If you want to connect to this database instance with SQL Server Management Studio (SSMS), use the following as your connection string: (LocalDB)\MAPToolkit. As long as you are the owner of the LocalDB instance, you will be able to connect just fine. If you are not the owner of the instance, you will get the following error:

If you run into this, verify who the owner of the LocalDB instance is, by excecuting SqlLocalDb.exe from a command line as follows:

SqlLocalDB.exe info MAPToolkit

Which should return details about who owns the instance, when it was last started, and its state (running, for instance):

LocalDB instance "MAPToolkit" started.

C:\Users\Phil.Brammer>SqlLocalDb.exe info MapToolkit
Name:               MAPToolkit
Version:            11.0.2100.60
Shared name:        
Owner:              MyDomain\Phil.Brammer
Auto-create:        No
State:              Running
Last start time:    4/23/2013 9:04:31 AM
Instance pipe name: np:\\.\pipe\LOCALDB#SH2C50D1\tsql\query

Here we can see that “MyDomain\Phil.Brammer” owns this instance. Any other user will not be able to connect to it. In order to allow other users to connect, follow these steps:

• Connect to the instance in SSMS (logged in as the owner shown above)
• For the users/groups you want to be able to connect, add them with the SQL “sysadmin” server role.
• Share the LocalDB instance. The following command will do the trick: SqlLocalDb.exe share MAPToolkit MAPToolkitShare
• Stop the MAPToolkit LocalDB instance if it is running. SqlLocalDb.exe stop MAPToolkit
• Start the MAPToolkit LocalDB instance. SqlLocalDb.exe start MAPToolkit
• As the other user, open up SSMS and connect to the shared LocalDB instance using the following connection string: (localdb)\.\MAPToolkitShare

To see details about the shared LocalDB instance, you can execute SqlLocalDb.exe info MAPToolkit -or- SqlLocalDb.exe info .\MAPToolkitShare

Either command will show the same results:

C:\Users\Phil.Brammer>SqlLocalDb.exe info .\MapToolkitShare
Name:               MAPToolkit
Version:            11.0.2100.60
Shared name:        MAPToolkitShare
Owner:              MyDomain\Phil.Brammer
Auto-create:        No
State:              Running
Last start time:    4/23/2013 9:17:35 AM
Instance pipe name: np:\\.\pipe\LOCALDB#SH80EAD1\tsql\query

You can unshare the database with SqlLocalDb.exe unshare .\MapToolkitShare

Note: LocalDB does not allow remote connections. You will need to connect from the machine where the LocalDB instance is running.

For more details see the official announcement here. You can download the bits here.

When installing, please follow the below steps:
– Double-click the downloaded file from the above link to start the install.
– When you see “SQL Server 2012 Setup” don’t panic. You’re at the correct place.
– Accept the license terms, click Next.
– After the setup files are installed and you are at the “Installation Type” screen, click Next. (Do not “add features to an existing instance of SQL Server 2012″)
– On the “Feature Selection” screen, select “SQL Server Data Tools…” from the Shared Features section. Click Next.
– In the “Error Reporting” screen, select the error reporting option if you desire, otherwise just click Next.
– After the install complete, you may be presented with a “computer restart required” dialog box. Click OK.
– At this point, the install should be complete. Click close and restart if necessary.
– Open Visual Studio 2012 and start a new BI project.

When VS 2012 is open, you should see the familiar BI templates as below:

Enjoy!

Out of the box, the new SSIS 2012 catalog database (SSISDB) has some basic indexing applied, with referential integrity set to do cascade deletes between most tables. (You can see the schema here.)

The schema was setup and designed to allow for a maintenance job, SSIS Server Maintenance Job, to purge records out of the SSISDB database to keep it at a manageable size. The problem, as documented previously here, is that the job relies on cascade deletes to do the purge. If you are experiencing problems with this job taking a considerable amount of time, or you feel it may be causing other issues, please read that post.

Also out of the box, there are a few reports shipped with the product – Integration Services Dashboard, All Executions, and All Validations. (As an aside, if these reports are not enough for you, please check out MVP Jamie Thomson‘s Codeplex project – SSIS Reporting Pack) These reports obviously query the SSISDB contents, and as your catalog grows in size, you may find that some of these reports start to get slower and slower over time. Especially the unfiltered “All Executions” report.

One last thing that I experienced in my environment were deadlocks across numerous SSISDB tables.

The thing that ties these things together is this – slow performance. Many months ago, I dove head-first into the SSISDB catalog and began to troubleshoot these performance and deadlock problems I was encountering. After a bit of looking around, it became clear that some basic indexing strategies were not implemented.

First, if you are on SQL Server 2012 RTM, you need to install Service Pack 1. This is a must. It contains fixes for several bugs – a few that I raised and escalated appropriately. Which is actually a good point – all of this data I’m posting here has been shared with the SSIS dev team. I do not always post Connect items for you to vote on, and these indexing strategies are one of those cases.

Here are the recommendations:

Index #1 – ncidxOperation_Id on internal.event_messages

CREATE NONCLUSTERED INDEX [ncidxOperation_Id] ON [internal].[event_messages]
(
	[operation_id] ASC
)

This index facilitates the join back to internal.operations. Additionally, it helps with the foreign key relationship.

Index #2 – ncidxExecution_id on internal.executable_statistics

CREATE NONCLUSTERED INDEX [ncidxExecution_id] ON [internal].[executable_statistics]
(
	[execution_id] ASC
)

This index facilitates the join back to internal.executions. Additionally, it helps with the foreign key relationship.

Index #3 – ncidxExecution_idSequence_id on internal.execution_component_phases

CREATE NONCLUSTERED INDEX [ncidxExecution_idSequence_id] ON [internal].[execution_component_phases]
(
	[execution_id] ASC,
	[sequence_id] ASC
)

This index helps with the deadlocks I was encountering on internal.execution_component_phases. There are a few queries that retrieve values from this table, and this index was necessary in that regard (see stored procedure – internal.append_execution_component_phases as one such query).

Index #4 – ncidxOperation_id on internal.operation_messages

CREATE NONCLUSTERED INDEX [ncidxOperation_id] ON [internal].[operation_messages]
(
	[operation_id] ASC
)

This index facilitates the join back to internal.operations. Additionally, it helps with the foreign key relationship.

Index #5 – ncidxEvent_message_id on internal.event_message_context

CREATE NONCLUSTERED INDEX [ncidxEvent_message_id] ON [internal].[event_message_context]
(
	[event_message_id] ASC
)

This index facilitates the join back to internal.event_messages. Additionally, it helps with the foreign key relationship.

As a bonus, there was one change to a stored procedure that I had to make to also help with the deadlocks on internal.execution_component_phases. At the beginning of the procedure, internal.append_execution_component_phases, I added “SET TRANSACTION ISOLATION LEVEL SERIALIZABLE;” to ensure that this procedure can only run single-file (one at a time).

Together with these changes, I’ve been able to increase the performance of reports, seen deadlocks reduced to nearly zero, and it helped the SSIS Maintenance Job perform its purges more efficiently.

If you’d like to take a look at my scripts, you can download them all in a singe .zip file here – http://www.ssistalk.com/SSIS_2012_Missing_Indexes.zip

Together, these two authors put together a very nice desk reference of the new features introduced in SQL Server 2012, with topics ranging from an overview of the various SQL Server editions to new functions, such as EOMONTH (yep, there’s a new “End of Month” datetime function).

I highly recommend this book for any SQL Server user who is new to SQL Server 2012.

Obtain your copy here: http://www.packtpub.com/sql-server-2012-professional-tips-and-tricks/book

• Warning / Event ID 1004 / MsiInstaller / Detection of product ‘{A7037EB2-F953-4B12-B843-195F4D988DA1}’, feature ‘SQL_Tools_ANS’, component ‘{0CECE655-2A0F-4593-AF4B-EFC31D622982}’ failed. The resource ” does not exist.
• Warning / Event ID 1001 / MsiInstaller / Detection of product ‘{A7037EB2-F953-4B12-B843-195F4D988DA1}’, feature ‘SQL_Tools_ANS’ failed during request for component ‘{6E985C15-8B6D-413D-B456-4F624D9C11C2}’
• Information / Event ID 1040 / MsiInstaller / Beginning a Windows Installer transaction: {A7037EB2-F953-4B12-B843-195F4D988DA1}. Client Process Id: 16708.
• Information / Event ID 11724 / MsiInstaller / Product: SQL Server 2012 Management Studio — Install started.
• Information / Event ID 10000 / RestartManager / Starting session 0 – 2013-01-07T12:40:46.549958800Z.
• Information / Event ID 11728 / MsiInstaller / Product: SQL Server 2012 Management Studio — Configuration completed successfully.
• Information / Event ID 1035 / MsiInstaller / Windows Installer reconfigured the product. Product Name: SQL Server 2012 Management Studio. Product Version: 11.1.3000.0. Product Language: 1033. Manufacturer: Microsoft Corporation. Reconfiguration success or error status: 0.
• Information / Event ID 1042 / MsiInstaller / Ending a Windows Installer transaction: {A7037EB2-F953-4B12-B843-195F4D988DA1}. Client Process Id: 16708.
• Information / Event ID 10001 / RestartManager / Ending session 0 started 2013-01-07T12:40:46.549958800Z.

If you have the above symptoms, please look at this non-security hotfix KB download: http://www.microsoft.com/en-us/download/details.aspx?id=36215. Do not install if you have already applied any post SP1 hotfixes, such as SP1 CU1. You can also view the SQL Server Connect bug here: http://connect.microsoft.com/SQLServer/feedback/details/770630.

Installing that fix will bring your SQL version to 11.0.3128, fix the problem with the MSI installer, and it should drop your CPU consumption a bit as well.

To celebrate the five year anniversary of the Omaha SSUG, the topic of the evening will be … YOURS! John & Naras are bringing back the “Choose Your Own Adventure” agenda, where you get to bring your own questions, problems, fodder, tips, and best practices to be discussed by all attendees. If you haven’t been to one of these sessions before, or are feeling as if you just can’t quite find the answer to your questions online, be sure to take part in this meeting. They are extremely informational and beneficial to the SQL Community.

I will be in attendance, and if there are any questions that need to be raised to Microsoft, I will do my best to ensure the appropriate people see it.

You can view the materials either through the SQL Saturday 165 Website (http://www.sqlsaturday.com/165/eventhome.aspx) or through this Website via the Speaking tab at the top of the page.

First, this is what I am referring to:

When you right-click on a project under “Integration Services Catalogs” in SSMS, you can choose to view its versions. From this window you can revert back to an older version while still retaining the latest version deployed.

If you want to export projects, you can do so by right-clicking on the project and selecting “Export…” In this window, just point it to an appropriate location. In my case, I saved it to Documents\SQLSaturday165. The resulting file you produce is an .ispac (Integration Services deployment package). A .ispac file is just a .zip. Rename it to use .zip instead of .ispac if you want to extract a single package out of a project.

If you want to export an older version of the project, you have two options – set the older version of the project to be the current version via the Project Versions window, or you can use an SSIS package to explicitly export the desired project version. I will demonstrate the SSIS package solution. The full sample and download links are at the end of the post.

Step One
Create a global, temporary table (##) on the SQL instance that has the SSISDB catalog.

CREATE TABLE ##projectBLOB (val VARBINARY(MAX));

Step Two
Open up the ExportProject.dtsx package and modify the OLE DB Source in the data flow to use the values from the following query:

SELECT ov.object_version_lsn AS project_version_lsn,
       ov.object_id AS project_id,
       *
  FROM internal.object_versions AS ov
  JOIN catalog.projects AS p
    ON ov.object_id = p.project_id
   AND p.name = N'SQLSaturday165_SSISDBInternals' -- replace with your project name
  JOIN catalog.folders AS f
    ON p.folder_id = f.folder_id
   AND f.name = N'SQLSaturday165_SSISDBInternals' -- replace with the folder name that contains your project
 ORDER BY ov.created_time DESC;
GO

Step Three
Edit the derived column component to set the path that you want to put the exported project in. Remember that in the derived column component you need to escape slashes. (“C:\Temp” should be written as “C:\\Temp”)

Step Four
Change the SSISDB connection manager object to point to your SQL instance containing the SSISDB catalog.

Step Five
Execute the package.

Step Six
Confirm your project was exported to the path entered in the derived column component. You should have a .ispac file with the name of your project in that location. From this point, you can double-click on it to launch the deployment wizard, or you can rename the file so that it has a .zip extention. If you choose to rename it to .zip, you will be able to individually pull out the packages.

The resulting .ispac output file from the export process.

Renaming the .ispac file to .zip allows you to inspect its contents.

Just a word of warning – THIS SSIS APPROACH IS NOT A SUPPORTED METHOD OF EXPORTING PROJECTS FROM THE CATALOG. YOU NEED TO BE A MEMBER OF SYSADMIN OR THE SSIS_ADMIN DATABASE ROLE. YOU TAKE SOLE RESPONSIBILITY FOR RUNNING THIS PACKAGE IN YOUR ENVIRONMENT. I AM NOT RESPONSIBLE FOR ANY DAMAGE THAT IS A RESULT OF RUNNING THIS PACKAGE. THIS IS AN UNSUPPORTED TECHNIQUE THAT USES INTERNAL, NON-DOCUMENTED PROCEDURES FOR EXPORTING PROJECTS.

With that said, I am merely posting this technique for educational purposes. The supported method for exporting an older version of a project is to revert to the older version and exporting it using the GUI in SSMS and then reverting back to the correct, current project.

The package path referenced an object that cannot be found: "????????????????????????????". This occurs when an attempt is made to resolve a package path to an object that cannot be found.

Since the characters will inevitably be munged when displaying this page, here is an image of the actual text:

If you encounter these errors and you are using SQL Server-based configurations (using the dbo.[SSIS Configurations] table), double-check that the data types of all columns in the config table are NVARCHAR, not VARCHAR. If they are not NVARCHAR, set them back accordingly. Occasionally you may need to modify this table to extend the max length of some of the columns and when that happens someone may inadvertently change the data type to VARCHAR.

I will be doing a presentation on the internals of the SSIS Catalog introduced in SQL Server 2012. For those that haven’t been much involved in the new catalog subsystem of SSIS, please plan on attending as I will be giving an overview and then going into some of the procedures, processes, and logging that are possible with this new framework. See you there!

http://www.sqlsaturday.com/165/eventhome.aspx