SQL Server

Microsoft SQL Server 2017 powers your entire data estate by supporting structured and unstructured data sources. It builds on previous versions of SQL Server, which have been industry leading for four years in a row and a leader in TPC-E. It scales to petabytes of data and allows customers to process big data through PolyBase using T-SQL over any data. SQL Server has also been the least vulnerable database during the last seven years. SQL Server 2017 brings data insights with business intelligence capabilities that provide analytics at a fraction of the cost on any device along with advanced analytics with support for R and Python.

More details about SQL Server are available in the SQL Server documentation.
If you have a technical issue, please open a post on the developer forums through Stack Overflow or MSDN.

  • Hot ideas
  • Top ideas
  • New ideas
  • My feedback
  1. 0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Bugs  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 3

    <=-=Aug 20 2008 10:03AM=-=>

    Hi Bill,
    Do you have SSMS configured in a Tab or MDI format? You can check this by selecting Tools → Options and looking at this setting on the General area. Perhaps a screen shot will also help.

    Thanks,
    Dan

    <=-=Aug 20 2008 11:22AM=-=>

    Tab (the default)
    I already attached a screen shot.
    Start SSMS and reduce the size to a non-full-screen window. Start a new query which opens a new window in the center area. Maximize SSMS. Note the Query window does not grow.

    <=-=Jul 5 2010 1:36PM=-=>

    I am also having what i believe is the same problem, and it’s extremely frustrating. It only happens when i have tablet drivers (i.e. Wacom) installed and the Microsoft Tablet PC Input Service running, Wacom tell me to stop the MS Tablet service and that does help but i don’t like the answer.

    If…

  2. SQL Server 2008 RTM - In SSMS Ctrl+N and/or F8 shortcuts not available

    Sometimes it happens that there is no Ctrl+N and/or F8 shortcuts available in SSMS. They are not visible in main menu (but the appropriate options: File - New - Query with Current Connection and View - Object Explorer ara available) and do not work.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Bugs  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 25

    <=-=Aug 26 2008 9:24AM=-=>

    Hi Brejk,

    Thank you for reporting the problem. If the table designer was open then the F8 may not have opened the Object Explorer. The current workaround in that case would be use the View→ Object Explorer menu option. CTRL+N consistenly opened the Query Editor window.

    Can you please update the cases when CTRL+N and F8 were not working as expected?

    Thanks,
    Karthik

    <=-=Sep 5 2008 4:26AM=-=>

    the proposed workaround of changing keyboard scheme in Tools→Options→Keyboard really worked for me

    <=-=Sep 5 2008 10:50AM=-=>

    Hello Brej K,

    Can you try the following: From the Tools→Options go to Environment→Keyboard and change the keyboard scheme to SQL Server 2000. This should bring back the back the CTRL+N behaviour to launch a new query editor.

    Thanks,
    Karthik

    Program Manager | SQL Server

    <=-=Sep 5 2008 11:26AM=-=>

    Hi Karthik,

    I have already given this workaround here. However, can…

  3. Report Builder Search does not search Description

    In Report Builder. The Search button above the list of Entities does not search any descriptions of fields. This should search this field so that a entity reference can be included in the Description which can be searched on. Also the search should include the reference field in the Serach Results for easier identification of the correct field.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Bugs  ·  Flag idea as inappropriate…  ·  Admin →
  4. SSMS find and replace alert (informational messages) settings cannot be accessed

    There is a problem concerning restoring settings in SSMS. When using the find functionality to find some text in a document in SSMS (both 2005 and 2008), when the last occurrence of the text is found in the document an alert window appears, notifying the user that "Find reached the starting point of the search". This alert window has a checkbox ("Always show this message"). If one unticks this checkbox, there seems to be no way to restore the original behaviour, i.e. to have an alert notifying the user that the end of the document has been reached.

    In Visual…

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Bugs  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 2

    <=-=Sep 29 2008 12:25PM=-=>

    Dear Andras Belokosztolszki,

    Thank you for reporting this problem. The Find and Replace Option setting is not present in SSMS to popup a dialog box indicating that the end of file has been reached after the checkbox is unchecked. However the status bar highlights the message after each time search reaches the starting point of the file. We will try to address this issue in a future release of SQL Server.

    Thanks,
    Karthik

    <=-=Aug 11 2011 12:27AM=-=>

    Hi

    Thanks for your request on SSMS to introduce an option to control display of informational messages for the “Find” operation.

    Given the work that would be involved in implementing this work, and triaging against our current set of deliverables, we do not think, we would be able to get to this in the near future.

    Having said that, we value your suggestions, and would like to…

  5. SQLGetTypeInfo ODBC API result set change for SQL 2008 ?

    It appears that the result set returned from SQLGetTypeInfo for the SQL_WVARCHAR type has changed between SQL 2005 and SQL 2008. Under SQL 2005, only "nvarchar" and "sysname" were returned; however, under SQL 2008, various date and time types are returned in addition to the nvarchar and sysname types.

    These extraneous types are breaking my app, and could cause problems for other applications that are not equipped to parse out non-character related types.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Bugs  ·  Flag idea as inappropriate…  ·  Admin →
  6. Costing issue with many-to-many merge joins

    The cost for the many-to-many merge joins that have large numbers of distinct values is to high. The optimizer chooses other algorithms which are slower in this case.

    Checked on SQL Server 2005 SP2 Dev and SQL Server 2008 Dev.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Bugs  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 1

    <=-=Aug 24 2008 12:22PM=-=>

    I agree that the cost estimates seem off – the merge join operator has a very high I/O estimate (313 in my case), whereas there is 0 I/O cost estimated for the hash join. But the I/O statistics don’t seem to show increased I/O for the merge join.

    While this is a costing bug, I think, the repro isn’t too compelling. The actual costs are close, and there is little impact from the wrong plan choice (the chosen plan was about 5% slower in my case). In general, choosing a hash join over a merge join is probably not a disaster (wrongly choosing a loop join could be more serious). If I find a modified repro that shows a real impact, I’ll post it.

    For the record, I also tried this with a unique clustered index on (c,rn) instead of the nonclustered index you…

  7. Default values of procedure/function parameters are not stored in catalog views

    Default values for parameters of stored procedures and functions are not stored in catalog views.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Bugs  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 16

    <=-=Aug 21 2008 11:07AM=-=>

    Hi,
    Thanks for your suggestion. We will consider it for a future version of SQL Server. Currently, the definitions for procedures, functions, triggers and views are stored as is in the catalog tables. They are used at the time of compilation/execution directly. So there is no way to get the default value for parameters of SPs, and functions other than parsing the text.


    Umachandar, SQL Programmability Team

    <=-=Aug 21 2008 1:15PM=-=>

    Hi,
    But when I use the function like this:

    select dbo.fnTest(default)

    SQL Server has to parse it to get the default value. I see no reason why SQL Server could not parse the definition during object creation and put the proper default values into system tables.

    Regards,
    Pawel Potasinski, SQL Server MVP

    <=-=Aug 22 2008 10:00AM=-=>

    Hi,
    I did not say that we cannot do the parsing and persist the default value…

  8. 2008 RTM, SSMS/Engine: Table designer doesn't script WHERE clause in filtered indexes

    When using the table designer to make changes to a table that would cause a recreate, the script will not include the WHERE clause portion of a filtered index.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Bugs  ·  Flag idea as inappropriate…  ·  Admin →
  9. Copy Paste Functionality in Report Builder Formula

    In all versions of SSRS Report Builder (including the 2008 RC1 Release) The ability to copy and paste in the formula dialog does not work. This is the one issue in which we have dozens of users being challenged by the work arounds and avoiding adoption of what is an excellent ad hoc reporting tool.

    This issue is also occurs for Report Model Developers through the design tools in BI Developer Studio.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Bugs  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 4

    <=-=Sep 11 2008 11:12AM=-=>

    Thanks for your feedback. The inability to copy/paste in the Formula dialog is a known issue and has not been implemented yet due to the need to handle embedded subformulas. Other customers have expressed desire for this feature as well. We hope to address this in a future release.

  10. Report Builder 2.0 cannot run reports using forms authentication

    Creating a report against a model and then trying to execute it against a report server configured for forms authentication does not work.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Bugs  ·  Flag idea as inappropriate…  ·  Admin →
  11. Data Collector : query statistics report duration can be WAY off

    I let a query run overnight (over 15 hours); when I viewed the query statistics report, it has divided the duration (in seconds) by 1000, showing that it ran in 56 seconds (56,000 milliseconds) instead of 56,000 seconds.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Bugs  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 1

    <=-=Nov 5 2008 6:27AM=-=>

    We are running into a similar problem, however in our case we consistenly see zero values for the snapshot_worker_time column for the first n runs where n is random. For example we tried with a very simple query like this one:
    SELECT *
    FROM Sales.SalesOrderHeader AS O1
    JOIN Sales.SalesOrderDetail AS O2
    ON O1.SalesOrderID = O2.SalesOrderID

    <=-=Nov 20 2008 5:26PM=-=>

    Aaron,

    This looks like an existing bug (bug # is 234755). The problem is an understating of execution time (and CPU) for in-progress queries. Only very-long-running queries that are still in-progress at the point of data collection are affected. Once a query has completed, the execution stats roll from dm_exec_requests into dm_exec_query_stats, and the correct stats will get harvested and reported after that point. (However, note that if you cancel a query — as in your repro scenario — the stats never get rolled…

  12. CREATE LOGIN fails when DB collation does not match AD collation

    Customer is using case sensitive SQL Server instances for their SAP rollout. When adding a Windows login to SQL Server, if the account was already used locally in some capacity on the cluster node, and the case of the login used was incorrect from what was used for Active Directory (upper or lower case), then SQL Server will fail in the attempt to add the Windows account.

    I found a solution in KB article 946358 to clear the local SID cache. This is working for them, however the customer has asked if we can submit a request to have the…

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Bugs  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 3

    <=-=Sep 11 2008 10:54AM=-=>

    Hi Joe,

    We’ve heard of this problem before and we’ll investigate a fix in a future release of SQL Server.

    Il-Sung.

    <=-=Oct 12 2009 5:02AM=-=>

    Hi ,
    Is this bug fixed now, if yes let me know how to resolve this issue plz..

    <=-=Oct 24 2009 12:24AM=-=>

    Hi. No, the issue still exists.

    <=-=Nov 12 2009 5:30AM=-=>

    A possible workaround for this issue is to use suser_sname(suser_sid(@name, 0)) as described in the BOL, http://msdn.microsoft.com/en-us/library/ms179889.aspx.

    Il-Sung.

    <=-=May 9 2013 5:40PM=-=>

    Hi, Thank you for providing your feedback. You suggestion is something that we have already started thinking about. It’s unfortunately too late to add any new features to this release but we’ll definitely consider it for a future version of SQL Server.

  13. RepeatColumnHeaders=True has no effect

    I have a table (tablix) with column headers. When the table detail rows span more than one page in Preview, the column headers should repeat on each new page. Setting the RepeatColumnHeaders property of the tablix to True has no effect.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Bugs  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 7

    <=-=Dec 12 2008 11:44AM=-=>

    Thank you for posting this feedback. The RepeatColumnHeaders property works if there is a column headings area in your report. Unfortunately, the default table does not contain a Column Header area so setting the property appears not to work. You can identify a Column Heading area if you see the double dotted lines in your tablix. A work around is to add a column group to your table. Then delete the group in the grouping pane but choose to leave the associated rows and columns. Now you can move your column header labels to the new cells above the double dashed lines and it should work.

    We know this is a usability issue and are looking at fixing it in the next release.

    <=-=May 12 2009 2:09PM=-=>

    Requiring the column headers to be included as 2008 wants us to seems a little strange. There…

  14. PIVOT fails with XML column in table even if the column is not referenced

    Executing PIVOT command fails if the source table includes columns that are not comparable e.g. XML. This failure occurs even if the XML column is not referenced in the query.

    This needs to be circumvented by by either using a subquery or CTE to limit the columns in the result that is passed to the PIVOT operation.

    There should be no need to do that.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Bugs  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 6

    <=-=Sep 23 2008 6:30PM=-=>

    Dear Adam

    Thanks for the report. This is currently by design since these uncomparable columns (also observable with other non-comparable types like the spatial types) serve for grouping purposes unless they are not included in the rowset. We are however looking into possible solutions for this and have filed it for investigation for now.

    Best regards
    Michael

    <=-=Sep 7 2009 9:55AM=-=>

    Forum discussion related to this issue http://social.msdn.microsoft.com/Forums/en-US/transactsql/thread/27286804-9411-4ac1-8a0d-165ce37c66db

    <=-=Mar 10 2011 6:18PM=-=>

    Hello Adam,

    I have resolved your request as duplicate of one below:

    http://connect.microsoft.com/SQLServer/feedback/details/127071

    We will track your feature request as part of that.


    Umachandar, SQL Programmability Team

  15. SSIS Data Mining Query transform seems to give wrong results

    Using a Decision trees model I can query the model in SSMS and the Excel AddIn giving me results that have varying Predict() and PredictProbability() values for the input dataset. With the SSIS Data Mining query transform I always get the same values in these prediction functions for the same dataset in all rows. I can copy + Paste the query into SSMS from the transform and it gives me different values to SSIS.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Bugs  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 1

    <=-=Oct 13 2008 4:47PM=-=>

    Thanks Allan for reporting the problem.

    in order to understand what is going on we need more information attached to this issue.

    Alternatively you can try and contact customer support immideate help.

    Thanks.
    The Analysis Services Team

    <=-=Oct 14 2008 6:13AM=-=>

    What information would you like.

    The problem is that the Excel add-in when running the prediction query gave me what I think was good results. It had a mix of getting things right and wrong and different probabilities.

    SSIS the DM Query transform for the exact same query told me that the expected prediction was the same for every single row as was the the PredictProbability() value.

    Does that help.

    <=-=Oct 14 2008 7:23AM=-=>

    Hi Guys. I have attached a file with some pictures. The data being used in XL and SSIS as the testing data is the same. The results are very…

  16. mimefilt.dll does not recognize HTML body mail in full Text Search

    mimefilt.dll works fine with pain text mail but not with HTML formatted body
    In this case it reads only the object of the mail

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    1 comment  ·  Bugs  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 1

    <=-=Nov 3 2009 10:53AM=-=>

    Thanks for the feedback.

    We succesfully indexed the email in question you attached in this bug. Can you please tell me what content you do not see later on at query time? we want to verify we index indexed all the content (including HTML formatted one), thus we need the query you are using that fails to you.

    Thanks!

    <=-=Nov 5 2009 1:59AM=-=>

    It is quite strange.
    Im using SQL 2008 10.0.2531 and a database with filestreams full text indexed.
    One of the file is the second mail I’ve just sent to you and I Cant’ find the text ‘vediamo’ inside the body (a freetext statament) … (“Pillole di Tecnologia, l’Hard Disk questo sconosciuto.eml”) but the oject of the mail is correct indexed, in fact ‘Pillole’ is found correctly and also ‘Hard Disk’
    Regards.
    Luca

    <=-=Jul 22 2013 5:36PM=-=>

    Thanks for the feedback.

    Unfortunately…

  17. SQL Server 2008 Service Pack 1 Cumulative Update 4 Hotfix to Version 10.00.2740.00

    SQL Server 2008 SP1 CU4 (http://support.microsoft.com/kb/973602) has been updated on 4th November 2009. On the the download page you can fetch an updated version (10.00.2740.00 instead of the original 10.00.2734.00)
    The filename (SQLServer2008SP1CUUpdatedRefKB976761) points to a knowledgebasearticle 976761 (http://support.microsoft.com/kb/976761) that does not (yet) exist.
    I could not find any information on the net about this revision. What happened?

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Bugs  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 8

    <=-=Nov 8 2009 12:43PM=-=>

    I wrote a pretty lengthy blog post about this today: http://is.gd/4Qk4o / http://sqlblog.com/blogs/aaron_bertrand/archive/2009/11/08/some-updates-to-sql-server-2008-sp1-cu4-you-may-not-have-noticed.aspx

    I think the PSS team should think hard about the servicing model and why files must be rushed out to the download servers before any documentation or announcement about those files has been released. There should never be a public reference to a KB that hasn’t been published yet, but letting me put 2470 on my servers (potentially by accident!), without first telling me what it is, is a nightmare waiting to happen.

    I realize that some of these tasks take longer than others, and are signed off on by different groups. But if the files are not a critical fix, then you should be able to talk about the update before it needs to be physically out there. And if the files are a critical fix, then all the more…

  18. SQL Server 2005 Slow Performance

    Thesql server 2005 is very slow for specific time duration particulary between 2 PM to 4 PM. It is caused even only very limited user connected. The server specification is IBM X-226 series (Xeon Processor, 2 GB RAM, 140 GB Hardddisk)

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Bugs  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 1

    <=-=Nov 20 2009 4:53PM=-=>

    Govindaraj,

    We don’t have enough information to understand this issue from your initial description. It could be a larger number of possible things that may involve application design, SQL Server, or both. In this forum, it’s important that you narrow down the issue by providing a detailed way to reproduce the problem. If you need additional help, consider contacting Microsoft Consulting Services (for design help) or Product Support Services (if you think the problem is a bug).

    Best regards,
    Eric

    <=-=Nov 22 2009 5:23PM=-=>

    Govind,

    For issues like this you will get better responses if you post this in the forums. As Eric mentioned there could be lot of factors that could be causing. SP2 and SP3 fixed lot of issues and you may want to look at those options too. Please feel free to post in here (http://social.msdn.microsoft.com/Forums/en-US/sqldatabaseengine/threads/) and one of…

  19. Can not edit data source in Report Model project

    I recently purchased and installed SQL 2008 (Standard) SP1 on a Win2008 server.
    I installed BIDS on my XP-SP2 workstation and have problems editing data sources. BIDS crashes with the error below.

    I can edit the Data Source created in step 2, as long as there is no Report Model in the project (i.e. skip step 4). As soon as a Report Model is added, I can't edit the Data Source.

    I used FilMon to watch what processes were accessing the DataSource file (ServiceCenter.ds). The only process that accesses the file is DevEnv.exe.
    When I don't have a Report Model…

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Bugs  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 3

    <=-=Nov 23 2009 7:49AM=-=>

    Thanks csharp_guru for the workaround! That helped.

    <=-=Nov 23 2009 3:13PM=-=>

    We have addressed this issue in the upcoming SQL Server 2008 R2 release, thank you for taking the time to file the bug.

    <=-=Nov 24 2009 6:42AM=-=>

    Fantastic!

    <=-=Mar 25 2010 1:47PM=-=>

    The workaround doesn’t work for me. I’m using TFS, so I tried unbinding. Same problem. I also tried leaving the DSV and SMDL files open. Any other suggestions?

    <=-=Mar 25 2010 1:51PM=-=>

    Found it. I had to open every DSV and SMDL in my project, even if they don’t use that DS.

  20. FISERV TAP - Shared Dataset Refresh Plan Doesn't Require Parameter Values

    According to BOL:

    "If the shared dataset has parameters, you must specify static default values for each parameter that is not marked read-only. Read-only parameters will always use the default value. To cache a shared dataset for multiple combinations of parameters, you must create a separate cache refresh plan for each combination of values. Parameters cannot contain references to other datasets."

    Yet, Report Manager doesn't require parameter values. When no parameter values are provided, the dataset query executes with empty parameter values.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Bugs  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 1

    <=-=Nov 17 2009 7:59AM=-=>

    Thanks Teo. Can you include the definition of the Shared Data Set you’re using in the bug report? I’m specifically interested to know if null is a valid value for your parameters.

    Thanks,
    -Lukasz

    <=-=Nov 17 2009 10:01AM=-=>

    Hi Lukasz

    You were right! All shared dataset parameters were configured to accept NULL. Once I re-configured them, I got the expected error message:

    “This data set requires a default or user-defined value for the data set parameter ‘@ReportYear’. To execute this data set, you must provide a parameter value. (rsDataSetParameterValueNotSet) "

    So, I’ld lke to change this bug report to a suggestion to NOT default the shared dataset parameters to NULL when you convert a dataset to a shared dataset to avoid confusion that specifying parameters is optional.

    <=-=Oct 24 2013 4:25PM=-=>

    Hello,
    Thank you for submitting this feedback. After carefully evaluating all of…

  • Don't see your idea?

SQL Server

Categories

Feedback and Knowledge Base