Resolved Problems in the WebLogic SIP Server 2.1

CR222494

The SIP Servlet container did not support ejb-link and resource-ref entries defined in the sip.xml deployment descriptor file. Instead the values had to be defined in weblogic.xml as a workaround. The code was modified to support these entries directly in sip.xml.

CR235377

Call overload controls were not enabled by default. This problem was address with a code fix.

CR236024

WebLogic SIP Server sometimes threw a NullPointerException when running a User Agent Client (UAC) against a proxy servlet that proxied back to the same engine tier server instance. The problem caused the exception:

<Client timer task failed with fatal status java.lang.NullPointerException at com.bea.wcp.sip.engine.server.SipServletMessageImpl.getDialogId(SipServletMessageImpl.java:274)

The problem was solved with a code fix.

CR236379

Configuration files used inconsistent naming conventions for the data tier and replicas within the data tier. The configuration file schema has changed to consistently use the term "data tier" to refer to the cluster of WebLogic SIP Server instances that manage call state data, "partition" to refer to a managed portion of the call state, and "replica" to refer to an individual WebLogic SIP Server instance within a partition. See Configuring Data Tier Partitions and Replicas and Data Tier Configuration Reference (datatier.xml).

CR236479

The SNMP MIB for WebLogic SIP Server was previously available only from Managed Servers running in a domain. The code was modified to make WebLogic SIP Server MIB entries available from the Administration Server as well as Managed Servers. See Configuring SNMP.

CR237487

WebLogic SIP Server did not listen for UDP messages on a non-default network channel that specified IP_ANY/0.0.0.0 as the listen address. The code was modified so that the server listens for incoming UDP messages on any IP interface when you define a network channel with 0.0.0.0 as the listen address. See Configuring Servers to Listen on Any IP Interface (0.0.0.0) in Configuring and Managing WebLogic SIP Server.

CR238527

When using the Hostpot 1.4.2_05 VM and running under heavy loads, the UDP NIO socket would sometimes fail with:

java.io.IOException: Interrupted system call

at sun.nio.ch.PollArrayWrapper.poll0(Native Method)

at sun.nio.ch.PollArrayWrapper.poll(PollArrayWrapper.java:100)

at sun.nio.ch.PollSelectorImpl.doSelect(PollSelectorImpl.java:64)

at sun.nio.ch.SelectorImpl.lockAndDoSelect(SelectorImpl.java:59)

at sun.nio.ch.SelectorImpl.select(SelectorImpl.java:70)

at

com.bea.wcp.sip.engine.connector.transport.UdpTransportModule.run(UdpTransportModule.java:413)

The recovery from this failure caused the failure to lose 10 seconds of network traffic. This problem was resolved with a code fix.

CR239030

The previously deprecated XML configuration elements for defining trusted hosts have been replaced with new configuration elements. See sip-security in Configuring and Managing WebLogic SIP Server.

CR239250

In a replicated environment, or in a single server environment with debugging turned on, adding sleep time at the end of a doMessage() call could result in the error:

<Error> <WLSS.Session> <BEA-331410> <Invalid CSeq header. request=NOTIFY

sip:8005551212@172.17.24.251;appsessionid=app-nnw8zlr1voya:840a17c2649c84d0a47f06f1d7062cd2%40172.17.24.251;pxxx=12341234,

CSeq header=1 NOTIFY, CSeq number in this dialog=1>

This problem was resolved with a code fix.

CR240087

When waiting for over 60 minutes between an INVITE and a BYE message, a load testing proxy application would sent a 481 response even though the call should not be stateful. For example:

2005-08-22 14:12:51: Aborting call on unexpected message for Call-ID

'1-8415@10.32.4.213': while expecting '200' response, received 'SIP/2.0 481

Call/Transaction Does Not Exist

To: testuser <sip:proxy@10.32.4.213:5060>;tag=1

Content-Length: 0

CSeq: 2 BYE

Call-ID: 1-8415@10.32.4.213

Via: SIP/2.0/UDP 10.32.4.213:5061;branch=z9hG4bK-1-6

From: sipp <sip:sipp@10.32.4.213:5061>;tag=1

Server: BEA WebLogic SIP Server 2.1.0.0

This problem was resolved by adding a new container configuration parameter, default-behavior, which defines wether WebLogic SIP Server should act as a proxy or a user agent (UA) in the absence of an available, matching application. See default-behavior in Configuring and Managing WebLogic SIP Server.

CR240670

Prior to version 2.1, a WebLogic SIP Server 2.1 engine tier server would start up even if no SIP network channels were targeted to the server (for example, if a new engine tier server was configured manually and no channels were created).

The code was changed so that engine tier servers now throw an exception and fail to start if no SIP channels have been configured for the server. The new error message is:

<Error> <WLSS.Engine> <BEA-330075> <There are no sip channels targeted to server "servername">

CR241600

The previous version of the findme example application did not work in a domain having multiple engine tier servers in a cluster. The example code and documentation were modified to support a clustered environment. See Build the Example.