IV35487: POLLER INCORRECTLY STOPPING THE MONITORS WHILE PROCESSING THE TOPOLOGY UPDATES UPON COMPLETION OF DISCOVERY.

IBM Tivoli Network Manager IP 3.9.0 Fix Pack 3, 3.9.0-TIV-ITNMIP-FP0003

APAR status

• Closed as program error.

Error description

• v3.9.0.68 (v3.9.0.71 -GA refresh build) - platform independent.
  
  Upon completion of discovery, poller incorrectly stops monitors
  resulting in no polling on many devices.
  
  Replication steps:
  1) Do a full discovery and enable few policies, in this example
  we are tracking 'SNMP Link State' policy.
  
  2) Take the dump of 'select * from monitors.current;' from
  service SnmpPoller using ncp_oql.
  
  After very first discovery - poller has following entity from
  above query output:
  
  {
          MAINNODEENTITYID=6620600;
          MAINNODEADDRESS='101.1.1.1';
          MAINNODEENTITYNAME='THIRD_PARTY_NETWORK';
          ENTITYCLASS='NetcoolMPLS';
          LINGERTIME=3;
          CHASSISMANAGED=NULL;
          ENTITYID=6620600;
          ENTITYNAME='THIRD_PARTY_NETWORK';
          ENTITYTYPE=1;
          ACCESSIPADDRESS='101.1.1.1';
          MONITOREDINSTID=1;
          ISMANAGED='true';
          LASTPOLLTIME='2013-01-17 09:59:17';
          LASTPOLLINTERVAL=120;
          LASTPOLLFAILURE='2013-01-17 09:59:17';
          TEMPLATEID=4;
          TEMPLATENAME='SNMP Link State';
          POLICYID=4;
          POLICYNAME='SNMP Link State';
          CURRENTTIME='2013-01-17 09:59:26';
  }
  
  
  3) Do another full discovery and wait for Poller sync with
  latest topology - at this time, take dump of another set of
  'select * from monitors.current;' query:
  
  The results are:
  
  {
          MAINNODEENTITYID=6620600;
          MAINNODEADDRESS='101.1.1.1';
          MAINNODEENTITYNAME='THIRD_PARTY_NETWORK';
          ENTITYCLASS='NetcoolMPLS';
          LINGERTIME=3;
          CHASSISMANAGED=NULL;
          ENTITYID=6620600;
          ENTITYNAME='THIRD_PARTY_NETWORK';
          ENTITYTYPE=1;
          ACCESSIPADDRESS='101.1.1.1';
          MONITOREDINSTID=1;
          ISMANAGED='true';
          LASTPOLLTIME='1969-12-31 19:00:00';  <==not polling
          LASTPOLLINTERVAL=120;
          LASTPOLLFAILURE='2013-01-17 10:07:17';
          TEMPLATEID=4;
          TEMPLATENAME='SNMP Link State';
          POLICYID=4;
          POLICYNAME='SNMP Link State';
          CURRENTTIME='2013-01-17 10:47:32';
  }
  
  
  As you can see above - 'LASTPOLLTIME='1969-12-31 19:00:00';' -
  which means, Poller stopped polling this device.
  

Local fix

• Restart Poller upon completion of discovery.
  

Problem summary

• ****************************************************************
  * USERS AFFECTED:                                              *
  * All ITNM 3.9 Users.                                          *
  ****************************************************************
  * PROBLEM DESCRIPTION:                                         *
  * Upon completion of discovery, poller incorrectly stops       *
  * monitors                                                     *
  * resulting in no polling on many devices.                     *
  ****************************************************************
  * RECOMMENDATION:                                              *
  * | fix pack | 3.9.0-ITNMIP-FP0003                             *
  ****************************************************************
  

Problem conclusion

• When notification is for interface we only process for polls
  that have interface filters.  When we do process
  we remove the entity from the monitor and if this renders the
  monitor empty we remove the monitor.
  

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

  NC/PRECISIONIP

• Fixed component ID

  5724O52RC

Applicable component levels

• R390 PSN

     UP

• R390 PSY

     UP