IZ02349: USING CLUSTERING INDEX, FOR SCENARIO THAT AN INSERT ROLLED BACK DUE TO LOCKTIMEOUT, DB2 MAY WAIT AS TWICE AS LOCKTIMEOUT VALUE

DB2 UDB Version 8.1 FixPak 16 (also known as Version 8.2 FixPak 9 )

APAR status

• Closed as program error.

Error description

• If the index is a clustering index, when we insert a row, there
  are the following steps involved:
  1. We will perform a scan on the clustering index and try to
  cluster the row using the clustering index.
  2. Once the row is inserted properly, we need to insert the key
  into index(es). When we insert the key into the clustering
  index, we will first try to avoid the index probe and try to
  insert the key directly onto the last index page we read during
  step 1.
  3. We need to lock the next key during the insert, and if the
  next key is already locked in a mode that's not compatible with
  the insert, then we will attempt to reprobe from the root again.
  If the row lock is still not released during the reprobe, the
  insert will fail with a lock time out error.
  
  For the above scenario, DB2 waits as twice as locktimeout value.
  

Local fix

• Not available
  

Problem summary

• Users affected: All
  
  Problem Description:
  Using clustering index, for scenario that an insert rolled back
  due to locktimeout, DB2 may wait as twice as locktimeout value.
  
  Problem Summary:
  As same as Error description.
  

Problem conclusion

• Problem was first fixed in Version 8.1 FixPak 16 (s080111).
  

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:

  IZ43462

Fix information

• Fixed component name

  DB2 UDB WSE SOL

• Fixed component ID

  5765F3502

Applicable component levels

• R810 PSN

     UP

• R820 PSN

     UP