Akshai

My feedback

  1. 4 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    2 comments  ·  SQL Server » Bugs  ·  Flag idea as inappropriate…  ·  Admin →
    Akshai commented  · 

    Thanks Michal, we will investigate and perhaps follow up with you if we cannot repro the issue.

    If this is a blocker for you, please go ahead and open a support case so that CSS is able to work with you directly on the investigation.

    Thanks,
    Akshai

  2. 1 vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    3 comments  ·  SQL Server » Bugs  ·  Flag idea as inappropriate…  ·  Admin →
    Akshai commented  · 

    See my last paragraph, there is a setting to remove/replace these invalid characters. There is no way to raise an error.

    Note that the default setting is to accept these characters, primarily because there is a perf overhead during processing to analyze every character that is read. For most columns, this is not an issue and we didn't want to pay the performance price for all columns when a tiny number of columns might end up having this rare problem.

    Akshai commented  · 

    This is expected behavior. Analysis Services uses the XMLA protocol, which is based on the SOAP protocol, on top of the XML specification. The XML specification (https://www.w3.org/TR/REC-xml/) disallows certain characters (https://www.w3.org/TR/REC-xml/#charsets).

    Character Range
    [2]
    Char
    ::=
    #x9 | #xA | #xD | [#x20-#xD7FF] | [#xE000-#xFFFD] | [#x10000-#x10FFFF]
    /* any Unicode character, excluding the surrogate blocks, FFFE, and FFFF. */

    Therefore, Analysis Services is unable to support characters that are outside the ranges allowed by the XML specification.

    There is a setting on Multidimensional dimension attribute objects which allows you to decide what to do about invalid XML characters -- you can choose to keep them, remove them or replace them with a '?' character:
    https://docs.microsoft.com/en-us/dotnet/api/microsoft.analysisservices.invalidxmlcharacters?view=analysisservices-dotnet

    Thanks,
    Akshai

  3. 1 vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    1 comment  ·  SQL Server » Suggestions  ·  Flag idea as inappropriate…  ·  Admin →
    Akshai commented  · 

    This feature is not needed in Tabular models since it has a very different design. Multidimensional measure groups need to be fully reprocessed when dimension objects are fully reprocessed, because the surrogate keys of the members are changed when a dimension by the full processing operation.

    In Tabular models, processing a dimension only requires the relationship objects in the fact table to be rebuilt. This is a very fast operation, and it doesn't require the fact table to be fully reprocessed.

    In effect, full processing of a dimension table in Tabular models behaves very similar to ProcessUpdate in MOLAP.

    Thanks,
    Akshai

  4. 1 vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    1 comment  ·  SQL Server » Bugs  ·  Flag idea as inappropriate…  ·  Admin →
    Akshai commented  · 

    Hi Michal, thanks for reporting the issue. I've filed a bug to investigate it.

    If you need this fixed more urgently, I would recommend opening a support case.

    Thanks,
    Akshai

  5. 122 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    18 comments  ·  SQL Server » Suggestions  ·  Flag idea as inappropriate…  ·  Admin →
    Akshai commented  · 

    This is something being actively looked at, but we don't have a timeline to share at this point.

  6. 1 vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    3 comments  ·  SQL Server » Bugs  ·  Flag idea as inappropriate…  ·  Admin →
    Akshai commented  · 

    Typically you would create a support case based on your Microsoft support contract https://support.microsoft.com/en-us/hub/4343728/support-for-business

    E.g. do you have an enterprise support contract, or you can open a pay-per-incident support case.

    Akshai commented  · 

    Hi Chris, you should open a support case for issues like this. Its not really possible to diagnose problems like this based on a forum post -- CSS will gather more information as necessary...

    This particular issue looks a bit similar to a problem where the processing thread pool shows a spike in the number of threads, and it becomes very slow. If you see this symptom, then a hotfix will become available in the next CU (apparently SP1 did not have this problem, so that might be another hint that it could be the same issue).

    Thanks,
    Akshai

  7. 2 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    2 comments  ·  SQL Server » Bugs  ·  Flag idea as inappropriate…  ·  Admin →
    Akshai commented  · 

    SQL Server 2012 is out of mainstream support, so I'd look at upgrading to a supported version of the product (preferably 2016/2017 since 2014 is also going out of support this year)...

  8. 0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    2 comments  ·  SQL Server » Suggestions  ·  Flag idea as inappropriate…  ·  Admin →
    Akshai commented  · 

    I'm afraid that this is expected. The engine can leverage the knowledge that both branches of the inner iif are guaranteed to be null if the [Internet Average Unit Price] value is null. It can take advantage of that to only evaluate the iif for the cells where the [Internet Average Unit Price] is not null.

    For the first scenario, the inner iif results in a 'dense' space because it returns a non-null value for all the cells, and it would be computed for all cells. Theoretically, the outer iif should guarantee that the inner iif only needs to be executed for cells that are not null, but that information isn't propagated down.

    The alternative approach is a great way to leverage the engine optimization. It would be nice to not need it, but it is a pretty straightforward solution.

    Thanks,
    Akshai

  9. 0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    under review  ·  3 comments  ·  SQL Server » Bugs  ·  Flag idea as inappropriate…  ·  Admin →
    Akshai commented  · 

    You would contact your Microsoft support representative (many larger enterprises have support contracts and you can use that to request assistance). This page may help but I've never used it myself :)

    https://support.microsoft.com/en-us/getsupport?oaspworkflow=start_1.0.0.0&wf=0&wfName=productselection&gprid=1044&st=1&wfxredirect=1&sd=gn&ccsid=636554509977088654

    This type of issue is difficult to diagnose without the tools that CSS can use to gather more data from your environment.

    Thanks,
    Akshai

    Akshai commented  · 

    Hi Ivan,

    Have you tried posting this on the msdn forums? There should be other customers and support folks who can help you there. Otherwise, you should probably open a support case.

    The feedback site is meant more for suggestions and new feature requests...

    Thanks,

    Akshai

  10. 3 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    under review  ·  7 comments  ·  SQL Server » Bugs  ·  Flag idea as inappropriate…  ·  Admin →
    Akshai commented  · 

    Hi Alex, please open a support case to investigate and fix this issue. It looks like a defect related to the name resolution error in the calc member at line 493 in the script -- if you're able to avoid the failures, you should be able to work around this problem.

    Customer support will need to handle this because it needs a proper repro -- the minidump is not enough.

    Thanks,
    Akshai

  11. 8 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    under review  ·  6 comments  ·  SQL Server » Bugs  ·  Flag idea as inappropriate…  ·  Admin →
    Akshai commented  · 

    Hi John, we believe this is a bug with Excel 2016, and they're working on investigating and fixing the issue.

    Due to the release cadence of Office, I can't guess when you would receive an update containing the fix -- hopefully within a few months...

    Thanks,
    Akshai

Feedback and Knowledge Base