Adrian Buckman. It's for folks who have an Agent job to log sp_BlitzFirst to tables every 15 minutes, saving wait stats, Perfmon counters, file stats, etc into tables. sp_BlitzAnalysis reads those tables and sums up activity in your chosen time period, like this:

To install it, run the sp_BlitzAnalysis.sql file included in the First Responder Kit. It's not a complex installation or anything - I just didn't want to include it in the main installer scripts because this is the first public release, and I haven't tested it myself yet either. (Been a little busy with the whole Iceland move.) To learn how to use sp_BlitzAnalysis, read the documentation. For questions or support on it, hop into the #FirstResponderKit Slack channel. (If you need a free invite, hit SQLslack.com. Be patient - it's staffed with volunteers who have day jobs.

When you find a bug or want something changed, read the contributing.md file.

To get the new version:

• Download the updated FirstResponderKit.zip
• Azure Data Studio users with the First Responder Kit extension: ctrl/command+shift+p, First Responder Kit: Import.
• PowerShell users: run Install-DbaFirstResponderKit from dbatools
• Download the updated Consultant Toolkit in your account
• EU customers: check your email for the updated version from Gumroad, our European distributor

Consultant Toolkit Changes

I updated it to this month's First Responder Kit, but no changes to querymanifest.json or the spreadsheet. If you've customized those, no changes are necessary this month: just copy your spreadsheet and querymanifest.json into the new release's folder.

sp_Blitz Changes

• Fix: the check for the most recent CHECKDB dates was running even when @CheckUserDatabaseObjects = 0. This was the only instance where we were using sp_MSforeachdb with @CheckUserDatabaseObjects = 0, so it'd slow sp_Blitz down on instances with hundreds or thousands of databases. (#2779)
• Fix: put comments around the term xp_cmdshell to fool a firewall. (#2788, thanks TLovie.)
• Fix: changed typo of BrenOzar.com to BrentOzar.com on one check. (#2786, thanks Michael Gowan.)

sp_BlitzAnalysis Changes

• All new! This is the first release that includes sp_BlitzAnalysis by Adrian Buckman. If you want to install this script, install sp_BlitzAnalysis.sql - it's not included in the normal installation scripts yet just to limit the blast radius in case something goes wrong.
• For tips on how to use it, check out the documentation.
• For questions or problems, see the "For Support" section of this doc.

sp_BlitzCache Changes

• Improvement: default output now includes PlanGenerationNum from sys.dm_exec_query_stats. Helps troubleshoot scenarios where the plan is frequently recompiling due to statistics changes. (#2792, thanks Tom Lovie.)
• Fix: a recent release broke @sortorder = 'recent compilations', and that's fixed. (#2772)

sp_BlitzFirst Changes

• Improvement: if more than 20% of the queries are runnable, waiting on CPU power to make progress, we warn you as shown above. sp_BlitzFirst makes two passes through to check - once at the beginning of the sample defined with @seconds (defaults to 5), and again at the end. (#2776, thanks Erik Darling.)
• Fix: if you passed in @OutputTableRetentionDays because you wanted to store more or less history in the output tables, that change wasn't being passed to sp_BlitzWho, which was still just storing 3 days. (#2758, thanks Emanuele Mezzo.)
• Fix: the high compilations/sec and recompilations/sec warnings had way too high of thresholds. They were looking for 1,000 batch requests per sec, but you can see high CPU usage due to compilations with even just 10 big ugly plans per second getting generated. Lowered the threshold to either 10 compilations/sec, or more compilations than batch requests per second - which happens when a proc has multiple statements with recompile hints in them. (#2770)

sp_BlitzIndex Changes

• Improvement: duplicate indexes are now sorted by row count descending so that the big bang-for-the-buck stuff comes first. (#2762, thanks Todd Chittenden.)
• Improvement: Aggressive Indexes warnings are now sorted by total lock wait time descending. (#2768)
• Fix: a few releases back, I added support for SQL Server 2019's new missing index DMV that tells you which queries are triggering the request. I had some challenges getting it to work reliably, and Microsoft just now documented it so I'm removing support for it temporarily. It'll be back. If you want to help with coding for it, check out #2185. (#2780)

sp_AllNightLog Changes

• Fix: the RPO and RTO now honor overrides set in the config table. (#2775, thanks Alin Selicean.)

sp_ineachdb Changes

• Improvement: Azure SQL DB compatibility in the sense that it'll run on Azure SQL DB, but it can't actually change databases since Azure SQL DB doesn't allow that. I'm referring to Azure SQL DB "Classic Flavor" here, not Managed Instances. (#2790)

For Support

When you have questions about how the tools work, talk with the community in the #FirstResponderKit Slack channel. If you need a free invite, hit SQLslack.com. Be patient - it's staffed with volunteers who have day jobs.

When you find a bug or want something changed, read the contributing.md file.

When you have a question about what the scripts found, first make sure you read the "More Details" URL for any warning you find. We put a lot of work into documentation, and we wouldn't want someone to yell at you to go read the fine manual. After that, when you've still got questions about how something works in SQL Server, post a question at DBA.StackExchange...

2021-01-18: Bug Fixes

You know how a lot of app release notes say "bug fixes and improvements"? We're not gonna lie: this one's all bug fixes. Huge shout out to the community contributors for catching stuff and contributing fixes.

New behavior for SQL Server 2008 & R2 users: sp_BlitzLock has never worked on SQL Server 2008, but starting this month, the installation of sp_BlitzLock simply fails with an error. In this month's release, one of the checkins uses THROW, which isn't available on 2008. This also means that if you use the Install-Core-Blitz-No-Query-Store.sql script to do installations, you're going to get an error. It's okay - all of the other stored procs install successfully. This behavior wasn't intentional by any means, and we'd like that not to happen. If you'd like to improve sp_BlitzLock's installer to bail out gracefully without an error, we'd gladly accept a pull request for that.

To get the new version:

• Download the updated FirstResponderKit.zip
• Azure Data Studio users with the First Responder Kit extension: ctrl/command+shift+p, First Responder Kit: Import.
• PowerShell users: run Install-DbaFirstResponderKit from dbatools
• Download the updated Consultant Toolkit in your account
• EU customers: check your email for the updated version from Gumroad, our European distributor

sp_Blitz Changes

• Fix: standardized the @Help = 1 behavior across this & all of the procs. When you run 'em with @Help = 1, they return the help info and immediately return. (#2717, thanks Todd Chitt.)
• Fix: offline databases no longer show up in auto-shrink or page verification checks. (#2750, thanks Erik Darling and Ali Hacks.)

sp_BlitzFirst Changes

• Improvement: if you turn on @debug = 1, it now raises "Running CheckID #___" messages in the messages tab in SSMS to help you debug. (#2739, thanks Adrian Buckman.)
• Fix: fix warning typo on queries with max memory grant and clarified that the query's memory grant is a percentage of max workspace. (#2732, thanks Ali Hacks.)
• Fix: forwarded fetches in table variables now show that it's a table variable rather than a temp table. (#2742)
• Fix: forwarded fetches in tempdb objects were reporting the cumulative number of fetches since the object was created rather than the forwarded fetches created during the sp_BlitzFirst sample. (#2743)

sp_BlitzIndex Changes

• Fix: if a missing index recommendation included columns whose names included ampersands, sp_BlitzIndex was failing with an error. (#2734, thanks snckirkmarken.)

sp_BlitzLock Changes

• Improvement: now runs better without sysadmin or db_owner permissions by wrapping an update statement in a try/catch block. (#2751, thanks Chad Baldwin.)

sp_ineachdb Changes:

• Fix: case sensitivity bug introduced in last month's version. (#2737, #2744, #2747, and #2748, thanks Jefferson Elias and sm8680.)
• Fix: GO was removed at the end of last month's version, which broke the Install-All-Scripts.sql installer when running on brand-new servers. (#2753, thanks TLSystemsLLC.)

sp_Who Changes

• Fix: we're supposed to sort queries from longest-running to shortest-running, but last month's version introduced a bug in that. This should fix it. (#2721, thanks Erik Darling and Ali Hacks.)

For Support

When you have questions about how the tools work, talk with the community in the #FirstResponderKit Slack channel. If you need a free invite, hit SQLslack.com. Be patient - it's staffed with volunteers who have day jobs.

When you find a bug or want something changed, read the contributing.md file.

When you have a question about what the scripts found, first make sure you read the "More Details" URL for any warning you find. We put a lot of work into documentation, and we wouldn't want someone to yell at you to go read the fine manual. After that, when you've still got questions about how something works in SQL Server, post a question at DBA.StackExchange.com and the community (that includes me!) will help. Include exact errors and any applicable screenshots, your SQL Server version number (including the build #), and the version of the tool you're working with.

2020-12-11: Bug Fixes & Improvements

sp_BlitzLock gets better identification of parallel deadlocks, sp_DatabaseRestore uses Ola's scripts for logging, and lots of bug fixes this month.

To get the new version:

• Download the updated FirstResponderKit.zip
• Azure Data Studio users with the First Responder Kit extension: ctrl/command+shift+p, First Responder Kit: Import.
• PowerShell users: run Install-DbaFirstResponderKit from dbatools

sp_Blitz Changes

• Improvement: documentation updates. (#2706, thanks AliHacks.)

sp_BlitzFirst Changes

• Improvement: in @expertmode = 1, when we show the top files by read & write stalls, show 20 files instead of 5, and also sort them by the worst stalls. (#2707)
• Fix: ignore FT_IFTSHC_MUTEX waits. (#2697)

sp_BlitzIndex Changes

• Fix: index suggestions on really long table names could be truncated. (#2680, thanks Ralf Pickel.)
• Fix: columnstore visualization only worked on tables in the dbo schema. (#2683, thanks Ali Hacks.)
• Fix: nonclustered columnstore visualization didn't report the correct columns. (#2684, thanks Ali Hacks.)
• Fix: if you passed in databases to be ignored, they were still counting against the 50-database limit before we made you pass in BringThePain = 1. (#2693, thanks skrishnan31.)
• Fix: temporal tables had table.schema in the detail names rather than schema.table. (#2694, thanks Mark Hions and Mikey Bronowski.)

sp_BlitzLock Changes

• Improvement: better identification of batch mode parallel deadlocks. (#2686, thanks Erik Darling.)
• Fix: no longer reports 0 parallel deadlocks. (#2682, thanks Erik Darling.)
• Fix: corrected time zone conversion issues. (#2674, thanks Merlin Bruno.)
• Fix: filter out incorrect results from the summary. (#2676, thanks Merlin Bruno.)
• Fix: reduce memory grant requirements. (#2691, thanks Ali Hacks.)

sp_DatabaseRestore Changes:

• Improvement: commands are now run with CommandExecute so that they get logged if things go wrong. (#2700, thanks Frederik Vanderhaegen.)
• Fix: the new @SkipBackupsAlreadyInMsdb switch was ignoring transaction logs if you were restoring them on the same server where you were taking backups (since they were already in msdb.) (#2710, thanks Greg Dodd, and you can watch a video of him fixing it.)

sp_ineachdb Changes

• Fix: now handles database names with spaces in 'em. (#2702, thanks renegm.)

For Support

When you have questions about how the tools work, talk with the community in the #FirstResponderKit Slack channel. If you need a free invite, hit SQLslack.com. Be patient - it's staffed with volunteers who have day jobs.

When you find a bug or want something changed, read the contributing.md file.

When you have a question about what the scripts found, first make sure you read the "More Details" URL for any warning you find. We put a lot of work into documentation, and we wouldn't want someone to yell at you to go read the fine manual. After that, when you've still got questions about how something works in SQL Server, post a question at DBA.StackExchange.com and the community (that includes me!) will help. Include exact errors and any applicable screenshots, your SQL Server version number (including the build #), and the version of the tool you're working with.

2020-11-14: Bundled with Azure Data Studio*

* - Not really, but we got a chuckle out of Microsoft bundling ADS with SSMS, like the classic toaster-fridge joke.

sp_Blitz Changes

• Improvement: new check for log backups to NUL, which would break point-in-time recovery. (#2640, thanks DBAdminDB.)
• Fix: stop alerting about usage of service accounts on the Launchpad service. (#2664, thanks Todd Chitt.)

sp_BlitzCache Changes

• Improvement: added @sortorder = 'unused grant', useful for when you have a lot of queries that are all hitting the max grant, and you're looking for low-hanging fruit of the ones that use the least memory. (#2646)
• Fix: removed extra LOWER calls. (#2633, thanks Maikel van Gorkom)
• Fix: when outputting the results to table, the PlanCreationTimeHours calculated column now correctly refers to the difference between when the plan was created vs when the data was saved to table, as opposed to the current time (which would have been artificially long). (#2655, thanks Adrian Buckman.)
• Fix: parameterized queries in different databases were flagged as multiple query plans for the same query. (#2653, thanks Erik Darling.)

sp_BlitzFirst Changes

• Fix: we were excluding wait types that weren't in our predefined list, which was a problem if you were hitting a new or obscure wait type. (Hello, EC.) (#2631)
• Fix: when saving sp_BlitzFirst's output to table, we now trim it to 4,000 characters to avoid problems with really long warnings. (#2660, thanks Jefferson Elias aka JeffChulg.)

sp_BlitzIndex Changes

• Improvement: revamped the sort orders to make the output more clear and actionable, and clarified the wording on some of the warnings. For example, "Cold Calculator" is now "Serial Forcer - Computed Column with Scalar UDF." Also added a new documentation file with the prioritized list of checks. (#2662)
  • @mode = 0 (default): now lists priorities 1-100, the most urgent things to fix.
  • @mode = 4: all priorities, 1-255, which includes not-easy-to-fix stuff (like bad clustering key designs or really wide tables), and informational warnings (like hey, you have temporal tables.)
• Improvement: added @SortDirection parameter that works in combination with the @sortorder parameter for @mode = 2, letting you sort by asc or desc. (#2651, thanks David A. Poole.)
• Fix: statistics updates will now show "never" if they've never been updated. (#2643, thanks Excelosaurus.)
• Fix: instead of showing 0 reads/writes in the summary, show "Not Tracked" on spatial indexes and "Disabled" on disabled indexes. (#2650)
• Work in progress, but not really working well yet: on SQL Server 2019, next to missing index requests in the table-level Mode 2 output and Mode 3 output, we show a sample query plan linked to that missing index request. (#2185, thanks Erik Darling.)
• Fix: statistics warning for low sampling rate was never firing because it was looking for rows < 1 rather than percent sampled < 1. (Caught & fixed this while I was working on #2662.)
• Fix: statistics checks now work in @GetAllDatabases = 1. (Caught & fixed this while I was working on #2662.)
• Fix: removed an extra space. (#2673, thanks John McCall.)

sp_BlitzLock Changes

• Improvement: better detection of parallel deadlocks. (#2636, thanks Erik Darling.)

sp_BlitzWho Changes:

• Fix: was throwing an error on AG readable secondaries when sp_BlitzFirst called sp_BlitzWho due to duplicate temp table names. (#2491, thanks Adrian B.)
• Fix: we now pad the number of days a query's been running with 2 digits instead of 0, so queries running more than 9 days (!!!) sort to the top of the list. (#2652, thanks Ahmet Rende.)

sp_DatabaseRestore Changes

• Improvement: @StoPAt can now skip full backups, too. (#2600, thanks Sergedess.)
• Improvement: new @SkipBackupsAlreadyInMsdb setting. When set to 1, we check MSDB for the most recently restored backup from this log path, and skip all backup files prior to that. Useful if you're pulling backups from across a slow network and you don't want to wait to check the restore header of each backup. (#2644, thanks Greg Dodd and Raphael Ducom.)

sp_ineachdb Changes

• Fix: prevents runtime error when using @database_list. (#2621, thanks Stuart Cuthbertson.)

For Support

When you have questions about how the tools work, talk with the community in the #FirstResponderKit Slack channel. If you need a free invite, hit SQLslack.com. Be patient - it's staffed with volunteers who have day jobs.

When you find a bug or want something changed, read the contributing.md file.

When you have a question about what the scripts found, first make sure you read the "More Details" URL for any warning you find. We put a lot of work into documentation, and we wouldn't want someone to yell at you to go read the fine manual. After that, when you've still got questions about how something works in SQL Server, post a question at DBA.StackExchange.com and the community (that includes me!) will help. Include exact errors and any applicable screenshots, your SQL Server version number (including the build #), and the version of the tool you're working with.

2020-10-11 Release: Lots of Columnstore & Sorting Features

In preparation for this month's new Fundamentals of Columnstore class, sp_BlitzIndex has a few cool new tricks! I'll show you how to use 'em for free in the upcoming How I Use the First Responder Kit class, too.

Columnstore indexes are kinda like an index on every column (called a segment), and plus they're kinda partitioned (called a row group). When you want to visualize how the data is stored, call sp_BlitzIndex just for that one table:

The last result set gives you a row per row group, plus columns for each column of the table, showing you what range of data is stored in each row group/segment combo. It helps you better understand what parts of the table SQL Server will be able to eliminate, and what parts it'll still need to dig through. This also affects how we load data and how we manage index rebuilds.

Next, when you want to analyze the sizes of your columnstore dictionaries, we're now breaking those out in the Index Size column. Columnstore indexes used to just show a total LOB size, but now it's broken out between columnstore and dictionary data:

That "Index Size" column shows in @modes 0, 2, 4, and table-level output. And hey, speaking of sizes, you're probably curious about which indexes & tables are the largest ones in your database. sp_BlitzIndex @mode = 2 now accepts a @sortorder parameter. You can now sort by rows or size, plus more stuff too.

To get the new version:

• Download the updated FirstResponderKit.zip
• Azure Data Studio users with the First Responder Kit extension: ctrl/command+shift+p, First Responder Kit: Import.
• PowerShell users: run Install-DbaFirstResponderKit from dbatools
• Download the updated Consultant Toolkit in your account
• EU customers: check your email for the updated version from Gumroad, our European distributor

Consultant Toolkit Changes

I updated it to this month's First Responder Kit, but no changes to querymanifest.json or the spreadsheet. If you've customized those, no changes are necessary this month: just copy the /Resources/SQLServerSetup checklist folder over your existing files to replace the SQL scripts.

sp_Blitz Changes

• Improvement: new check for >= 35 databases being VSS snapshotted at the same time. (#2581)
• Improvement: if you're outputting the results to database, the @OutputXMLasNVARCHAR parameter now lets you save XML columns of the table output as NVARCHAR. Useful if you're using a single table to store results over linked servers and locally. (#2601, thanks David Prestridge.)
• Improvement: added warnings for non-default settings for SQL Server 2019's new database scoped configuration options. (#2573, thanks mtndew01.)

sp_BlitzCache Changes

• Improvement: if you use the @sortorder parameter, rows won't be shown unless they actually have that problem. For example, if you used @sortorder = 'spills', we were ordering by spills - but if not a single query plan in cache was having any spills, we were still sorting by that column, and showing plans where Spills = 0. This sounds trivial, but it's actually a big deal for those of us who do a ton of plan cache analysis because now we can glance quickly at the output and go, "Your plan cache has been cleared since the spills problem happened, or the plan has aged out, or it's using a recompile hint. (#2174, thanks Erik Darling.)

sp_BlitzFirst Changes

• Fix: CHECKDB running wasn't always shown due to a cross apply (rather than outer apply) to query plan DMFs. (#2609, thanks Erik Darling.)

sp_BlitzIndex Changes

• Improvement: when you specify the @TableName parameter, there's a new result set that visualizes the row groups and segments. (#2584)
• Improvement: new @sortorder parameter lets you sort the inventory output from @mode = 2 by rows, reserved mb, forwarded fetches, lock time, reads, writes, and more. (#2619)
• Improvement: the Index Size column now separates columnstore index size vs its dictionary size. (#2585)
• Improvement: the stats histogram output now includes the modification_counter. (#2612, thanks Erik Darling.)
• Fix: the stats histogram & columnstore visualizations weren't working when sp_BlitzIndex was installed in a database other than master. (#2616, thanks Erik Darling.)

sp_BlitzLock Changes

• Improvement: added @ExportToExcel parameter like sp_BlitzCache has. When set to 1, we don't show the XML results in the output, making it easy to copy/paste the results to Excel. (#2613, thanks Rebecca Lewis.)

sp_BlitzWho Changes

• Improvement: added @sortorder parameter that accepts most of the column names, so you can sort by things like session_id or memory grant. (#2603, thanks Dylan Petrusma.)

sp_DatabaseRestore Changes

• Fix: if you set @RunRecovery = 0 and also pass in a @DatabaseOwner name, we ignore you and don't bother trying to change the database owner. (#2598, thanks Greg Dodd.)

For Support

When you have questions about how the tools work, talk with the community in the #FirstResponderKit Slack channel. If you need a free invite, hit SQLslack.com. Be patient - it's staffed with volunteers who have day jobs.

When you find a bug or want something changed, read the contributing.md file.

When you have a question about what the scripts found, first make sure you read the "More Details" URL for any warning you find. We put a lot of work into documentation, and we wouldn't want someone to yell at you to go read the fine manual. After that, when you've still got questions about how something works in SQL Server, post a question at DBA.StackExchange.com and the community (that includes me!) will help. Include exact errors and any applicable screenshots, your SQL Server version number (including the build #), and the version of the tool you're working with.

2020-09-13: September to Remember

Let's give a big round of virtual germ-free applause to Adrian Buckman, Erik Darling, Garry Bargsley, Greg Dodd, Ian Manton, John McCall, Kuldar Andares, Pierre Letter, Stephen Anslow, Tom Butler, and Walden Leverich for volunteering their time to help make your job easier this month. Thanks to them, the September release is chock full of features - I'm not even going to try to sum them up here, and you should read the release notes. Kudos to them!

To get the new version:

• Download the updated FirstResponderKit.zip
• Azure Data Studio users with the First Responder Kit extension: ctrl/command+shift+p, First Responder Kit: Import.
• PowerShell users: run Install-DbaFirstResponderKit from dbatools
• Download the updated Consultant Toolkit in your account
• EU customers: check your email for the updated version from Gumroad, our European distributor

Consultant Toolkit Changes

I updated it to this month's First Responder Kit, but no changes to querymanifest.json or the spreadsheet. If you've customized those, no changes are necessary this month: just copy the /Resources/SQLServerSetup checklist folder over your existing files to replace the SQL scripts.

sp_Blitz Changes

• Improvement: new check looks for Query Store problems where is_query_store_on reports 1, but sys.database_query_store_options doesn't agree. (#2525, thanks Walden Leverich.)
• Improvement: more readable info in the drive free space checks. (#2557, thanks Pierre Letter.)
• Fix: sp_Blitz was failing on Amazon RDS SQL Servers because it was trying to run sp_validatelogins, which Amazon doesn't allow. (#2526, thanks Erik Darling.)
• Now alerts if there's even just 1 memory dump in the last year (as opposed to 5.) (#2564, thanks Garry Bargsley.)

sp_BlitzCache Changes

• Improvement: @OutputServerName parameter lets you push the sp_BlitzCache results to a central server via linked server. The linked server will need to already exist and have RPC enabled. Linked servers don't support writing via XML, so when writing out to a linked server, the query plan column is created as NVARCHAR(MAX) rather than XML. (#2528, thanks Pierre Letter.)
• Fix: "Many Duplicate Plans" check was reporting numbers higher than 100%. I didn't do a great fix, but if anybody wants to dig into the real underlying problem, they can take the code from issue 2480 and apply it to this same problem. I'd be glad to take a pull request for that. (#2532, thanks Tom Butler.)
• Fix: removed unqualified join to get the Microsoft data migration wizard to stop complaining. (#2558, thanks Ian Manton.)

sp_BlitzFirst Changes

• Improvement: new warnings for queries running with memory grants > 15% of the buffer pool. You can configure that threshold with the @MemoryGrantThresholdPct parameter, too. (#2505, thanks Adrian Buckman.)
• Fix: the recently-updated-statistics check (ID 44) was being blocked by index rebuild operations due to sys.dm_db_stats_properties ignoring isolation level hints. (#2548, thanks Adrian Buckman.)

sp_BlitzIndex Changes

• Improvement: missing indexes now show both the data type for the column, and have commas in between the columns. (#2522, thanks Greg Dodd.)
• Improvement: if lock escalation has been disabled on a table, add text mentioning that. (#2553, thanks Stephen Anslow.)
• Fix: if Clippy came up with a missing index request more than 4,000 characters long, we were truncating it. Now, we preserve Clippy's insanity in all its blazing glory. (#2533, thanks Kuldar Andares for the report and Greg Dodd for the fix.)
• Fix: suggested index names are now capped at 128 characters long. This can produce duplicate index names, but frankly, that's the price you should pay for creating column names that long, buddy. (#2552, thanks Greg Dodd.)
• Fix: @mode = 3 (missing indexes compilation) had a few extra diagnostic columns in the output left over from working on last month's release. Removed those. (#2539)

sp_BlitzWho Changes

• Improvement: when you use the @OutputTable parameters, sp_BlitzWho now automatically creates a deltas view like the other procs. The deltas view shows the last collection of each query on a session, so you see the total CPU/reads/time impact of each query. Not collected across multiple sessions - just helps you identify long-running queries that aren't showing up in the plan cache, like stuff with option recompile hints or servers with a lot of memory pressure. (#2315, thanks Adrian Buckman.)
• Fix: when a database has RCSI turned on, some DMV (don't ask me which one) isn't always reporting all transactions are in RCSI. John McCall came up with a slick workaround to report the right isolation levels. (#2534, thanks John McCall.)

sp_DatabaseRestore Changes

• Fix: if you pass in an empty string for the @RestoreDatabaseName parameter, we use the existing database name instead. (#2536, thanks Greg Dodd.)

SqlServerVersions Changes

• Improvement: when creating the SqlServerVersions table, we now also add extended properties for documentation. (#2574, thanks John McCall.)

For Support

When you have questions about how the tools work, talk with the community in the #FirstResponderKit Slack channel. If you need a free invite, hit SQLslack.com. Be patient - it's staffed with volunteers who have day jobs.

When you find a bug or want something changed, read the contributing.md file.

When you have a question about what the scripts found, first make sure you read the "More Details" URL for any warning you find. We put a lot of work into documentation, and we wouldn't want someone to yell at you to go read the fine manual. After that, when you've still got questions about how something works in SQL Server, post a question at DBA.StackExchange.com and the community (that includes me!) will help. Include exact errors and any applicable screenshots, your S...

2020-08-08: The Kick Drum Release

A couple of big new features this month: sp_Blitz's @SkipChecksServer parameter lets you centralize a list of checks that you want to skip across different servers, and sp_BlitzIndex's missing index recommendation list now includes the datatypes, making it easier to do index analysis for clients even when you're not connected to their servers. I'll blog about both of those this week in separate posts to get you up to speed.

If you're working directly with the releases on Github, note that the "master" branch has been renamed to "main" instead.

To get the new version:

• Download the updated FirstResponderKit.zip
• Azure Data Studio users with the First Responder Kit extension: ctrl/command+shift+p, First Responder Kit: Import.
• PowerShell users: run Install-DbaFirstResponderKit from dbatools
• Download the updated Consultant Toolkit in your account
• EU customers: check your email for the updated version from Gumroad, our European distributor

sp_AllNightLog Changes

• Fix: if @BackupPath is set to null during installation, throws an error and stops now rather than failing silently. (#2474, thanks Curtis Browne.)

sp_Blitz Changes

• Improvement: @SkipChecksServer is now implemented, so you can have sp_Blitz check a centralized SkipChecks table for a list of checks, databases, and servers to skip across your entire estate. (#2496, thanks Pierre Letter.)
• Fix: reprioritize a few of the checks so that I can say with a straight face that you should drop everything and focus if you have alerts between priorities 1-50. (#2492)

sp_BlitzCache Changes

• Improvement: add plan_generation_num to ExpertMode output and to the logged table results. Adds the column to existing tables if necessary. (#2514)
• Improvement: add query_hash to the default output. Been using this a lot lately during my parameter sniffing classes. (#2513)
• Fix: @sortorder = 'query hash' wasn't working on case-sensitive servers. (#2482, thanks Adrian Buckman.)
• Fix: USERSTORE_TOKENPERM calculation was incorrect for pages_kb. (#2487, thanks Mike Vernon.)
• Fix: Single-use plans were reporting over 100%. (#2480, thanks Simon Yeats.)

sp_BlitzIndex Changes

• Improvement: added new @ShowAllMissingIndexRequests parameter to show 'em even if the query was only run a few times, or had low impact. (#2498, thanks valued customer Greg Dodd.)
• Improvement: added data types to missing index requests. (#2227 and #2512, thanks Greg Dodd and Oleg Strutinskii.)

sp_BlitzLock Changes

• Fix: databases & tables with periods in the name wouldn't show correctly in the detail results. (#2452, thanks Filip Cornelissen.)

sp_BlitzWho Changes

• Fix: sleeping sessions no longer show as accruing wait time. (#2465, thanks Adrian Buckman.)
• Fix: parallel blocked sessions weren't showing the blocking spid. (#2489, thanks Adrian Buckman.)

sp_ineachdb Changes

• Improvement: handle distribution databases as system databases. (#2507, thanks Andy Mallon and Aaron Bertrand.)

SqlServerVersions Changes

• Fix: The SqlServerVersions table has been updated with a few SQL Server 2012 security patches. Not new, just missed 'em in the original table. (#2481, thanks TheVladdestVlad.)
• Fix: Uninstall now removes the SqlServerVersions table. (#2502, thanks rebambus, and congrats on your first Github pull request ever!)

For Support

When you have questions about how the tools work, talk with the community in the #FirstResponderKit Slack channel. If you need a free invite, hit SQLslack.com. Be patient - it's staffed with volunteers who have day jobs.

When you find a bug or want something changed, read the contributing.md file.

When you have a question about what the scripts found, first make sure you read the "More Details" URL for any warning you find. We put a lot of work into documentation, and we wouldn't want someone to yell at you to go read the fine manual. After that, when you've still got questions about how something works in SQL Server, post a question at DBA.StackExchange.com and the community (that includes me!) will help. Include exact errors and any applicable screenshots, your SQL Server version number (including the build #), and the version of the tool you're working with.

2020-07-12: Fixing sp_BlitzFirst ExpertMode

I don't normally do multiple releases per month, but sp_BlitzFirst had a pretty big bug in the last release for those of us who do live performance tuning. In the July release, a well-meaning contributor tried to make sp_BlitzFirst faster when @OutputType = 'NONE', but inadvertently also made it faster when @expertmode = 1 - by skipping the output. Oopsie.

If you're working directly with the releases on Github, note that the "master" branch has been renamed to "main" instead.

To get the new version:

• Download the updated FirstResponderKit.zip
• Azure Data Studio users with the First Responder Kit extension: ctrl/command+shift+p, First Responder Kit: Import.
• PowerShell users: run Install-DbaFirstResponderKit from dbatools
• Download the updated Consultant Toolkit in your account
• EU customers: check your email for the updated version from Gumroad, our European distributor

sp_BlitzFirst Changes

• Fix: wait stats weren't being returned when @expertmode = 1. (#2463, thanks Rich Benner and Joseph LaBonde.)

sp_BlitzIndex Changes

• Fix: SQL Server has a bug that ignores isolation level hints on sys.identity_columns (note: no response from Microsoft on that one despite being open for several months, and hello, blue badges, that is why people get so pissed off about feedback.azure.com being like yelling into the void. This is a bug in your product, and you're not even taking the time to read the bugs that get reported. HELLOOOO) Anyhoo, we get blocked by offline index rebuilds. To fix it, we now have a 1-second lock timeout, and sp_BlitzIndex will simply return if there is blocking. (#2176)

For Support

When you have questions about how the tools work, talk with the community in the #FirstResponderKit Slack channel. If you need a free invite, hit SQLslack.com. Be patient - it's staffed with volunteers who have day jobs.

When you find a bug or want something changed, read the contributing.md file.

When you have a question about what the scripts found, first make sure you read the "More Details" URL for any warning you find. We put a lot of work into documentation, and we wouldn't want someone to yell at you to go read the fine manual. After that, when you've still got questions about how something works in SQL Server, post a question at DBA.StackExchange.com and the community (that includes me!) will help. Include exact errors and any applicable screenshots, your SQL Server version number (including the build #), and the version of the tool you're working with.

2020-07-04: Independence From Bugs Day Release

This month, 15 community folks contributed code - I think that's an all-time high for a single month's release! Good work, y'all. Thanks for payin' it forward. The bad news is that they're almost all bug fixes, hahaha. I do love that, though - given enough eyeballs, all bugs are shallow.

If you're working directly with the releases on Github, note that the "master" branch has been renamed to "main" instead.

To get the new version:

• Download the updated FirstResponderKit.zip
• Azure Data Studio users with the First Responder Kit extension: ctrl/command+shift+p, First Responder Kit: Import.
• PowerShell users: run Install-DbaFirstResponderKit from dbatools
• Download the updated Consultant Toolkit in your account
• EU customers: check your email for the updated version from Gumroad, our European distributor

Consultant Toolkit Changes

Updated the First Responder Kit with this month's script updates, plus:

• Improvement: the config file now lets you specify default values for the --upload and --deepdive command line switches. This way, you can pass even less instructions on to your clients - just have 'em run the utility, and you get their diagnostic data in your Amazon S3 bucket within minutes. Read this month's updated documentation for more details.
• Fix: no arithmetic overflow even if you have truly ginormous bigint sp_BlitzCache metrics. (#2270.)

sp_Blitz Changes

• Improvement: the installation scripts create the SqlServerVersions table if it doesn't exist, and repopulate it. (#2429, thanks Curtis Browne.)
• Fix: Update descriptions of operating systems and no longer call Windows 2012 & R2 "pretty modern." (#2418, thanks Randolph West.)
• Fix: when alerting about tables in the master database, ignore the Dynamics NAV license table. (#2426, thanks Johan Parlevliet.)

sp_BlitzCache Changes

• Maybe fix: possible arithmetic overflow when checking for many duplicate plans in the cache. Not entirely sure whether this actually fixes it, though, because I couldn't reproduce it. (#2425, thanks smcnaughton, Randolph West, and Santi Swarup Choudhury.)

sp_BlitzFirst Changes

• Improvement: now runs faster if @OutputType = 'NONE'. (#2423, thanks Jefferson Elias.)
• Improvement: new check for >10% of your memory being used by the USERSTORE_TOKENPERM cache. (#2134)
• Fix: last month's new update-stats check would fail if you had a LOT of stats updates at once, like, well, an update-stats job. Doh! Now we just show the first 4,000 characters of updated stats, sorted by rows in the table descending, so you can get a rough idea of when you had big churn. (#2409, thanks CJR aka camaro322hp.)
• Fix: remove Page Life Expectancy warning. It's a bad metric to monitor in the year 2020. (#2433)
• Fix: no longer alert on >10000x cardinality UNDER-estimations because they're false alarms when the query just started and hasn't retrieved data yet. (#2438, thanks Nicklas Bjälemark.)
• Fix: skips last month's new update-stats check when you have >20 databases. (#2439, thanks Will McCardell.)

sp_BlitzIndex Changes

• Improvement: In-Memory OLTP indexes now have a call to sp_BlitzInMemoryOLTP in the "More Info" column. (#296)
• Improvement: in the readme, documented the levels of support for specialized index types (columnstore, graph, spatial, temporal tables.) I'm not really actively adding support for those, but just wanted to make it clear in the documentation that they show up in the results, just with varying levels of details. For example, the sizes of most of 'em aren't shown.
• Fix: for disabled indexes, the "Create TSQL" column now shows a create rather than a drop. (#2447, thanks Filip Cornelissen.)

sp_BlitzLock Changes

• Fix: now works on case sensitive instances. (#2440, thanks pnauta.)

sp_BlitzWho Changes

• Fix: handles long database & schema names when creating the output tables. (#2421, thanks Jefferson Elias.)
• Fix: adds compatibility with Azure SQL DB by avoiding querying sysjobs. (#2435, thanks Jacob Golden.)

sp_DatabaseRestore Changes

• Improvement: file paths now take a comma-delimited list of paths for striped files. (#2180, thanks CubsRep.)
• Fix: won't error out if you specify both @StoPAt and @OnlyLogsAfter at the same time. (#2348, thanks Greg Dodd.)

sp_ineachdb Changes

• Fix: incorrect description text. (#2445, thanks John McCall.)

For Support

When you have questions about how the tools work, talk with the community in the #FirstResponderKit Slack channel. If you need a free invite, hit SQLslack.com. Be patient - it's staffed with volunteers who have day jobs.

When you find a bug or want something changed, read the contributing.md file.

When you have a question about what the scripts found, first make sure you read the "More Details" URL for any warning you find. We put a lot of work into documentation, and we wouldn't want someone to yell at you to go read the fine manual. After that, when you've still got questions about how something works in SQL Server, post a question at DBA.StackExchange.com and the community (that includes me!) will help. Include exact errors and any applicable screenshots, your SQL Server version number (including the build #), and the version of the tool you're working with.