DB2 FOR LUW / DB2FORLUW / 980 - DB2

Diagnostic logging from the CF occurring immediately after a 
member has been ejected from the cluster may degrade the CFs 
ability to 
service requests from member. This may manifest as 
a short tansient period of degraded throughput from the cluster. 
 
 
The problem occurs because the CF worker threads get stalled 
writing the diagnostic messages. The stall is most severe when 
they are writing diagnostic messages to a GPFS filesystem which 
is undergoing recovery. The problem can be further complicated 
in  environments with a low number of CF worker threads or when 
the IO system is constrained. 
 
These messages may look like: 
 
The affected messages are similar to the following: 
2011-10-13-21.20.07.0391667000+540 E123456789A299    LEVEL    : 
Error 
PID      : 14811370 TID :      1800 
HOSTNAME  : c5cf01 
FUNCTION  : CA trace, log_error 
MESSAGE  : CA server has encountered an error. 
DATA #1  : 
DMA Error detected in File (cshwarm.c) at line (2443), Reason 
code: (800b0020) 
 
and 
 
2011-10-13-21.20.13.0815893000+540 E123456789A369    LEVEL    : 
Error 
PID      : 14811370 TID :      3599 
HOSTNAME  : c5cf01 
FUNCTION  : CA trace, log_error 
MESSAGE  : CA server has encountered an error. 
DATA #1  : 
do_xi_list - delayed-ack() RDMA Write of XI byte 
(dat_ep_post_rdma_write) failed for index 2 of XI list (3 
entries). LCID: 2. DAT Status: 0x80030002 
 
 
The fix provided by this APAR removes these messages from being 
written to disk. 
The solution is to filter specified msgids in the trace log

**************************************************************** 
* USERS AFFECTED:                                              * 
* Users of DB2 V9.8                                            * 
**************************************************************** 
* PROBLEM DESCRIPTION:                                         * 
* See Error Description.                                       * 
**************************************************************** 
* RECOMMENDATION:                                              * 
* Upgrade to DB2 9.8 Fix Pack 5                                * 
****************************************************************

Problem was first fixed in Version 9.8 Fix Pack 5

keiner bekannt / siehe Local-Fix

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

01.11.2011
21.06.2012
21.06.2012

9.8.FP5