DB2 FOR LUW / DB2FORLUW / 970 - DB2

In a federated DB2 database server environment, if multiple 
users or clients are collecting remote application snapshots 
concurrently, it may sometimes cause a dead latch in the DB2 
server which may effectively lead to the instance hanging. 
 
When the instance is hung, the stacks and the latches of the 
two agents may look like below: 
 
1) the agent 1 is holding the master application latch - 
SQLO_LT_sqeApplication__masterAppLatch, but waiting on the 
remote latch - SQLO_LT_sqlmon_dbcb__xlatch_remote_list. 
 
thread_waitlock 
sqloXlatchConflict 
sqm_snap_appl_remote 
sqlmonssagnt 
sqlmonssbackend 
sqlesrvr 
sqleMappingFnServer 
sqlerKnownProcedure 
sqlerCallDL 
 
2) While the agent 2 is holding the remote latch - 
SQLO_LT_sqlmon_dbcb__xlatch_remote_list, but is waiting on the 
master application latch - 
SQLO_LT_sqeApplication__masterAppLatch. 
 
thread_waitlock 
sqloXlatchConflict 
estim_appl_remote_snap_sz 
sqlmonszagnt 
sqlmonszbackend 
sqlesrvr 
sqleMappingFnServer 
sqlerKnownProcedure 
sqlerCallDL 
 
 
Thus there is a dead latch situation between these two agents. 
As these agents may also be holding other latches, such as the 
database latch, the application service latch, etc., this may 
eventually cause the instance to appears as hung, that is, 
terminating an existing connection or new database connections 
may get hung on these latches. The only way to get out of this 
hang is to recycle the db2 instance.

**************************************************************** 
* USERS AFFECTED:                                              * 
* Users in a federated DB2 database server environment where   * 
* multiple users or clients are collecting remote application  * 
* snapshots concurrently.                                      * 
**************************************************************** 
* PROBLEM DESCRIPTION:                                         * 
* See Error Description                                        * 
**************************************************************** 
* RECOMMENDATION:                                              * 
* Upgrade to DB2 Version 9.7 Fix Pack 6 or newer               * 
****************************************************************

1. Set database manager configuration variable FEDERATED to NO. 
Please note that this will disable the federation setup and will 
not allow connection to data sources through the federation 
server. 
 
2. Recycle the instance to resolve the hang and avoid collection 
of remote application snapshots as much as possible. Please note 
that some tools such as db2top may collect such snapshots 
undercover, avoid to run these tools if possible.

DB2 Version 9.7 Fix Pack 6 for Linux, UNIX, and Windows
 DB2 Version 9.7 Fix Pack 7 for Linux, UNIX, and Windows
 DB2 Version 9.7 Fix Pack 8 for Linux, UNIX, and Windows
 DB2 Version 9.7 Fix Pack 9 for Linux, UNIX, and Windows
 DB2 Version 9.7 Fix Pack 9a for Linux, UNIX, and Windows
 DB2 Version 9.7 Fix Pack 10 for Linux, UNIX, and Windows
 

Problem was first fixed in Version 9.7 Fix Pack 6

See above Local Fix

forerunner  : APAR is sysrouted TO one or more of the following: IC81192 IC84305 
follow-up : 

Date  - problem reported    :
Date  - problem closed      :
Date  - last modified       :

09.01.2012
18.06.2012
27.05.2013

9.7.FP6