CheckDB not completing because of resource errors

  • I am testing a new SQL 2016 Std build in AWS. Part of the testing includes running checkdb on all the databases.  Two of the databases failed repeatedly with errors like 701 - "There is insufficient system memory in resource pool 'internal' to run this query.", and errors 802 and 17300.  With max memory set at 24GB, there should be plenty of resources for checkdb to complete.  These databases complete checkdb weekly on our current production box with the same amount of memory.

    As a test I cut the maximum memory in half -- 24gb reduced to 12gb -- and the checkdb passed repeatedly.

    I don't know what to do with this result. Any thoughts on what to do next?

    Thanks,
    Dale

  • DaleB - Friday, October 19, 2018 6:50 AM

    I am testing a new SQL 2016 Std build in AWS. Part of the testing includes running checkdb on all the databases.  Two of the databases failed repeatedly with errors like 701 - "There is insufficient system memory in resource pool 'internal' to run this query.", and errors 802 and 17300.  With max memory set at 24GB, there should be plenty of resources for checkdb to complete.  These databases complete checkdb weekly on our current production box with the same amount of memory.

    As a test I cut the maximum memory in half -- 24gb reduced to 12gb -- and the checkdb passed repeatedly.

    I don't know what to do with this result. Any thoughts on what to do next?

    Thanks,
    Dale

    I have had this error recently, for the first time, on a 2017 instance. It's only a QA box & I haven't tried to resolve it yet, so I am interested to hear what others may say about it.

    The absence of evidence is not evidence of absence
    - Martin Rees
    The absence of consumable DDL, sample data and desired results is, however, evidence of the absence of my response
    - Phil Parkin

  • DaleB - Friday, October 19, 2018 6:50 AM

    I am testing a new SQL 2016 Std build in AWS. Part of the testing includes running checkdb on all the databases.  Two of the databases failed repeatedly with errors like 701 - "There is insufficient system memory in resource pool 'internal' to run this query.", and errors 802 and 17300.  With max memory set at 24GB, there should be plenty of resources for checkdb to complete.  These databases complete checkdb weekly on our current production box with the same amount of memory.

    As a test I cut the maximum memory in half -- 24gb reduced to 12gb -- and the checkdb passed repeatedly.

    I don't know what to do with this result. Any thoughts on what to do next?

    Thanks,
    Dale

    Dont do all at once. Try doing with some dbs and next other batch of dbs

  • Tthank-you for your reply.  

    I am using the ola.hallengren scripts, so the databases are processed in sequence.  The same script i have been running on my production server for months without errors.

  • DaleB - Monday, October 22, 2018 9:30 AM

    Tthank-you for your reply.  

    I am using the ola.hallengren scripts, so the databases are processed in sequence.  The same script i have been running on my production server for months without errors.

    Is the testing server and production having same size and number of databases? If not, try split checckDB.

    Muthukkumaran Kaliyamoorthy
    https://www.sqlserverblogforum.com/

  • Check the settings of min and max server memory. If the value of maximum server memory is close to min server memory, then increase the max server memory value. Check virtual memory paging file and increase the size of the file, if possible. Following commands can also be run to free the memory:

    • DBCC FREESYSTEMCACHE
    • DBCC FREESESSIONCACHE
    • DBCC FREEPROCCACHE

Viewing 6 posts - 1 through 5 (of 5 total)

You must be logged in to reply to this topic. Login to reply