IV30360: WHEN AN ENDPOINT IS UNAVAILABLE, ONE MORE DISTRIBUTION IS ATTEMPTED.

APAR status

• Closed as program error.

Error description

• When an endpoint status is UNAVILABLE or INTERRUPTED, one more
  retry distribution is made.
  <configuration>
  retry_ep_cutoff = 180
  conn_retry_interval = 60
  
  <current behavior :incorrect>
  1.start distribution
  2.Endpoint becomes unavailable
  (60sec elapsed)
  3.unavailable time(60) < retry_ep_cutoff(180)-->1st retry start
  4.Endpoint becomes unavailable
  (60sec elapsed)
  5.unavailable time(120) < retry_ep_cutoff(180)-->2nd retry start
  6.Endpoint becomes unavailable
  (60sec elapsed)
  7.unavailable time(180) = retry_ep_cutoff(180)-->3rd retry start
  ** this attempt should not be done **
  8.Endpoint becomes unavailable
  
  <correct behavior>
  1.start distribution
  2.Endpoint becomes unavailable
  (60sec elapsed)
  3.unavailable time(60) < retry_ep_cutoff(180)-->1st retry start
  4.Endpoint becomes unavailable
  (60sec elapsed)
  5.unavailable time(120) < retry_ep_cutoff(180)-->2nd retry start
  6.Endpoint becomes unavailable
  (60sec elapsed)
  

Local fix

Problem summary

• When an endpoint status is UNAVILABLE or INTERRUPTED, MDist2
  repeater retries the distribution after conn_retry_interval.
  When conn_retry_interval elapsed, if the total down time is
  equal to or greater than retry_ep_cutoff, MDist2 repeater should
  stop the retrial of the distribution.
  When conn_retry_interval elapsed, MDist2 compares the total down
  time with retry_ep_cutoff to determine whether or not to start
  the retrial of the distribution. But the total downtime is not
  updated just before it is compared. It is updated after MDist2
  starts the retial. So, the total down time becomes less than the
  correct value. That causes the Mdist2 to make one more retrial
  even though the correct down time exceeds retry_ep_cutoff.
  

Problem conclusion

• The gateway has been fixed. This fix will be included in the
  next Tivoli Management Framework 4.1.1 and 4.3.1 patch for
  TMR/MN/GW.
  

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

  MGMT. FRAMEWORK

• Fixed component ID

  5698FRA00

Applicable component levels

• R411 PSY

     UP

• R431 PSY

     UP