PQ60039: OSE REMOTE PLUGIN "DEAD CLONE" DETECT PROBLEM.

Fixes are available
PQ71589: Incorrect response code is logged in IIS access log
PQ67072; 3.5.4: Invalid error number reported in plug-in traces
PQ67072; 3.5.4: Invalid error number reported in plug-in traces
PQ60039: Connect timeout enablement for OSE remote with WebSphere 3.5.x
PQ61926, 3.5.x: Connect timeout enablement for OSE remote
WebSphere Application Server Version 3.5 Fix Pack 7 (3.5.7)
PQ60991, 3.5.4, 3.5.6: Connection handshake enablement for OSE remote
PQ57425: Timeout enablement on the Apache/IHS webserver with plug-ins.
PQ76785: Cumulative plug-in fix for WebSphere Application Server V 3.5.4-3.5.7

APAR

APAR status
Closed as program error.

Error description
In AIX IHS WAS "OSE Remote plugin" does not have capability to
detect connect timeout to remote queue port.
tcp_keepinit system default is 75 seconds. so we need at least
75 seconds to detect dead clone by default.
Since tcp_keepinit is system wide parameter, We can not set
short value for this.
Another problem is, AIX IHS is running in multi process mode.
and each process have clone status independently. therefore each
process need to wait at least 75 seconds to detect dead clone.
All process must share single clone status information.
Local fix
Problem summary
****************************************************************
* USERS AFFECTED: All WebSphere Application Server users       *
*                 of OSE                                       *
****************************************************************
* PROBLEM DESCRIPTION: Failover delayed when an application    *
*                      server machine removed from network.    *
****************************************************************
* RECOMMENDATION:                                              *
****************************************************************
When a application server machine is removed from the network
the webserver machine has to wait the system tcp/ip timeout
period before it realizes that the appserver machine is
unavailable.  This delays failover.
Problem conclusion
A directive was added to allow the user to specifiy how
long the plugin should wait before timing out when trying to
communicate with a machine not connected to the network.
User can set ose.connect.timeout=N (where N is the time in
seconds) in the bootstrap.properties file.  After N seconds
the plugin will assume the machine is unavailable and
failover will occur.
Temporary fix
Comments
APAR information
APAR numberPQ60039
Reported component nameWAS ADVANCED AI
Reported component ID5648C8400
Reported release350
StatusCLOSED PER
PENoPE
HIPERNoHIPER
Submitted date2002-04-16
Closed date2002-05-13
Last modified date2002-08-29

APAR is sysrouted FROM one or more of the following:

APAR is sysrouted TO one or more of the following:APAR is sysrouted FROM one or more of the following:


Modules/Macros
PLUGIN
APAR is sysrouted TO one or more of the following:Modules/Macros

Fix information
Fixed component nameWAS ADVANCED AI
Fixed component ID5648C8400

Applicable component levels
R350 PSYUP











Document Information

Product categories: Software, Application Servers, Distributed Application & Web Servers, WebSphere Application Server, General
Software version: 350
Reference #: PQ60039
IBM Group: Software Group
Modified date: 2002-08-29