1 2 3 Previous Next 32 Replies Latest reply on Aug 24, 2018 6:33 AM by ryan nicosia Go to original post
      • 30. Re: Remedy 9.1 Upgrade Recommendations
        Stefan Hall

        Hi Johannes,

        patch 001 for 9.1SP2 ... good to know. I will check my emails tomorrow.

        We run SP2 as is, currently without Hotfix Version Stamp: 9.1.02 201701050505. Once bitten, twice shy - most new BMC HF brings new errors. We implement some additional QA checks, especially for patches and the new major CU_ALL HFs.

         

        Back to SP2 patch 001 - I've checked BMC documentation (known & correctes issues), but no SP2.001 information, but I can download it. Hope we will see some detailed information early next week.

         

        I remember that normalization should be patched, because of incorrect handle of null values. I've documented our findings at office. I'll take a look tomorrow.

        :-) Stefan

        • 31. Re: Remedy 9.1 Upgrade Recommendations
          Stefan Hall

          Hi ryan nicosia,

          We are updating to 9104 and I have a similar problem.

          Ultimately, my team on the ground found that any form, menu, or query that was leveraging AST:Attributes form was resulting in wide open table scans on 3 and 4 way joins.  Where results that should have been 100,000 rows were 17 and 18 million rows.  This resulted in TempDB growth to the point that the database ran out of allocated space and crashed.

          How did you solve the problem and optimize the joins with AST:Attributes?

          • 32. Re: Remedy 9.1 Upgrade Recommendations
            ryan nicosia

            Stefan,

             

            The root cause was thousands of records in AST:Attributes that all had a ReconID of 0.  This results in a many to many relationships and causes what appears to be a wide open table scan in SQL when just launching asset management console.   Our fix was finding all the records in AST:Attributes where ReconID = 0 and deleting them.  Fixed the problem for us and the correct records got recreated after an ID job was ran.

            3 of 3 people found this helpful
            1 2 3 Previous Next