I've setup a cluster of two WAS 5.0 servers as described in this
document:
http://www-106.ibm.com/developerworks/linux/library/l-ss4ip-was5clus/
Everything seems to be fine (the cluster starts and I can see the
information exchange between the server about updating the configuration
and everything) until the moment I try to start the Default Application.
If anyone has a suggestion on what I'm doing wrong it would be greatly
appriciated if you provide help. Below is the beginning of the error
message I get. (I can copy in the whole message if someone asks, but it
goes on for several pages).
Many thanks in advance,
Louis
[3/1/04 10:26:29:347 CET] 43cf3839 ApplicationMg A WSVR0200I: Starting
application: DefaultApplication.ear
[3/1/04 10:26:29:393 CET] 43cf3839 EJBContainerI I WSVR0207I: Preparing
to start EJB jar: Increment.jar
[3/1/04 10:26:29:597 CET] 43cf3839 EJBContainerI I WSVR0037I: Starting
EJB jar: Increment.jar
[3/1/04 10:26:29:618 CET] 43cf3839 Helpers W NMSV0605W: A
Reference object looked up from the context "java:" with the name
"comp/PM/WebSphereCMPConnectionFactory" was sent to the JNDI Naming
Manager and an exception resulted. Reference data follows:
Reference Factory Class Name:
com.ibm.ws.naming.util.IndirectJndiLookupObjectFactory
Reference Factory Class Location URLs: <null>
Reference Class Name: java.lang.Object
Type: JndiLookupInfo
Content: JndiLookupInfo: jndiName="eis/DefaultDatasource_CMP";
providerURL=""; initialContextFactory=""
Exception data follows:
javax.naming.NameNotFoundException: eis/DefaultDatasource_CMP
at
com.ibm.ws.naming.jndicos.CNContextImpl.doLookup(CNContextImpl.java:1503)
at
com.ibm.ws.naming.jndicos.CNContextImpl.doLookup(CNContextImpl.java:1457)
at
com.ibm.ws.naming.jndicos.CNContextImpl.lookup(CNContextImpl.java:1167)
at com.ibm.ws.naming.util.WsnInitCtx.lookup(WsnInitCtx.java:132)
at javax.naming.InitialContext.lookup(InitialContext.java:359)
at
com.ibm.ws.naming.util.IndirectJndiLookupObjectFactory$1.run(IndirectJndiLookupObjectFactory.java:216)
at java.security.AccessController.doPrivileged(Native Method)
at
com.ibm.ws.naming.util.IndirectJndiLookupObjectFactory.getObjectInstance(IndirectJndiLookupObjectFactory.java:92)
at
javax.naming.spi.NamingManager.getObjectInstance(NamingManager.java:313)
at
com.ibm.ws.naming.util.Helpers.processSerializedObjectForLookup(Helpers.java:884)
at
com.ibm.ws.naming.urlbase.UrlContextHelper.processBoundObjectForLookup(UrlContextHelper.java:135)
at
com.ibm.ws.naming.urlbase.UrlContext.lookup(UrlContext.java:1227)
at
com.ibm.ws.naming.urlbase.UrlContext.lookup(UrlContext.java:1203)
at
com.ibm.ws.naming.urlbase.UrlContext.lookup(UrlContext.java:1257)
at javax.naming.InitialContext.lookup(InitialContext.java:359)
I don't think this has anything to do with the cluster - but the
datasource definition for this sample datasource was defined only at the
server level and not at the cluster level.
Ken
From what I can find that could very well be the problem. However,
being a newbie at clustering, this generates another question: "How do I
define the datasource at cluster level?"
Louis
on the datasource page, delete the server and/or node name, and press
apply. The arrow will then point to the cluster. Define your datasource
then.
Ken