IV67984: RTEMS SLOW SHUTDOWN DUE TO STOP EVENT PER SIT SENT TO HUB TEMS

IBM Tivoli Monitoring 6.3.0 Fix Pack 6 (6.3.0-TIV-ITM-FP0006)

APAR status

• Closed as program error.

Error description

• Severity: 3
  Approver:BEH
  Compid:  5724C04MS Tivoli Enterprise Management Server
  
  Abstract:
  RTEMS slow shutdown due to stop event per Sit sent to HUB TEMS
  
  Environment:  ITM 6.30 FP1 on WindowsPage 20 of 21
  
  Problem Description:
    When a RTEMS is shutdown, for each running
  Situations a 'stop' event is being sent to HUB TEMS. This is
  inefficient since regardless of how many Situations are running
  at RTEMS they will all be stopped given the RTEMS is shutting
  down; sending a single, 'master' stop event to HUB TEMS would
  be a more logical, more efficient means to inform HUB TEMS that
  all the running Situations at RTEMS are being stopped.
  
  If there is a slow network link between RTEMS and HUB TEMS the
  aggregate processing time for the collection of individual
  'stop' events is exacerbated and in turn results in a
  protracted time to complete the RTEMS shutdown.
  

Local fix

Problem summary

• Remote monitoring server may be very slow to complete shutdown
  processing.
  
  
  When a remote monitoring server is shutdown, for each Situation
  running at remote monitoring server a 'stop' event is being sent
  to HUB monitoring server. This is inefficient since regardless
  of how many Situations are running at remote monitoring server
  they will all be stopped because the remote monitoring server is
  shutting down.
  
  If there is a slow network link between remote monitoring server
  and HUB monitoring server the aggregate of data transfer time
  over the network plus processing time for the collection of
  individual 'stop' events is exacerbated and in turn results in a
  protracted time lapse before remote monitoring server completes
  shutdown.
  

Problem conclusion

• A solution was implemented that when remote monitoring server
  receives a shutdown request it sends a single, 'master' stop
  event to HUB monitoring server and the HUB monitoring server
  then creates a 'stop' event for each situation that was running
  on the remote monitoring server.  This solution requires remote
  monitoring server to only have to send a single 'stop' event to
  HUB monitoring server.  It is a more efficient means to inform
  HUB monitoring server that all the running Situations at remote
  monitoring server are being stopped.
  
  The fix for this APAR is contained in the following maintenance
  packages:
  
    | fix pack | 6.3.0-TIV-ITM-FP0006
  

Temporary fix

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

• APAR is sysrouted TO one or more of the following:

  OA49286

Fix information

• Fixed component name

  TEMS

• Fixed component ID

  5724C04MS

Applicable component levels

• R630 PSY

     UP