SQL Server Performance

CUBE processing

Discussion in 'SQL Server 2005 Analysis Services' started by erajendar, Jul 25, 2007.

  1. erajendar New Member

    Hi – I made some changes to CUBE and started the processing. Ideally it should not take more than 2 hrs to process. But, it is under processing even after 15 hrs without any success/failure message. Can someone tell me what could be the Problem?

    --Rajendar E.


    ok
  2. ranjitjain New Member

    1>Open connection to Analysis server profiler and look for the queries.
    2>Also open performance monitor and look whether as is doing any read or write activities on hard disks.
    3>Also check in task manager whether msmdsrv.exe is consuming any processing power.

    Otherwise you can stop the processing task in between by restarting the as services.
    run sp_updatestats on OLTP db and rerun the processing with all above traces connected.
  3. satya Moderator

    Have you observed any errors or warning in SQL error log?
    Also are you running any specific DISTINCT user measure on thsi cube?

    Satya SKJ
    Microsoft SQL Server MVP
    Writer, Contributing Editor & Moderator
    http://www.SQL-Server-Performance.Com
    @http://www.askasqlguru.com/

    This posting is provided AS IS with no rights for the sake of knowledge sharing. Knowledge is of two kinds. We know a subject ourselves or we know where we can find information on it.
  4. erajendar New Member

    The following message is logged in SQL log:

    "The time stamp counter of CPU on scheduler id 3 is not synchronized with other CPUs."

    Apart from above I didn't see any message in logs.

    One more thing, If I process the changed measuregroup partition alone, it is getting processed successfully, but, If I do the Full Processing of CUBE it keep continuing as I described above:

    Thanks



    ok
  5. erajendar New Member

    The following message is logged in SQL log:

    "The time stamp counter of CPU on scheduler id 3 is not synchronized with other CPUs."

    Apart from above I didn't see any message in logs.

    I am not running any DISTINCT measure. Only thing I did is added one new column to the fact table and made the Regular relationship between fact and dimension. If I process the changed measuregroup partition alone, it is getting processed successfully, but, If I do the Full Processing of CUBE it keep continuing as I described above:

    Thanks



    ok
  6. satya Moderator

    Fyi on the error message:http://blogs.msdn.com/psssql/archiv...e-error-log-related-to-timing-activities.aspx though it talks about SQL 2005, check whether the hardware matches to it.

    KBAhttp://support.microsoft.com/kb/931279 fyi.

    Satya SKJ
    Microsoft SQL Server MVP
    Writer, Contributing Editor & Moderator
    http://www.SQL-Server-Performance.Com
    @http://www.askasqlguru.com/

    This posting is provided AS IS with no rights for the sake of knowledge sharing. Knowledge is of two kinds. We know a subject ourselves or we know where we can find information on it.
  7. erajendar New Member

    I ran the below command and processed the cube.
    exec sp_configure 'affinity mask', 0x00000003
    GO
    reconfigure
    GO
    Now, new problem begins :)

    OLE DB error: OLE DB or ODBC error: Protocol error in TDS stream; HY000; Protocol error in TDS stream; HY000; Protocol error in TDS stream; HY000; Communication link failure; 08S01; TCP Provider: An existing connection was forcibly closed by the remote host.
    ; 08S01.
    I have sqlserver Servicepack2 installed in my machine.
    Pls let me know the work around for it.
    Thanks
  8. satya Moderator

    What is the connection string you use in your data source?
    What is provider you are using?

    Have you tried testing connection?
    I would guess the problem you are having issues that might be related to some network issues with your network.
  9. erajendar New Member

    Test connection is successful. And more over CUBE is getting processed for dimensions and the issue is coming when it processing the measures
    thanks
  10. satya Moderator

    You should check the network related issues too in addition to local SQL resource management such as memory & disks.
  11. dineshasanka Moderator

    I would guess the problem you are having might be related to some network issues with your network.
  12. jbocachica New Member

    Hi, i received the same error but when my SSAS was processing the last partition on my cube does not matter the order, another issue is that i have my relational engine and SSAS in the same machine, then i have to discard network problems...
    Now, if i send that partitions alone, everything it's ok...
    Some ideas?
    I have a SAN with 1TB of free space, and my cube is not bigger that 1/4TB, 2 QuadCore procs and 16 GB RAM
    Regards
  13. satya Moderator

    Welcome to the forums.
    Did you check SQL error log f& event viewer log or any other hardware or software related errors?
    What is the service pack level on sQL Server?

Share This Page