IV69475: RESTART DISCO PROCESS OPTION ON BUT DISCO FAILS TO RESTART WITH SOLID DB ERROR

4.1.1-TIV-ITNMIP-Linux-FP0002
4.1.1_README_NLS_FP0002
4.1.1_README_FP0002
4.1.1_README_NLS_FP0001
4.1.1_README_FP0001
4.1.1-TIV-ITNMIP-Linux-FP0001
IBM Tivoli Network Manager IP Edition 4.1.1 Fix Pack 2, 4.1.1-TIV-ITNMIP-FP0002

APAR status

• Closed as program error.

Error description

• If the restart discovery process option is set to 1 then on
  each full discovery the process should be stopped and then
  restarted by ncp_ctrl. The issue is that occasionally the disco
  process can be killed by ncp_ctrl but doesn't fully die
  immediately. Ctrl then tries to restart the discovery process
  but this fails as the old ncp_disco still has open the defined
  solidDb port. The result is the new discovery fails to start
  and the old discovery eventually does die resulting in no
  discovery process. The solid Db error is a common one:
  
  Wed Feb 11 22:09:53 2015  Warning: Error found in file
  CNcpSolidDbh.cc at line 602 - Unable to start solid server as
  port number:12516 is already in use. Please check DbLogins
  
  Wed Feb 11 22:09:53 2015  Warning: An unexpected nil pointer
  reference has been found. found in file CNcpDbFactory.cc at
  line 1027
  Wed Feb 11 22:09:53 2015  Warning: Error found in file
  CDiscoStitcherMgr.cc at line 1442 - Failed to connect to
  embedded db for dbIdDNCIM
  Wed Feb 11 22:09:53 2015  Warning: Error found in file
  CNcpSolidDbh.cc at line 602 - Unable to start solid server as
  port number:12516 is already in use. Please check DbLogins
  

Local fix

• stop and start manually the process
  

Problem summary

• ****************************************************************
  * USERS AFFECTED:                                              *
  * ITNM 411 users                                               *
  ****************************************************************
  * PROBLEM DESCRIPTION:                                         *
  * If the restart discovery process option is set to 1 then on  *
  * each full discovery the process should be stopped and then   *
  * restarted by ncp_ctrl.The issue is that occasionally the     *
  * disco                                                        *
  * process can be killed by ncp_ctrl but doesn't fully die      *
  * immediately. Ctrl then tries to restart the discovery        *
  * process                                                      *
  * but this fails as the old ncp_disco still has open the       *
  * defined                                                      *
  * solidDb port. The result is the new discovery fails to start *
  * and the old discovery eventually does die resulting in no    *
  * discovery process. The solid Db error is a common one:       *
  ****************************************************************
  * RECOMMENDATION:                                              *
  * | fix pack | 4.1.1-ITNMIP-FP0001                             *
  ****************************************************************
  

Problem conclusion

• The ticket was resolved by altering the restart_disco_process.pl
  script so that rather than waiting 20 seconds and assuming the
  process was by then dead it instead uses ps to monitor the old
  process and when it is sure it is dead it restarts the new
  process.
  

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:

Fix information

• Fixed component name

  TIV NETWK MGR I

• Fixed component ID

  5724S4500

Applicable component levels

• R411 PSN

     UP

• R411 PSY

     UP