SQL Server

Microsoft SQL Server 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 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. No dependies are stored for procedures called in INSERT-EXEC

    When a stored procedure includes a called to a second stored procedure through
    INSERT-EXEC, no dependencies are added. Neither do you get any warning message
    if the called procedure does not exist.

    2 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 →
  2. Reporting Services Aggregate Function

    The Aggregate function allows direct access to non-detail level values from an MDX source. It can work, however, only with the .Value property of the field. This makes it impossible to reflect "bubble-up" format calculations from the cube at any level other than the detail level. It also makes it impossible to display formatted values from the cube at any level other than the detail.

    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 →
  3. Wrong calculation for unary operator with many-to-many relationships

    This is a problem similar to ID 200152 and ID 200154. I open another feedback just to send another scenario that demostrate the issue from another point of view.
    The problem is that a parent-child dimension with unary operator does wrong calculation on '+' operator when applied to measures belonging to measure groups related through a many-to-many relationship with the parent-child dimension.

    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. Reference dimension through different data views

    If you define a referenced dimension for a measure group and the dimension is defined in a data source view different than the one containing the fact table, you get an error (dimension not defined) that does not specify the real problem (you need to add the dimension in the same data source view of the fact table).
    The problem happens even if the dimension is present in both DSVs but has different names - please note that the name is automatically generated by BI Studio and the only way to change it is to manually edit the XML in…

    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 →
  5. attempted to unlock a resource it does not own

    This happens to me all the time on various machines, various databases, both x64 and x32. Both standard and Development all SP1. Always only me on db.
    Error: 1203, Severity: 20, State: 1.
    Process ID 53 attempted to unlock a resource it does not own: OBJECT: 52:1182223562:0 . Retry the transaction, because this error may be
    caused by a timing condition. If the problem persists, contact the database administrator.

    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: 0

    <=-=Nov 29 2006 2:57PM=-=>

    Thanks for reporting this.
    While we are trying to repro this, can you make this database available to help pinpoint the issue, if possible?

    Best Regards
    Andrew
    [MSFT]

    <=-=Dec 8 2006 2:39PM=-=>

    Hi – we are currently tracking the same issue as a RFC. If you have the database available so we can repro this issue, this would be helpful. I am closing this one as a duplicate of the other issue we are tracking.

    Thanks,
    Tomer Verona
    SQL Server Development

    <=-=May 22 2007 3:26AM=-=>

    Ok, I just got this problem again on a brand new server with exactly the same style of changescript as below.
    Using SP2 this time

    <=-=Sep 5 2007 10:12AM=-=>

    This has been fixed in SP2 Cumulative Update.

    <=-=Sep 5 2008 12:58PM=-=>

    Hi,

    I am facing the same issue on SQL 2005 SP2 Enterprise Edition (64-bit) (9.00.3042.00).

  6. Partition Table using min/max functions and Top N - Index selection and performance

    Partitioned Tables performance issues - For select statements using Min and Max functions and Top N with ordering over an index.

    Poor performance is being detected for Queries on partitioned tables utilising the min and max functions and select Top N clause with "order by" matching columns of the index.

    The candidate index is either not being used or is being used to scan or seek all rows rather than a subset.

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

    We’ll send you updates on this idea

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

    Upvotes: 125

    <=-=Aug 10 2007 9:17PM=-=>

    Hi Microsoft, please confirm this optimizer problem by running the scripts, it is self evident and can be optimised significantly better.

    Eg. For a minimum, the optimal plan is to obtain the minimum for each partition using the index and then taking the minimum of all the (partition) results. The same principle can apply to maximum, and for Top N over an index …

    The plan should stand out as optimal, because the number of reads over the index is very small, and there will be one set of page reads (usually only one page) for each partition…. you can use the assumption that the number of partitions is significantly smaller than the number of rows (or even count the number of partitions), to estimate the cost (page reads) – which comes out at about N pages where N is the number of partitions.…

  7. New stored procedures not replicating to client if not accompanied by a new table and new proc associated w/ that table in pull me

    If you add a new stored procedure that is not related directly to a table, it will fail during the replication process.

    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: 0

    <=-=Nov 22 2006 9:23AM=-=>

    Thank you for letting us know of the issue. We will investigate it.

    Thank you.

    <=-=Nov 28 2006 12:45PM=-=>

    I am able to reproduce this error in house and we will take a look at it and triage it. Meanwhile please continue to use the workaround of adding a dummy table when you add a proc article.

    <=-=Jun 18 2009 11:17AM=-=>

    This problem is now resolved in SQl 2008 sp1 release. Thanks

  8. Moving system databases data and log files to separate directories makes Service Pack 2 fail

    I have had a lot of problems installing Service Pack 2 for SQL Server 2005 and I have found the same bug in the installation as in Service Pack 1 (reported as Bug ID 126672 and was supposed to be fixed in SP2).

    The service pack installation of the "Database Services" component fails and I get the following two error messages in the file SQL9HotfixKB921896sqlrunsql.msp.log:

    Failed to copy file "D:\SQLDATA\INSTANCENAME\mssqlsystemresource.ldf" to "D:\SQLDATA\INSTANCENAME\mssqlsystemresource1.ldf". Error 2

    Failed to copy file "D:\SQLDATA\INSTANCENAME\distmdl.ldf" to "D:\SQLDATA\INSTANCENAME\distmdl1.ldf". Error 2

    The problem seems to be that we have moved the data and logfiles…

    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

    <=-=Dec 7 2006 4:52PM=-=>

    This bug has been fixed in CTP3 of SP2.

    <=-=Feb 20 2007 12:52AM=-=>

    No it has NOT been fixed! I have just tried the offical SP2 and the EXACT same problem still exists! Have you even tried putting the Mssqlsystemresource log and data files into separate directories and applying SP2 because it does not work?!

    <=-=Jun 8 2007 7:40AM=-=>

    this still isn’t fixed as I downloaded Sp2 last night and got this problem. Any ideas when it might be?? It’s a bit of a pain having to do the workround when your got 20 servers to install Sp2 on!

    <=-=Jul 4 2007 10:28AM=-=>

    +1 This should be fixed as soon as possible.

  9. Incorrect database prefix is allowed in a query of a system view

    The query processor allows and ignores the specified database in a query of a system view even though no table or view in that database is in the query's FROM clause.

    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

    <=-=Dec 10 2007 2:13PM=-=>

    Hi Steve,

    This has been fixed in the 2008 release, but has not been fixed in the 2005 branch. We are keeping this defect alive for now and will triage it with other 2005 issues.

    Thanks for submitting this.

    Cheers,
    -Isaac

  10. MS Sales - Use of most of the available resources (specially, memory) being used during long cube processing and long running quer

    Need to use all available resources in 64 bit server (scale-up).
    I do not see most of the available resources (specially, memory) being used during long cube processing and long running queries.
    The configuration of the box is as follow:
    64 bit, HP RX5670 4 PROC @ 1.3 GHZ, Memory = 42 GB, Itanium 2
    Is this possible to consider this to Katmai ?
    Please contact me scyrus@microsoft.com for more info.

    Thanks,
    Surush

    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. Drillthrough on a multilingual cube return only the default language

    When I drillthrough on my cube which is a multilingual cube. Only the default language (english in my case) is returned instead of the user language.
    This bug is about the metadata and the result returned by the drill action

    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 →
  12. bug on the Data dimension security set up?

    tried to set up the security on the members of dimension attributes throught the "Dimension Data" tab of the "Edit role" form in the OLAP database.

    If I set up an "Allowed Set" directly on a "cube-dimension", everything works fine, whilst if I set up an "Allowed Set " on the "Database dimension" (which I suppose should apply to all the cubes), then, if the "cube dimension" name is different from the "database dimension" name, then I get an error in the string parsing (referring to the first element in the "Allowed Set", which cannot be found in the cube,…

    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

    <=-=Nov 22 2006 1:21PM=-=>

    Hi Vania,

    Yes – this is a known problem we’re looking at. One approach is to do something like we already have for cubes – currentcube is an abstraction that refers to the cube in context. We’re considering something similar like CurrentDimension.

    We hope to get to this is the next major release – buit no firm decision yet.

    Thanks,
    Richard (richtk@microsoft.com)

  13. MS Sales - In some queries, query performance of TOPCOUNT is slower that SP1!

    In some queries, query performance of TOPCOUNT is slower that SP1!
    I see this issue mostly with cold cache, warm cache is faster than SP1 though!
    Tested Build, �3024�

    Please contact me (scyrus@microsoft.com) for details and permission to the Dev test box

    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 →
  14. Save As does not always update Query tab with new filename

    I'm running SQL 2005 Management Studio SP 2 on Windows XP SP 2. What happens is when I use Save As to resave a file, the filename on the tab does not always change to that new name. This usually happens when I have opened a query I had saved in previously in Management Studio, made some changes, and resaved it. Just saving a query with one name, then using Save As to save it with something else does not produce the issue.

    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 →
  15. DisplayFolder in report model generated from a cube doesn't work sometimes

    Build a cube and generate a report model off of it using the Generate Model button in Report Manager. Under the following circumstances, it doesn't behave as expected...

    If there are no calculated measures assigned to a folder in the measure group (via the calculation properties dialog), then setting the DisplayFolder attribute for a physical measure in that measure group will have no impact on the report model generated from that cube. Likewise, if there are no physical measures assigned to a DisplayFolder, then assigning calculated measures to a folder via the calculation properties dialog will have no effect. I…

    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 →
  16. Setting and using Context_Info

    Currently,
    Contextinfo can be set by executing command "set contextinfo" and it can be seen in sysprocess.contextinfo or dmexecsession.contextinfo. But it is a binary value. I think this restriction makes it useless or difficult to implement.

    It should be nice
    if it can be set to a textdata
    if it can be seen as textdata (varchar or char type) in sysprocess and other dmvs
    if it can be seen and filtered in sql*profiler. This will be usefull especially for monitoring a specific part/module of an application.

    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

    <=-=Oct 11 2006 10:10AM=-=>

    Thanks for your feedback. It is a common need to be able to tag requests for monitoring / tracing purposes. We’ll consider this for future releases of SQL Server.

    - Christian Kleinerman

    <=-=Jan 20 2011 7:26AM=-=>

    The ability to collect and filter on context_info has been added in Extended Events for the next version of SQL Server.

    Thanks,
    Jerome Halmans
    SQLOS

    <=-=Jun 25 2015 1:12AM=-=>

    How is it possible to filter on context_info in Extended Events with a LIKE comparison?
    Since this is a binary value, one can only filter based on equality or inequality to another binary value.
    I would like to filter and fire only events for which the context_info contains the text “user=guy”.
    Is it possible?

    Thanks,
    Guy

  17. Incorrect output of xsi:nil=true

    We have a schema model built upon attributes rather than element values. For example:
    <family>
    <mother name="jane" age="33" />
    <father name="john" age="34" />
    <children>

    &lt;child name=&quot;sara&quot; grade=&quot;first&quot; /&gt;
    
    &lt;child name=&quot;thomas&quot; grade=&quot;fifth&quot; /&gt;

    </children>
    </family>

    With a schema created for the above xml data installed in sql 2005, all select queries against the table containing the xml column will return xsi:nil=true in the node names.

    Normally this is not a problem. However, couple that with VS2005 xmlserializer and the relevant objects will not be created.

    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 9 2006 3:16PM=-=>

    Dear Chris,

    I assume that the schema produced by xsd.exe erroneously generates a nillable content model for the type. Can you remove the nillable=“true” from the XML schema? You can use the XML-DML to do so…

    We are looking at improving the fidelity of our handling of the xsi:nil attribute for a future release, so only attributes that are provided in the instances are being returned. But for the time being, we recommend to only use nillable=“true” on elements that may have also non nilled content.

    Best regards
    Michael

    <=-=Oct 9 2006 5:51PM=-=>

    Unfortunately we are also using Biztalk 2006.

    Turning off nillable=true would cause Biztalk to not be able to process our xml files any more.

    If I set the elements to nillable=false or just flat remove that part of the declaration then Biztalk refuses to handle the xml file and…

  18. Analysis Services generates internal error when running managed code

    We have a stored procedure/UDF written in C# that we are accessing in various calculated measures in our cube. The sproc interates through the root members of a dimension, summing up the measure of interest in the tuples of those members in the current context until hitting a threshold (percentile calculation). The query intermittantly generates internal error exceptions. This may be related to Bug #480669

    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 →
  19. Format string issue with non calculated members (SSAS)

    I have some calculated members and some non calculated members with their format string set to "#,#". When I get them from SQL Server Management Studio, a very weird thing happens:


    • Calculated members are displayed correctly ( for example: 1.000.000)

    • Non calculated members are allways displayed in U.S. english ( for example: 1,000,000)

    I've read some people have exactly the opposite problem, and changing locale settings for SSAS service is working for them... OK, this workaround doesn't work for me.

    My regional settings are:
    - Windows XP PRO with regional settings set to SPANISH (os Spain)
    - Windows SQL Server…

    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: 5

    <=-=Oct 5 2006 12:20PM=-=>

    Thank you for reporting this issue. We will investigate and get back to you.

    <=-=Apr 18 2007 7:18AM=-=>

    This is not fixed in SP2 – both calculated and real measures now display the US format regardless of the client locale.

    <=-=Sep 19 2010 6:22AM=-=>

    You may override the LCID with the cube script to whatever you need. The list of supported LCID can be found on: http://msdn.microsoft.com/en-us/goglobal/bb964664.aspx.
    Unfortunately this approach applies to all clients alike and cannot be individualized (user-specific).

    LANGUAGE (THIS) = 2055 /* German – Switzerland */

  20. Empty Drillthrough result when unrelated dimension attribute has a default member

    Whenver a dimension attribute that is unrelated to a measure group has a default member which is not �All�, and the measure group has IgnoreUnrelatedDimensions set to false, no rows are ever returned when drilling through on the MeasureGroup (even if the value of the unrelated dimension is explicitly set to �All� at the cell from which the drillthrough takes place).

    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

    <=-=Oct 2 2006 12:55PM=-=>

    This issue got reported twice – keeping the first report.

    <=-=Mar 1 2011 12:24PM=-=>

    I can repro the same issue with 2008 SP2. In my case, drillthrough on some cells returns no rows when it should or returns repeating rows. I think after four years it’s time for a fix. The only workaround I have is to set IgnoreUnrelatedDimensions back to True.

  • Don't see your idea?

SQL Server

Categories

Feedback and Knowledge Base