4 Replies Latest reply on Mar 25, 2008 8:29 AM by RickDavie

    Inventory Service Crashing



      The Inventory Service on our Core server is crashing every couple of weeks or so.  I checked the error log and there is a set of reoccuring errors being reported by the Inventory Service within a minute of each other before the crash:



      First error: "the size of AppSoftwareSuites.PRODUCTID is too small, increase its size by at least 10."






      Second error: "Database exception:




      Multiple-step operation generated errors. Check each status value.


      Update Table:AppSoftwareSuites Column:PRODUCTID PK:Easy Grade Pro

      Thread ID: 12576."






      Third error: "Could not read the metaschema.  Make sure the database is a valid LANDesk 7.5 or greater version."






      Fourth error: "Database exception: ."






      The third and fourth errors occur less frequently then the first two, which occur quite often without causing a crash.  However, the group of four errors often occur all within the same minute of each other before the Inventory Service crashes.







      It appears that every time the third and fourth errors occur, the Inventory Scanner on the Core server reports: "LDIScn32: The inventory server OTSCYLANDESK did not respond.".  Clients can not send in inventory scans until the service is started again manually.  I have already set the service to restart on every failure  (Computer Management>Services>LANDesk Inventory Service).  I have also set configure>Inventory>Advanced Settings>set "DB Error Recovery" to "10000".  After restarting the service manually, the service runs for a week or two and then crashes again.






      I have also seen at least one occasion where the last error reported by the Inventory Service says "Out of memory".   The server has 3.5GB of memory.






      I should mention that we are currently getting approx. 1500 rejected delta scans due to Deepfreeze being on half of our computing environment.  We have not yet turned off Delta scanning as we are going to try to resolve the issue without turning off Delta scanning.  Every other day or so, I empty the ErrorScan folder on the Core server.  Could this be causing the problem?