Recently, I switched to the websphere liberty 8.5.5.9. The adminCenter replaced the admin console. I used to configure many things in the admin console of lower Websphere version. but in the new admincenter, I couldn't find them any more, e.g. configure a datasource. any suggestions?
The Admin Console is the UI for WebSphere Traditional whereas Admin Center is the UI for WebSphere Liberty. Just like there are differences between the functionality of WebSphere Traditional and Liberty, same can be said for Admin Console and Admin Center.
Although Admin Center does not provide a 'wizard' type of experience that Admin Console does, you can use the Server Config tool to create a datasource and it can help guide you through the process by showing all the fields that can be set (including any defaults). If you stay in the Design view for the configuration file (likely server.xml) of the server that you want to alter, you can select 'Add Child' and select 'Datasource.' If will then populate all the fields with descriptions and defaults:
Data Source in Design View of Server Config in Admin Center
You can get similar assistance in the 'Source' view by activating content assist via ctrl+space. This functionality is not unique to datasource, but is provided for all elements.
In WebSphere Liberty you can configure your DataSource directly in the server.xml (similar to most server-level configurations in Liberty).
Here is an example from the Liberty documentation:
<dataSource id="DefaultDataSource" jndiName="jdbc/derbyEmbedded">
<jdbcDriver libraryRef="DerbyLib"/>
<properties.derby.embedded databaseName="C:/databases/SAMPLEDB" createDatabase="create"/>
</dataSource>
<library id="DerbyLib">
<fileset dir="C:/db-derby-10.8.1.2-bin/lib"/>
</library>
To use a datasource, be sure to enable a JDBC feature in your server.xml, such as this:
<featureManager>
<feature>jdbc-4.1</feature>
</featureManager>
If you're not sure where to find your server.xml, by default it is in this location:
${wlp.install.dir}/servers/${your.server.name}/server.xml
For full doc on configuring data sources in Liberty, see: Configuring relational database connectivity in Liberty
In addition to editing the server.xml as described by aguibert, you can create and edit data sources using the Liberty AdminCenter as described in this IBM KnowledgeCenter topic.
Related
I was following a tutorial for setting up a WebSphere Liberty Server Here and didn't really know what a part of the tutorial did. I completed the tutorial and it works fine.
On step 3 it has me modify the server.xml with these two lines and I dont really know what they do.
<applicationMonitor updateTrigger="mbean" />
<feature>localConnector-1.0</feature>
I Found the documentation for localConnector-1.0 but its a little over my head
https://www.ibm.com/support/knowledgecenter/en/SSEQTP_liberty/com.ibm.websphere.liberty.autogen.nd.doc/ae/rwlp_feature_localConnector-1.0.html
I think localConnector allows IntelliJ to run the server somehow but i dont know what updateTrigger="mbean" does.
If anyone has an explanation that would be great. Thanks!
The localConnector-1.0 feature enables the local JMX connector on Liberty so that the JMX Client (IntelliJ) can connect to and administer Liberty.
You can find more documentation on the feature here: https://www.ibm.com/support/knowledgecenter/en/SSEQTP_liberty/com.ibm.websphere.wlp.doc/ae/twlp_admin_localconnector.html
updateTrigger="mbean" is setting the application updates to only occur when trigger by an mbean call (whereas the default is to poll for changes).
You can find more documentation here:
https://www.ibm.com/support/knowledgecenter/SSAW57_liberty/com.ibm.websphere.wlp.nd.multiplatform.doc/ae/twlp_setup_dyn_upd.html
I created a simple project using ATG 10.2 .I want to know how to deploy it in weblogic. Please provide detailed procedure with screenshots,if possible.
To provide a 'detailed' procedure is beyond the scope of what StackOverflow is trying to provide. That said, if you have an understanding of the Weblogic Management Console you should be able to follow these steps to setup your initial deployment:
Create a Server
1.1 Specify a server name (eg. commerce) and the port number this server will run on (eg. 8180). Select it as a 'Stand-alone server'.
1.2 Once created go to Configuration > Server Start for the newly created server and modify the 'Arguments' block and include the following setings (assuming you are running windows, for Unix update your own paths)
-Datg.dynamo.data-dir=c:\ATG-Data -Datg.dynamo.server.name=commerce -d64 -XX:ParallelGCThreads=8 -Dsun.rmi.dgc.client.gcInterval=3600000 -Dsun.rmi.dgc.server.gcInterval=3600000 -Xms1152m -Xmx2048m -XX:NewSize=128m -XX:MaxNewSize=256m -XX:PermSize=128m -XX:MaxPermSize=256m
1.3 Save your Server
Create DataSources
2.1 In the Console click on 'Services > Data Sources'
2.2 Create 'New' datasources for each of your connections. As a minimum you will need connections for ATGSwitchingDS_A, ATGSwitchingDS_B (Assuming you are doing switching datasources) and ATGProductionDS. These names should match your JNDI names in your property files. Remember to specify the 'commerce' server as the target for each of the datasources.
Create Deployment
3.1 Assuming you've already built your EAR (eg. ATGProduction.ear) and it is available in c:\deployments you need to create a deployment in Weblogic. You need to create the deployment in the console and specify the target as 'commerce'. Once done you need to also 'start serving requests' on the deployment.
Start Server
You should now be able to see your server running on port 8180 with the log files being written to c:\ATG-Data\servers\commerce\logs.
If after this things aren't running, post specific questions about your issues and someone here might be able to help you.
When configuring WL HTTP Adapters, the domain and port are part of the adapter configuration .xml file build and uploaded on the WL server.
For our use case (especially beta testing and demos) the endpoint server url needs to be configurable for the end user. Example, same builds are tested by QA on test envs, while BA connects to demo.
We have only one WL Server up and setting environment specific servers is not an option.
Is it possible to change domain/hostname dynamically at application launch or runtime ? Ideally it would be to get and use the domain/hostname value from a drop down or free input from the client and use it.
<connectivity>
<connectionPolicy xsi:type="http:HTTPConnectionPolicyType">
<protocol>http</protocol>
<domain>{hostname}</domain>
<port>80</port>
</connectionPolicy>
<loadConstraints maxConcurrentConnectionsPerNode="2" />
</connectivity>
Update: This answer is useful, so I leave it here for reference, but accept that it doesn't correctly answer this question!
There is a specific Worklight feature designed to address your scenario (for the Infocenter detail, see here).
You can do this by using a combination of worklight.properties and JNDI properties.
For example, let's say you had this setup in your adapter XML:
<connectivity>
<connectionPolicy xsi:type="http:HTTPConnectionPolicyType">
<protocol>${my.adapter.protocol}</protocol>
<domain>${my.adapter.domain}</domain>
<port>${my.adapter.port}</port>
</connectionPolicy>
<loadConstraints maxConcurrentConnectionsPerNode="2" />
</connectivity>
You then define default values for these in your worklight.properties file (in the server/conf directory of your Worklight project, and "burnt in" to the .WAR file when you build it):
my.adapter.protocol=http
my.adapter.domain=some.host.com
my.adapter.port=80
You can then override these values in individual environments, by setting JNDI properties. For example, if you are using WebSphere Liberty, you might put this in your server.xml:
<jndiEntry jndiName="my.adapter.protocol" value="https"/>
<jndiEntry jndiName="my.adapter.domain" value="some.other.host.com"/>
<jndiEntry jndiName="my.adapter.port" value="8080"/>
You could create 3 adapters: 2 adapters connect to each backend servers, and one "proxy" adapter.
Your application would call the proxy adapter, passing some variable (the dropdown). Then the proxy would call one of the 2 real adapters.
You can´t. An option is to use MashUps.
See:
IBM Worklight 6.1 - Can a HTTP adapter call another HTTP adapter on server-side?
http://public.dhe.ibm.com/software/mobile-solutions/worklight/docs/v620/04_11_Advanced_adapter_usage_and_mashup.pdf
The idea is to have an Adapter responding to app mobile than this adapter call other deployed adapter (QA, Test Env, Prod Env, etc) considering any logic.
I think if you override WL.Server.invokeHttp so you can specify the domain name in each request instead of using the one set in the XML file, would a good alternative . You can get the domain name as parameter in the invocation request.
There is question that may interest you regarding that, you should check it because I'm not sure if it's possible or not. Worklight Adapter Override Origin of request
I want to know how we can configure Liberty Profile 8.5.5 (Dev version not the WAS ND version) to be load balanced by apache http server.
I have tried to search but havent been able to come across any useful links. Any help will be much appreciated.
Thanks,
Vishalendu
Currently, you'll have to generate a plugin-cfg.xml from each liberty server (the license has info about how many servers you can aggregate in this way for load balancing and failover) and merge the result to make it appear like a cluster to the WAS Plugin.
Other editions provide a merge tool, if you have access to them.
The WAS plugin installation has an XSD file for the plugin-cfg.xml.
1) note the http and https transports in both plugin configurations
2) make a copy of one of the XML's to edit
3) Find the <ServerCluster
<Config...
<ServerCluster CloneSeparatorChange="false" GetDWLMTable="false" IgnoreAffinityRequests="true" LoadBalance="Round Robin" Name="cluster1" PostBufferSize="64" PostSizeLimit="-1" RemoveSpecialHeaders="true" RetryInterval="60" ServerIOTimeoutRetry="-1">
<!-- copy generated Server stanza for your other XML -->
<Server ...
<PrimaryServers>
<!-- add a 2nd primary server, from your other XML -->
<Server Name="node1_serv1"/>
...
</PrimaryServers>
</ServerCluster>
4) Copy the stanzas from the other file inside the ServerCluster
5) Add the servers name to the field
If your servers have the same apps on them, you're done. Otherwise, you have to merge the other elements (Route, URIGroup, etc) but usually they'll be the same.
Worklight 5.0.5
Liberty 8.5.0.1
We are experiencing some timeout problems, i see in the liberty infocenter it says that you can turn on access logging on liberty with
<httpAccessLogging/>
and with
<httpAccessLogging enabled="true" filePath="e:\\Temp\\http_access.log" />
I have inserted the tag in my server.xml, but it does not create an access log file.
The main thing im looking for the is the IP address of the device that is making the call.
That was i can find out if the call made it to the server at a certain time..
httpAccessLogging is enabled on a per-httpEndpoint basis. Try:
<httpEndpoint id="defaultHttpEndpoint">
<accessLogging/>
</httpEndpoint>
...or:
<httpEndpoint id="defaultHttpEndpoint" accessLoggingRef="hal"/>
<httpAccessLogging id="hal"/>