PM42959: WLM EXTERNAL CLIENT DATA PROPAGATION INCOMPLETE

Fixes are available

APAR status

Closed as program error.

Error description

In the scenario in which a customer has an external client to
the target cluster (defined as either a java thin client, or a
WebSphere Application Server process acting as a client which
is either in a separate, non-bridged cell, or in the same cell
but in a separate, non-bridged  coregroup), the cluster data
passed back to the client after certain events may be
incomplete, or new data may not be passed back at all.  An
example of this is shutting down a cluster member.  The
information that the cluster member was shut down may not be
passed back to the thin client.  In most scenarios this will
not cause more than minor issues, as the client side workload
management (WLM) code will be able to detect the server is
unavailable via an exception attempting to route to that
member, but this will not cause the client to permanately
remove that downed cluster member from the routing list.
Occasionally the client will attempt to reconnect to the
downed member to see if it has become available, in an
interval controlled by the jvm property described in this
article:

http://www-01.ibm.com/support/docview.wss?uid=swg21380854

Each attempt to the downed member would take a small amount of
time based on the client connection and socket timeouts.  If the
server has been shut down, the return will usually be quick,
one second or less. If the server is unavailable for another
reason, for example a Solaris Zone shutdown, this path
required a socket-connection timeout, which the default is
generally much longer.

Overall the problem experienced by a customer in this scenario
would be a reduced overall throughput due to the additional
attempts to the downed server.  In environments using the
default WLM timeout and client connection parameters, this
reduction in throughput would be slight, if even measurable,
but there are environments in which this can cause a
significant degredation in throughput and response times.

Local fix

Problem summary

****************************************************************
* USERS AFFECTED:  Users of java thin client applications and  *
*                  IBM WebSphere Application Server Network    *
*                  Deployment edition with clustering          *
****************************************************************
* PROBLEM DESCRIPTION: Workload management (WLM) data          *
*                      propagated to thin clients is           *
*                      incomplete                              *
****************************************************************
* RECOMMENDATION:                                              *
****************************************************************
In the scenario in which a customer has an external client to
the target cluster (defined as either a java thin client, or a
WebSphere Application Server process acting as a client which
is either in a separate, non-bridged cell, or in the same cell
but in a separate, non-bridged  coregroup), the cluster data
passed back to the client after certain events may be
incomplete, or new data may not be passed back at all.  An
example of this is shutting down a cluster member.  The
information that the cluster member was shut down may not be
passed back to the thin client.  In most scenarios this will
not cause more than minor issues, as the client side WLM code
will be able to detect the server is unavailable via an
exception attempting to route to that member, but this will
not cause the client to permanately remove that downed cluster
member from the routing list. Occasionally the client will
attempt to reconnect to the downed member to see if it has
become available, in an interval controlled by the  jvm
property described in this article:
http://www-01.ibm.com/support/docview.wss?uid=swg21380854
Each attempt to the downed member would take a small amount of
time based on the client connection and socket timeouts.  If
the
server has been shut down, the return will usually be quick,
one
second or less.
If the server is unavailable for another reason, for example a
Solaris Zone shutdown, this path required a socket-connection
timeout, which the default is generally much longer.
Overall the problem experienced by a customer in this scenario
would be a reduced overall throughput due to the additional
attempts to the downed server.  In environments using the
default WLM timeout and client connection parameters, this
reduction in throughput would be slight, if even measurable,
but there are environments in which this can cause a
significant
degradation in throughput and response times.

Problem conclusion

The code was modified to ensure in the start and stop scenario
that the data will be updated and propagated to the java thin
clients properly to allow for more correct routing behaviors.

In order to enable this function, you must set a cell-level
custom property by going through the administrative console
from:

System Administration -> Cell -> Custom Properties

and define a new custom property with the name:

IBM_CLUSTER_ENABLE_THIN_CLIENT_UPDATES

and a value of true.  Save, and synchronize to the nodes and
the next time each of the cluster members are restarted, the
custom property will be picked up and the change in behavior
come into effect.

The fix for this APAR is currently targeted for inclusion in
fix packs 6.1.0.43, 7.0.0.23, 8.0.0.3.  Please refer to the
Recommended Updates page for delivery information:
http://www.ibm.com/support/docview.wss?rs=180&uid=swg27004980

Temporary fix

Comments

APAR Information

APAR number
PM42959
Reported component name
WEBS APP SERV N
Reported component ID
5724H8800
Reported release
610
Status
CLOSED PER
PE
NoPE
HIPER
NoHIPER
Special Attention
NoSpecatt
Submitted date
2011-07-04
Closed date
2011-10-31
Last modified date
2011-10-31

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

Fix information

Fixed component name
WEBS APP SERV N
Fixed component ID
5724H8800

Applicable component levels

R61A PSY
UP
R61H PSY
UP
R61I PSY
UP
R61P PSY
UP
R61S PSY
UP
R61W PSY
UP
R61Z PSY
UP
R700 PSY
UP
R800 PSY
UP

[{"Business Unit":{"code":"BU059","label":"IBM Software w\/o TPS"},"Product":{"code":"SSEQTP","label":"WebSphere Application Server"},"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"610","Line of Business":{"code":"LOB45","label":"Automation"}}]

Document Information

Modified date:
27 October 2021

Tips

PM42959: WLM EXTERNAL CLIENT DATA PROPAGATION INCOMPLETE

Fixes are available

Subscribe

APAR status

Closed as program error.

Error description

Local fix

Problem summary

Problem conclusion

Temporary fix

Comments

APAR Information

APAR number

Reported component name

Reported component ID

Reported release

Status

PE

HIPER

Special Attention

Submitted date

Closed date

Last modified date

APAR is sysrouted FROM one or more of the following:

APAR is sysrouted TO one or more of the following:

Fix information

Fixed component name

Fixed component ID

Applicable component levels

R61A PSY

R61H PSY

R61I PSY

R61P PSY

R61S PSY

R61W PSY

R61Z PSY

R700 PSY

R800 PSY

Document Information

Share your feedback

Need support?