DB2 FOR LUW / DB2FORLUW / A10 - DB2

This problem can lead to very huge leaks but needs the following 
to be true: 
 
  - Statement must use 'input array' (SQL_ATTR_PARAMSET_SIZE in 
CLI) 
  - Statement must use 'non atomic' type 
(SQL_ATTR_PARAMOPT_ATOMIC in CLI) 
 
With those attributes any time one row in the array fails to 
insert/update/delete a new 
'sqlca' is allocated for that row. At the end of the 'execute' 
the 'sqlca' should be freed. 
 
This is true EXCEPT if at some point while we are dealing with 
rows in the array 
one of them fails with a 'fatal' error like for example running 
out of log space. 
In that case we internally perform a 'rollback' and the 'sqlca' 
allocated for each rows 
will be left not freed until the connection is closed. 
 
A 'db2pd -db <dbname> -memb -appl' will should blocks in 
'1582738789' 
each one the size of a 'sqlca' structure. 
The following is sample output for db2pd -memblocks -appl. 
Address            DataAddress        PoolID     PoolName 
BlkAge Size(Bytes)  I LOC   File 
0x000007FAAF5A8EE8 0x000007FAAF5A8F00 1          apph 
28583 136          1 4150  1582738789 
...

**************************************************************** 
* USERS AFFECTED:                                              * 
* All platforms                                                * 
**************************************************************** 
* PROBLEM DESCRIPTION:                                         * 
* See Error Description                                        * 
**************************************************************** 
* RECOMMENDATION:                                              * 
* Upgrade to DB2 Version V10.1 Fixpack 5 or higher.            * 
****************************************************************

Terminate connection or restart db2 instance

Fixed in DB2 Version V10.1 Fixpack 5

keiner bekannt / siehe Local-Fix

Datum - Problem gemeldet    :
Datum - Problem geschlossen :
Datum - der letzten Änderung:

05.06.2014
13.07.2015
13.07.2015