PM74655: READ TIMEOUTS CAN CAUSE A CLASSCASTEXCEPTION WHEN USING IPC AS THE LOCAL CONNECTOR.

Fixes are available

APAR status

Closed as program error.

Error description

A request is sent from the DMGR to the local Node agent using
the IPC connector.
When the response is not received within the set timeout
(default 600 seconds) the READ will time out and the DMGR
closes the connection. The connection is put back into the pool
and eventually gets reused for new connections.
At some point the Node agent is finally sending the response to
the DMGR which is put on the inbound message queue of the
current connection, and the READ will result in a
ClassCastException.

Here is an example flow:
The DMGR sends a request  to the NA at 09:50:50.361. Let's call
this REQ1.
After sending this request the DMGR eventually timed out on a
READ request waiting for the response, as seen in this trace
entry:

Trace: 2012/10/05 10:00:50.366 01 t=8B0580 c=UNK key=S2
(13007002)
   ThreadId: 0000001b
   FunctionName:
com.ibm.ws.management.connector.ipc.ClientAccessor
   SourceId: com.ibm.ws.management.connector.ipc.ClientAccessor
   Category: FINEST
   ExtendedMessage: Exception while receiving or de-serializing
response; java.io.IOException: ZAioTCPChannel: sync read
operation timed out.
    at
com.ibm.ws.tcp.channel.impl.ZAioTCPReadRequestContextImpl.read(Z
AioTCPReadRequestContextImpl.java:227)
 ...... rest of stack trace omitted.


The connection from the DMGR to the NA is re-established at
10:00:50.381.   It drove a new request into the NA (REQ2) and
issued a synchronous READ to wait for the response (RSP2).
The original request finally finished and drove a response at
10:03:42.940 (RSP1).
The DMGR was expecting RSP2 on its inbound message queue, and
received RSP1 instead, causing the ClassCastException.
The DMGR is left with an additional response in its pipe and
the status of the node will not be reflected correctly after
that. This is because the isAlive request from the DMGR to the
Node agent will continue to fail after the exception.

If the DMGR and Node agent are restarted it will work for a
while.

Local fix

Use SOAP instead of IPC as the local connector

Problem summary

****************************************************************
* USERS AFFECTED:  All users of IBM WebSphere Application      *
*                  Server V7.0, V8.0, and V8.5                 *
****************************************************************
* PROBLEM DESCRIPTION: WebSphere Application Server for z/OS   *
*                      local communication received stale      *
*                      data                                    *
****************************************************************
* RECOMMENDATION:                                              *
****************************************************************
A timing window in the local communication code can lead to the
successful sending of data using a stale connection.  The
delivery of this data across a stale connection can cause
unexpected behavior on the receiving side of a new connection.
It has been observed, on the Admin IPC channel over local
communication, that this data was mistakenly treated as valid
reply data to new requests.  This in turn lead to a
ClassCastException after a number of erroneous "successful"
requests.

Problem conclusion

Additional code has been added in the local communication path
to detect and reject attempts to send data over a stale
connection.

APAR PM74655 is currently targeted for inclusion in Fix Packs
7.0.0.29, 8.0.0.6, and 8.5.0.2 of WebSphere Application Server.

Please refer to URL:
//www.ibm.com/support/docview.wss?rs=404&uid=swg27006970
for Fix Pack availability.

Temporary fix

Comments

APAR Information

APAR number
PM74655
Reported component name
WEBSPHERE FOR Z
Reported component ID
5655I3500
Reported release
700
Status
CLOSED PER
PE
NoPE
HIPER
NoHIPER
Special Attention
NoSpecatt
Submitted date
2012-10-09
Closed date
2012-12-11
Last modified date
2013-07-03

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
WEBSPHERE FOR Z
Fixed component ID
5655I3500

Applicable component levels

R700 PSY UK94926
UP13/06/20 P F306

Fix is available

Select the PTF appropriate for your component level. You will be required to sign in. Distribution on physical media is not available in all countries.

[{"Business Unit":{"code":"BU059","label":"IBM Software w\/o TPS"},"Product":{"code":"SS7K4U","label":"WebSphere Application Server for z\/OS"},"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"7.0","Line of Business":{"code":"LOB45","label":"Automation"}}]

Document Information

Modified date:
29 October 2021

Tips

PM74655: READ TIMEOUTS CAN CAUSE A CLASSCASTEXCEPTION WHEN USING IPC AS THE LOCAL CONNECTOR.

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

R700 PSY UK94926

Fix is available

Select the PTF appropriate for your component level. You will be required to sign in. Distribution on physical media is not available in all countries.

Document Information

Share your feedback

Need support?