IZ99025: ITM 6.2.2 FP1 TEPS WON'T START AFTER REBOOT

IBM Tivoli Monitoring 6.2.2 Fix Pack 8 (6.2.2-TIV-ITM-FP0008)
IBM Tivoli Monitoring 6.2.2 Fix Pack 9 (6.2.2-TIV-ITM-FP0009)
IBM Tivoli Monitoring 6.2.2 Fix Pack 6 (6.2.2-TIV-ITM-FP0006)

APAR status

• Closed as program error.

Error description

• Severity:         2
  
  Approver:        sm
  Reported Release:622
  Compid:          5724C04PS Tivoli Enterprise Portal Server
  Abstract:        ITM 6.2.2 FP1 TEPS won't start after reboot
  
  Environment:
  Problem Description:
    The node status record for the managed system smse4111:KUL has
  a bad
  value for the HOSTINFO column.  It contains 'smse4111~6.1', but
  it
  should contain 'AIX~6.1'.  The TEPS uses the value of the
  HOSTINFO
  column to determine what system group folder to put a box level
  node
  under.  It maps the values to the "Windows Systems", "UNIX
  Systems",
  etc. folders you see under the Enterprise node.  For this node,
  it could
  not determine what system group to put it under, so it created a
  new
  system group named 'smse4111'.  This caused a recursive edge in
  the
  managed system graph, which led to an infinite loop.
  
  
  
  Detailed Recreation Procedure:
    recreated using TMES tables and migrate-export of TEPS.
  
  Related Files and Output:
  /ecurep/pmr/6/3/63994,075,724/2011-04-11 has the migrate-export
  
  /ecurep/pmr/6/3/63994,075,724/2011-04-08 has pdcollect with tems
  tables.
  ************************ End of APAR Template
  *************************
  

Local fix

• tacmd clean ms then check tables for HOSTINFO information.
  Remove any information that is incorrect.
  

Problem summary

• During startup or shortly after startup completes, the portal
  server begins to consume high CPU for several minutes.
  Eventually, the process crashes.  There are no messages in the
  log indicating a problem.  A dump of the node status table from
  the hub shows one or more agents with bad data in the HOSTINFO
  column.  The first part of the value, which should correspond to
  the operating system, instead contains the hostname of the
  machine.  For example, instead of "OS~V#.#", it contains
  "hostname~#.#".
  

Problem conclusion

• Prevent insertion of recursive edge in managed system graph.  If
  the condition is encountered, the following message will appear
  in the portal server log:
  
  ERROR: invalid platform for agent '<hostname>' due to bad
  HOSTINFO value of '<hostinfo value>', setting platform to
  '<object id>@<hostname>'
  
  The agent with bad HOSTINFO data will appear under a new
  platform level node in the physical navigator.  One or more
  nodes for agents that are also running on that system may show
  up under the new platform node.
  
  
  
  The fix for this APAR is contained in the following maintenance
  packages:
  
    | fix pack | 6.2.2-TIV-ITM-FP0006
  

Temporary fix

• Identify the agent(s) that contain bad data in the HOSTINFO
  column.  Stop those agents, then use the ITM CLI command
  'cleanms' to delete the node status entries for these agents.
  Then restart the agents and verify the HOSTINFO column contains
  valid data.
  

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

  TEPS

• Fixed component ID

  5724C04PS

Applicable component levels

• R622 PSY

     UP