23 Replies Latest reply on Jun 16, 2014 9:04 AM by aLTeReGo

    Rather annoying alert status in Solarwinds MP 2.1.0.0

    Jonorowe

      Hello guys

       

      I have been trying to integrate our Solarwinds SAM Monitors into our SCOM 2012 SP1 environment.

       

      Unfortuantely, a huge number of component monitors sit in a warning state with the error status "monitor has been forced into a reset state". The SCOM dashboard does not render the status of the SAM tests properly. the SCOM dashboard does however render performance metrics OK.

       

      I see this has been a known issue related to permissions in the past. having followed the recommendations to fix it (add the SDk, Read and write accounts into the local admins), I still have the same problem. Annoyingly, it seems to work for some tests and not for others.

       

      Is there anything else I can try to get this integration working? I did log a ticket with the support guys (523891) and provide the diags from the MP but they couldnt help. They asked me to post here.

       

      Running Solarwinds SAM 5.0.1, NPM 10.4.1,

      SCOM 2012 Sp1 Cu1.

      Solarwinds MP 2.1.0.0

       

        • Re: Rather annoying alert status in Solarwinds MP 2.1.0.0
          aLTeReGo

          Have you ensured that the SCOM service accounts (datareader, datawrite, mgmtsvraction) were added properly to the local admins group of the SCOM RMS box?

          • Re: Rather annoying alert status in Solarwinds MP 2.1.0.0
            lukas.cerny

            Hello,

            can you check that in the performance view, when you select the Status counter, it changes its value accordingly to the entity status (0/1/2). If this performance counter is changing its value, it means that for some reason the SCOM is not correctly updating the entity status. In that case I recommend to first try to reset health of those affected entities (after health is reset, status should go to up and during next poll of data status should be changed to a correct one). If this doesn't help, try to recalculate the health (that will set status to warning and during next poll it should change to correct one). If neither resetting nor recalculating health doesn't help, run the configuration wizard, remove nodes which are experiencing the issue, save the changes. Then wait some time to ensure that these nodes were removed from the SCOM as well. Then run configuration wizard again and re-add the nodes.

            • Re: Rather annoying alert status in Solarwinds MP 2.1.0.0
              hingy

              I have installed NPM with SCOM 2012 SP1 Management Pack at a customer and am seeing a similar issue. The product installs and passes data to SCOM, but most of the nodes, interfaces and other entities show as Not Monitored. I have an open case with Support at the moment, but in the meantime have uninstalled everything and started again with exactly the same results. Here is a screen shot of the Network Nodes View. I have looked in Orion NPM and the nodes that show as monitored do not have anything different to the ones that are not, and the data is showing fine in the Orion Web Console. Also, the data is present in the config.xml and performancedata.xml files. The alerts show in the SCOM Active Alerts View, but not in the Solarwinds Alerts View. It looks like the OrionMP service, or the connector is not passing data to the SCOM SDK. I also get the graphs showing properly, just like Jono.

               

              Nodes.jpg

              • Re: Rather annoying alert status in Solarwinds MP 2.1.0.0
                hingy

                Jono

                 

                I may have solved the issue.

                 

                The first thing that I noticed was that on all SCOM Management servers, except for the server on which I have installed the Solarwinds Management Pack, I was receiving errors in the Operations Manager Log such as this:

                 

                 

                "Loading managed module type in assembly "SolarWinds.Orion.SCOM.DataSources, Culture="", PublicKeyToken="e3753070c0e63e97", Version="2.1.0.0"" with type name "SolarWinds.Orion.SCOM.DataSources.OrionDataEmitter" failed with error "The module assembly "SolarWinds.Orion.SCOM.DataSources, Culture="", PublicKeyToken="e3753070c0e63e97", Version="2.1.0.0"" could not be loaded.  The exception was: \nCould not load file or assembly 'SolarWinds.Orion.SCOM.DataSources, Version=2.1.0.0, Culture=neutral, PublicKeyToken=e3753070c0e63e97' or one of its dependencies. The system cannot find the file specified..".

                 

                This may be because the type or assembly could not be found or the type does not have the MonitoringModuleAttribute.

                 

                Workflow: SolarWinds.Orion.SCOM.Rule.ResponseTimeAlert

                Instance: ERS-8610

                Instance ID: {7CA30B0F-BCC6-4EEA-EE99-29713724D0B3}

                Management Group: ***

                Full Exception Text:

                Microsoft.EnterpriseManagement.HealthService.HealthServiceException: The module assembly "SolarWinds.Orion.SCOM.DataSources, Culture="", PublicKeyToken="e3753070c0e63e97", Version="2.1.0.0"" could not be loaded.  The exception was: \nCould not load file or assembly 'SolarWinds.Orion.SCOM.DataSources, Version=2.1.0.0, Culture=neutral, PublicKeyToken=e3753070c0e63e97' or one of its dependencies. The system cannot find the file specified.. ---> System.IO.FileNotFoundException: Could not load file or assembly 'SolarWinds.Orion.SCOM.DataSources, Version=2.1.0.0, Culture=neutral, PublicKeyToken=e3753070c0e63e97' or one of its dependencies. The system cannot find the file specified.

                   at System.Reflection.RuntimeAssembly._nLoad(AssemblyName fileName, String codeBase, Evidence assemblySecurity, RuntimeAssembly locationHint, StackCrawlMark& stackMark, IntPtr pPrivHostBinder, Boolean throwOnFileNotFound, Boolean forIntrospection, Boolean suppressSecurityChecks)

                   at System.Reflection.RuntimeAssembly.InternalLoadAssemblyName(AssemblyName assemblyRef, Evidence assemblySecurity, RuntimeAssembly reqAssembly, StackCrawlMark& stackMark, IntPtr pPrivHostBinder, Boolean throwOnFileNotFound, Boolean forIntrospection, Boolean suppressSecurityChecks)

                   at System.Reflection.RuntimeAssembly.InternalLoad(String assemblyString, Evidence assemblySecurity, StackCrawlMark& stackMark, IntPtr pPrivHostBinder, Boolean forIntrospection)

                   at System.Reflection.RuntimeAssembly.InternalLoad(String assemblyString, Evidence assemblySecurity, StackCrawlMark& stackMark, Boolean forIntrospection)

                   at System.Reflection.Assembly.Load(String assemblyString)

                   at Microsoft.EnterpriseManagement.HealthService.Internal.ManagedHostHelper.LoadModuleType(String moduleAssemblyName, String moduleTypeName)

                   --- End of inner exception stack trace ---

                   at Microsoft.EnterpriseManagement.HealthService.Internal.ManagedHostHelper.LoadModuleType(String moduleAssemblyName, String moduleTypeName)

                   at Microsoft.EnterpriseManagement.HealthService.Internal.ManagedHostHelper.Microsoft.EnterpriseManagement.HealthService.Internal.IManagedHostHelper.LoadAndInitializeModule(String moduleAssemblyName, String moduleTypeName, UInt32 countInputPorts, Boolean doesModuleHaveOutput, String configuration, Byte[] stateBlock, UInt32 stateBlockLength, Object& moduleReference, Object& moduleInputReference, Object& moduleInputBatchedReference)"

                 

                 

                This meant that all our SCOM Management servers were attempting to access the assembly SolarWinds.Orion.SCOM.DataSources. So I installed the Solarwinds Management Pack msi on all our Management Servers. I did not run the Configuration Wizard, and disabled the Orion MP Service. Immediately, devices started showing in the Solarwinds views as Warning. However, I had been tinkering a lot with permissions and by now the Health Service on the Management Servers kept becoming unhealthy. I had given the SolarWindsOrionDatabaseUser dbo access to the OperationsManager database and this had killed AzMan. Removing this access and giving it db_datareader, and then restarting the System Center Data Access Service and System Center Management Service on the RMS Emulator box immediately fixed the Health Service and the nodes started receiving data from Orion! NodesMonitored.jpg

                 

                I am still investigating whether the Orion MP Service needs to be running and whether I can control the behaviour of the data import using overrides, but I think I am getting there.

                • Re: Rather annoying alert status in Solarwinds MP 2.1.0.0
                  hingy

                  Hi Jono

                   

                  In my last post, I thought I got to the bottom of it. However since the monitors started working, I still get a large number of them remaining in the Warning state, just like yours. I will continue to investigate.

                   

                  Thanks

                   

                  Hingy

                  • Re: Rather annoying alert status in Solarwinds MP 2.1.0.0
                    sowens82

                    Hi,

                     

                    Was a solution ever provided for this?  Seems we are in the same position:

                     

                     

                    Loading managed module type in assembly "SolarWinds.Orion.SCOM.DataSources, Culture="", PublicKeyToken="e3753070c0e63e97", Version="2.1.0.0"" with type name "SolarWinds.Orion.SCOM.DataSources.OrionDataEmitter" failed with error "The module assembly "SolarWinds.Orion.SCOM.DataSources, Culture="", PublicKeyToken="e3753070c0e63e97", Version="2.1.0.0"" could not be loaded.  The exception was: \nCould not load file or assembly 'SolarWinds.Orion.SCOM.DataSources, Version=2.1.0.0, Culture=neutral, PublicKeyToken=e3753070c0e63e97' or one of its dependencies. The system cannot find the file specified..".

                     

                    Thanks

                    • Re: Rather annoying alert status in Solarwinds MP 2.1.0.0
                      Jonorowe

                      Still an issue - memory dumps from our RMS have been with the SW devs for a month now.

                       

                      In this time we have replaced quite a few of our SAM tests with SCOM tests - soon we wont need to integrate SW at all.