1 Reply Latest reply on May 8, 2017 6:58 AM by Jeff Strauss

    Upgrade fails when restarting administrative server on Workers

    Andrew Dobbins

      Hello All,

       

      We are currently down in our production environment so any help would be greatly appreciated!

       

      This past weekend we attempted to install 9.0.21 and uninstall 9.0.0. In the process everything went smooth until the process where the primary builds the bundles for the workers and installs them. At this point it attempts to restart the service and fails to ever get a response. We tried reverting back and ran into the same issue. I suspect something with our server configuration is to blame but I have no answer at this point. I am attaching some lines from the logs if anyone can help it would be greatly appreciated!

       

      A big question - our IT team recently disabled NetBIOS and multicasting - does the install process use either? Also, does NetBIOS need to be running for Tableau to function properly?

       

      Thanks!

       

       

      2017-05-07 19:13:44.381 -0400_EXACT_192.168.3.5:infodashboard_:_pid=2388_0x859ea42__user=__request=__ infodashboard:

        Status: STOPPED

      192.168.3.7:

        com.tableausoftware.tabctl.dll.NoServiceError: No service named 'tabsvc' is installed (1060).

      192.168.3.12:

        com.tableausoftware.tabctl.dll.NoServiceError: No service named 'tabsvc' is installed (1060).

       

       

       

       

      2017-05-07 19:12:05.800 -0400_DEBUG_192.168.3.5:infodashboard_:_pid=1068_0x859ea42__user=__request=__ Failed to load configuration from C:/ProgramData/Tableau/Tableau Server/data/tabsvc/config/workgroup.yml

       

       

      2017-05-07 19:11:25.180 -0400_DEBUG_192.168.3.5:infodashboard_:_pid=10084_0x859ea42__user=__request=__ Exception contacting worker 192.168.3.7, DRb::DRbConnError: connection closed

       

       

       

       

      2017-05-07 19:11:25.196 -0400_ERROR_192.168.3.5:infodashboard_:_pid=10084_0x859ea42__user=__request=__ Error: Workers not responding: 192.168.3.7

        • 1. Re: Upgrade fails when restarting administrative server on Workers
          Jeff Strauss

          See if this helps.  I'm not sure it's the same circumstance or not, and by the way the issue has been fixed (I forget what release, but surely it's been fixed since 9.0.x).

           

           

          Have a look at the worker task scheduler and see if there's any pending tasks out there for the restart of the worker tabsvc.  We ran into this prior where the primary would issue a run of a task on the worker, the task would go into pending status, the worker would never restart and then the primary would fail with "worker not responding".  The temp workaround was during the install we would have to "babysit" the worker task scheduler and when we saw it pop-up in pending status, then force it to run.