3 Replies Latest reply on Oct 14, 2010 3:40 AM by Jason B

    nsh error within BL Audit Job

      Hi,

       

      I have several audit jobs which use nsh scripts.  Randomly, some targets get errors like the following;

       

      Error Oct 12, 2010 10:51:41 PM Audit of target part 'Server Reboot Status//EventLogInfo/Rebooted' failed on target hostname: Caught exception running command - nsh -c "/D/DATA/ExtendedObjectScripts/script.nsh" hostname
      Error: Exception creating connection to: hostname; nested exception is:
      java.net.BindException: Address already in use: connect

      the script is nsh, and when run from a nsh shell manually connects and runs fine against targets Blade Logic reports the above error on.

       

      What does this error mean and why is it caused ?  It appears to be a conflict of socket on a target machine when the connect method (blcli under the covers ?) is called ?  If this is the case, how can you handle this in a script to tell it to try again ?  Additionally, if this is the case, how do you find a 'free time' to schedule the jobs for to ensure they won't conflict with another job.

       

      If this is not the case, why does this occur.  It is radmon and does not necessarily occor on the same targets each night.

       

      Thanks

       

      Jason.

        • 1. Re: nsh error within BL Audit Job

          Correction !

           

          Error Oct 12, 2010 10:51:41 PM Audit of target part 'Server Reboot Status//EventLogInfo/Rebooted' failed on target targethost: Caught exception running command - nsh -c "/D/DATA/ExtendedObjectScripts/script.nsh" targethost
          Error: Exception creating connection to: JOBSERVER; nested exception is:
          java.net.BindException: Address already in use: connect

          • 2. Re: nsh error within BL Audit Job
            Bill Robinson

            Something is bound to the port that you are trying to use on the appserver.

            • 3. Re: nsh error within BL Audit Job

              so, when we run a job, as this is the only thing that can be causing it, we should not use unlimited as the number of  targets to execute that job against ?  There must be a limit within BL that is causing this.  ie the job is launched from the app server, and yet it is the app server that is throwing this error, randomly, whilst it executes fine for some targets and not for others.  How can I find out what is bound to it, as it will only be something within Blade Logic.  these errors mean we can not trust any of the results the scripts return, as the nsh exit code of non zero does not mean the script, if allowed to run, would return a negative result.