How ATG Nucleus layers modules on start up - atg

I want to know how ATG modules get started & are layered by Nucleus on server startup without explicitly specifying -Datg.dynamo.modules on server start up. I have observed that it picks up the modules in the order in which the build was done. But from where it is picking up that dynamo_module list? Till now I have found out "startNucleus.bat" where I saw dynamo_modules property.
:startNucleus
%JAVA_VM% -classpath %CLASSPATH% -Datg.dynamo.home=%NUCLEUS_HOME% -Datg.dynamo.root=%NUCLEUS_ROOT% -Datg.dynamo.server.home=%NUCLEUS_HOME% %JAVA_ARGS% atg.applauncher.dynamo.DynamoServerLauncher %DYNAMO_MODULES% %CONFIGPATH%
With Regards,
Nutan

When you assemble an application, nucleus records the order in the which modules were assembled. This information is stored in a file within the EAR located # \atg_bootstrap.war\WEB-INF\ATG-INF
Here there is a file named dynamo.env, which is used by nucleus on server start up. You can override this configuration using -Datg.dynamo.modules on server start up.
Regards,
Aazim

Related

sling run modes use?

what is the use of sling run modes property in sling.properties file?
I have a osgi felix bundle that installed onto aem admin bundle console through aem cq5 package manager.
**
configuration properties of one of the bundle service is not available
unless I put the following line in cq5/config/sling.properties file.
sling.run.modes=author,sandbox why is this so ? what is the importance
of sling.run.modes ?
**
Thank you,
Sri
Run modes allow you to tune your AEM instance for a specific purpose; for example author or publish, test, development, intranet or others.
Exampe: For dev: sling.run.modes=author,dev
use of run mode is, example - i have a config.author.prod and config.author.dev in crxd/e. Based on the run mode instance, the OSGI Bundle will pick the corresponding config.author.dev or prod configuration settings defined in nt:unstructured and start working.
Ref: https://docs.adobe.com/docs/en/cq/5-6-1/deploying/configure_runmodes.html
Ref: https://helpx.adobe.com/experience-manager/kb/RunModeDependentConfigAndInstall.html
Define a respository-based configuration for a single instance
There are two ways to configure CQ5.
Configure the Apache Felix Web Management Console
The configuration on the Apache Felix Web Management Console (http://:/system/console/configMgr) is always specific for the current instance.
You can find a description in the documentation: http://dev.day.com/content/docs/v5_2/html-resources/cq5_guide_system_administrator/ch05s03.html
Repository-based configuration
It is also possible to store configuration in the CRX repository as nodes of nodetype sling:OsgiConfig.
For more information, see http://dev.day.com/content/docs/v5_2/html-resources/cq5_guide_system_administrator/ch05s02.html
With this method, it is possible to share configuration among several instances.
The name of these nodes must be equal to the Persistent Identity (PID) of the configuration (for example, the name of the service). If you look at http://:/system/console/config, you see these names listed as service.pid properties. These configuration nodes have to be child-nodes of nodetype nt:folder with a name starting with config followed with a dot. All the run-modes that the config applies to are also separated with a dot.
Examples: config.author, config.publish, config.author.dev, config.author.foo.dev, and so on.

Cant access to FTP using Eclipse

I am using the Remote System software on Eclipse. I can successfully log in to my FTP account but when I try to view the directories, I get the following message:
Message: Operation failed due to network I/O error
'java.net.SocketException: Connection reset by peer: socket write
error'
Any ideas are welcome.
Looks like there could be some negotiation issue.
Try following solution:
I've got the same exception and in my case the problem was in a
renegotiation procecess. In fact my client closed a connection when
the server tried to change a cipher suite. After digging it appears
that in the jdk 1.6 update 22 renegotiation process is disabled by
default. If your security constraints can effort this, try to enable
the unsecure renegotiation by setting the
sun.security.ssl.allowUnsafeRenegotiation system property to true.
http://www.oracle.com/technetwork/java/javase/overview/tlsreadme2-176330.html
Setting the System Properties/Mode Configuration The various modes are
set using the corresponding system properties, which must be set
before the SunJSSE library is initialized. There are several ways to
set these properties:
From the command line:
% java -Dsun.security.ssl.allowUnsafeRenegotiation=true Main Within
the application:
java.lang.System.setProperty("sun.security.ssl.allowUnsafeRenegotiation",
true); In the Java Deployment environment (Plug-In/Web Start), there
are several ways to set the system properties. (See Java Web App and
Next Generation Web Browser Plugin for more information.)
Use the Java Control Panel to set the Runtime Environment Property on
a local/per-VM basis. This creates a local deployment.properties file.
Deployers can also distribute a enterprise-wide deployment.properties
file by using the deployment.config mechanism. (See Deployment
Configuration File and Properties.)
To set a property for a specific applet, use the HTML subtag
"java_arguments" within the tag. (See Java Arguments.)
To set the property in a specific Java Web Start application or applet
using the new Plugin2 (6u10+), use the JNLP "property" sub-element of
the "resources" element. (See Resources Element.)

JBoss 6.0.0 blank page during war redeploy for 3 seconds

I have the following problem.
I am using Jboss 6.0.0. I need to daily update the application that runs 24/7 (only WAR file).
During redeploy the server responds with a blank page (0 bytes). It takes about 3 seconds.
13:25:39,001 INFO
[org.jboss.web.tomcat.service.deployers.TomcatDeployment] undeploy,
ctxPath=/ 13:25:42,021 INFO
[org.jboss.web.tomcat.service.deployers.TomcatDeployment] deploy,
ctxPath=/
This is obviously the most adverse scenario, as users, who at that moment send the form lose all of the inputed data.
I tried to find a solution setting various configuration options (autoDeploy, unloadDelay, reloadable context attributes) with no success.
Is there something I'm missing or this is normal behavior and the server does not distinguish between redeploy and separate undeploy and deploy operations. It would be ideal if it could distinguish them and if container during the redeploy would queue requests until the next instance of servlet is initialized.
Details of the environment:
JBoss 6.0.0 Final single instance (also tried on Tomcat 7.0.28 - the same effect)
2 x quad core
32 gb ram
WAR:
Context of the web on a virtual host
War file size ~ 20mb
Number of classes ~ 1,5k
Number of files together ~ 7k
I tried to clean up the file and I went down to 700kb (I removed all the img / js / css etc and all lib/), the time shortened to ~ 2 seconds.
Thanks in advance

Change log file name during runtime - ent lib

I have a WCF service that will serve multiple clients. I'm using ent lib for the logging.
I'd like to have a different log file for each client. is there a way to change the file name back and forth?
I found a few threads but they all talk about editing the config file during runtime.
ALso found this: Enterprise Library Logging but it talks about environment variables. I will set the log name according to the client id.
Thanks
Avi
You can have distinct categories linked to individually configured FlatFile or RollingFile tracelisteners for each client.
If filenames are unknown till runtime, consider using fluent API for configuration, like so:
http://msdn.microsoft.com/en-us/library/ff664363(PandP.50).aspx#fluent_api_logging

Binding a queue to an EJB 3.0 MDB in WebSphere 7

I'm writing, or trying to write, Baby's First MDB on WebSphere 7. I have nearly no hair left, having pulled it all out trying to get the thing to work. It appears that I've got everything set up right, but I get no response when I put a message to the associated queue.
Here's the EAR file setup:
simplemdb.ear
META-INF
Manifest.mf
application.xml
simplemdb.jar
META-INF
Manifest.mf
ejb-jar.xml
com
[ classes go here ]
I can't find any syntax for defining the queue's JNDI name in ejb-jar.xml, so instead I:
Define a WebSphere activation spec. Name SimpleMDBActivationSpec, JNDI name jms/SimpleActivationSpec, Destination jms/SimpleMDBQueue.
Define a WebSphere queue. Name SimpleMDBQueue, JNDI name jms/SimpleMDBQueue, Queue name SIMPLE.MDB.QUEUE.
Define an MQ queue, name SIMPLE.MDB.QUEUE.
Deploy the EAR file. During the deployment, I'm asked to enter binding information. I select Activation Specification, then point the Target Resource JNDI Name and Destination JNDI name at the activation spec and queue, respectively.
(The MDB code has no annotations.) At this point, the app points to the spec and queue, and the spec points to the queue - belt and suspenders. Naturally, I imagine that the app therefore knows about the queue. Full of hope, I put a message on the queue, and ... nothing. The onMessage event is supposed to use System.out to log a message. I see no message.
Clear documentation on this is conspicuous by its absence. Google gives LOTS of results, but none of them details how the configuration all fits together. There's lots of hand-waving about ibm-ejb-jar-bnd.xmi, but examples of the file are arcane, full of opaque numbers with no explanation about how they were generated, or how they relate to other parts of the configuration.
For goodness' sake. All I want to do is deploy an MDB, and have it write "Hello, world" when I put a message to a queue. I'm using vi and ant as my development and build tools. Can anybody out there give me an idea about what I'm missing?
Edit: "zos" tag added.
I found the problem. It's specific to WebSphere running on z/OS. For an activation spec to be fully available in that environment, the Control Region Adjunct (CRA) process must be started. I told WAS to start it up, recycled the app server, and lo! My MDB started responding.
To make the CRA start via the WebSphere Admin Console, go to ...
Application servers > [server name] > Communications > Messaging > WebSphere MQ CRA Settings
... and check the box that says "Start CRA". Hit OK, save it to the master configuration, and to make the CRA actually start, bring the app server down and back up. (This is for WAS 7.0.)
Thanks to everyone for their time and thoughtspace.
have a quick look at this and see if there is anything here that helps you.
http://publib.boulder.ibm.com/infocenter/ieduasst/v1r1m0/topic/com.ibm.iea.wasfpejb/wasfpejb/6.1/DevelopmentTools/WASv61_EJB3FP_MDBLab.pdf
I haven't played with this for the last one year so i am not able to comment straight away but i thought the PDF might be of some assistance to you.
HTH
Manglu