Saurabh Gokhale

Hi

I have following questions about Part 3 exam interface

1> How is the interface designed for Part 3 exam? Does it has question and the text area below to give answer?
2> Does the answer area has rich text support like creating bullets or spell check?
3> Is there a limit on how much you can write in the answer area? Or you can write as much as the time permits
4> I understand from the previous post that you can move back and forth between all the questions? is it correct?
5> Are there fixed 8 questions to be answered in 90 mins? or the number of questions vary?


Thanks

Saurabh

Hi,

I have following setup in my project

remote java client running RMI over HTTP tunneling calls to weblogic server sitting behind apache proxy server.

Client --> Apache HTTP Proxy Server (running weblogic proxy) --> multiple Weblogic 9.2 servers.

The client is supposed to lookup EJBs deployed on multiple weblogic servers using single entry point which is an HTTP apache server.

Client looks up initial context using PROVIDER_URL as http://APACHE_IP:PORT/ABC where ABC tells apache server to forward the call to weblogic server A. If client EJB lookup URL has XYZ (http://APACHE_IP:PORT/XYZ) then apache needs to route the call to weblogic server B.

Apache server has settings to lookup the end of the URL and forward the call to appropriate weblogic server.

This setup is not working because whenever a request is made from the client, regardless of the provider URL given, all calls from the client gets converted by weblogic.jar (running on the clients classpath) to something like

/bea_wls_internal/HTTPClntLogin/a.tun?wl-login=http+dummy+WLREQS+8.1.5+dummy+%0A&rand=3524334727933247191&AS=255&HL=19 HTTP/1.1

I never get my unique sting ABC or XYZ when the call reaches apache server (access log shows something like below instead of actual URL from the client.
127.0.0.1 - - [13/Jan/2011:19:28:20 -0600] "GET /bea_wls_internal/HTTPClntLogin/a.tun?wl-login=http+dummy+WLREQS+8.1.5+dummy+%0A&rand=3524334727933247191&AS=255&HL=19 HTTP/1.1" 200 342)

Is there any way I can get my real URL at the server and then logically forward the call to the appropriate backend weblogic server? Has any one faced this issue before?

Hi All,

Few months back, we talked on this forum how some people received Sun certified security administrator certificate when they downloaded SCEA part 2 assignment project.

I was one of them. I tried following up with Sun by sending mails to [email protected] and [email protected] but I never got any reply back.

Now I am ready to give part 3 and submit part 2 assignment. I am wondering does Sun has any information as to what assignment I downloaded because as per their certification I7 site, I gave and passed security administrator exam.

Do I have to wait until I get a clarification/confirmation from them which I dont think they will or should I just go ahead with part 3 and then send part 2 assignment in the email. My only worry is, after sending part 2 project assignment they should not argue that I never downloaded part 2 assignment from the site.

Thanks

Saurabh

Yes, we can display jsp pages as part of class diagram. Though displaying JSP pages is not mandatory, remember jsp does get compiled into regular java class so technically it is not wrong to display them on the class diagrams.

I dont think, just by implementing design patterns, we can address NFRs. Though in case of Performance I can definitely have my class diagram following proper J2EE design patterns like session facade, or service locator caching home/remote references to avoid multiple lookups but that will not be enough.

What I would do instead is as follows:

Performance
Class Diagram:
1> Draw class diagram following J2EE design patterns [as given above]
2> Display some specific performance monitoring classes like Aspect Oriented Programming classes for performance and then comment on how they help segregate the code and help improve performance.
3> Talk about how performance test will be perform simulating production load in testing environments to identify performance bottlenecks before code reaching production.

Availability:
Deployment diagram:
1> Draw deployment diagram with multiple instances of servers balanced by load balancer.
2> Draw failover nodes.
2> Draw other clustered back end environment.
3> Talk about how the hot fix process will be set to perform rolling bounce of production servers [bouncing servers one after the other and not bounce everything together] to have 24*7 availability.

I think NFRs have to be managed by software and hardware changes and not just one alone.

yes, it can be shown with <<external>> as a stereotype but along with that I would also recommend to add a text at the left of the box [provided you draw one] with the text "External Systems (External to JVM)".
This way it will be very clear to the examiner that this component is external to your SuD.

Hi,

I am developing an application with the following structure

JSF2 (xhtml pages) --> Managed Bean --> Session facade (EJB3) --> JPA --> Database. (multiple JSF pages have multiple MBeans)

I want to enable performance logging in this application by writing timings to a text file which will have each logical activity against the time taken in seconds.

I am not able to find a single entry point in this architecture where I can write the method entry and exit logging statements. In such architecture what is the best location to write logging methods?

For example:
In case of Struts application, I can have a generic struts controller (action class), which all my application actions can extend. This controller can call preExecute, Execute and PostExecute methods on the application action classes. In this controller I can calculate start and end time and can print the same at the end of the method call. This way I do not have to write this code in each and every action class doExecute methods.

But in case of JSF/ ManagedBean application I am not able to come up with such a single entry point (Except my session facade) for logging performance data.

Thanks Das,

Yes, assumptions is definitely a good section to detail about these NFRs

Thanks for the suggestion.

Saurabh

Hi Remko ... thanks for the reply.

But I am not sure I completely understood your point.

So if you want to show something about performance just model only performance and explain why. Don't put performance modelling in a class model which is used to explain how the software is put together. Nothing more.

Are you saying that I should have a separate UML diagrams to depict these characteristics instead of adding them into the existing ones (especially existing class diagram) which would complicate the existing diagram?

I was thinking of displaying a Performance measuring Servlet (example: PerformanceServlet) as part of my class diagram hierarchy.

Hi,
While working on UML diagrams as part of SCEA Part 2 submission, how can a stress be given on the non-functional characteristics of the system like
■ Performance
■ Scalability
■ Reliability
■ Security
■ Availability
■ Maintainability

Which diagrams (Class, component, sequence etc) should be stressed more to display information about these requirements?

As per my understanding:
■ Performance --> Class diagram (by specifying performance specific class details) and sequence diagram (displaying performance logging mechanism)
■ Scalability --> Deployment diagram (Displaying array / cluster of hardware)
■ Reliability/ Availability --> Deployment diagram (displaying parallel / fail over architecture)
■ Security --> Class diagram, sequence diagram (displaying a flow when invalid/ unsecure request is received) and deployment diagram (showing network artifacts like LDAP and HTTPS traffic)
■ Maintainability --> Class Diagram (displaying good design pattern structure for easy maintainability)

Can some one shed more light if I am going in wrong direction?

Thanks... this information is really helpful

Hi,

I have following questions regarding SCEA Part 2 project work:

1> How many UML diagrams are expected or considered to be enough to fulfill the requirements
-- I know it would depend on the problem statement but in general is there a "good/enough" count?
-- Is it expected to have only 1 class diagram capturing entire system or we can break it down into multiple class diagrams. If we can break down, how can we display the relationship between them. (NOTE: Deliverable section says Create "a" class diagram ...
-- Assignment specifies multiple use case scenarios so those many use case diagrams are must but can we have any addition to those diagrams or can we merge those diagrams due to their logical grouping?

2> Also as per assignment requirements, the submissions should have HTML pages only. Can we add an explanation in plain English for the diagram below the diagram itself on the page?

3> Can there be only 1 big HTML page or every diagram type can have its own page ? or it doesn't matter?

4> In general what precautions should be taken while submitting assignment?

Hi .. sorry .. I was travelling so could not reply...

I am using my own setup by creating stack structure to store tags and begin and end events to keep track of which tag was read and whether it matches the XPath expression.

I will definitely try reading it as a byte stream and see if it makes any time difference. Thread is also another good option rightly suggested by you but its going to be little tricky to use XPath later.

Hi William, thanks for replying.

Yes I am using a non standard interpreter which is giving me an ability to read the XML using SAX and implement XPath on it.

One good thing about this XML issue is, I at least dont have to worry about namespaces.

You are right, I may be able to consider it as a text search instead of considering as an XML. It will at least give me some idea whether the tag I am looking for exists in the XML or no. But I may still end up reading a same tag (same tag name) under a different parent.

Will it improve the performance if I treat this as a text search instead like an XML? How should I go about this route?

Hi,

I am working on some project which requires me to read an xml as big as 2GB in size and parse it in less than few seconds.

My requirements are:
1> Read large XML (in GB size)
2> Use XPath to locate a node and a value
3> Parse the XML within a timeout period of 5 seconds.
4> Use JDK 1.4
5> Limited memory usage (can not use DOM technique)

Currently I am using SAX and XPath combination which is giving me good results if that particular node exists. But the problem occurs when either the node does not exists and I have to parse the entire XML to find that it is not there or it exists at the very end of the XML. Parsing complete XML goes way above 5 seconds.

Any one has any idea as to what combination technique can give me this ability to read GB XML in less than few seconds with limited memory usage with JDK 1.4?