|
 |
Thursday, December 16, 2004 |
Yes, it's time to breathe some life back into the Buttso Blog.
With the release of the OC4J
10g (10.1.3) Developer Preview 3, there's quite a bit of new
information to share, so the Buttso Blog will be reinstated to active
duty.
Some topics likely to seen here in the not too distant future are:
1. Using the new <oracle:deploy> ant tasks for application deployment operations with OC4J
2. Management stuff around JMX, MBeans, JSR77 stats, etc.
3. Shared libraries, classloading
4. A bit about clustering using the new application level clustering feature we have in 10.1.3.
And whatever else comes to mind when I get to my desk in the mornings.
12:22:06 PM
|
|
A Code Snippet to Create an AttributeChangeNotification
protected void setOrderProcessing(boolean val)
{
orderProcessing = val;
AttributeChangeNotification notification =
new AttributeChangeNotification(
this,
System.currentTimeMillis(),
System.currentTimeMillis(),
"OrderProcessing Changed",
"OrderProcessing",
Boolean.TYPE.getClass().getName(),
Boolean.valueOf(orderProcessing),
Boolean.valueOf(val));
sendNotification(notification);
}
12:13:18 PM
|
|
 |
Friday, June 04, 2004 |
A Trip to the Corner Shop .... An Expedition to Everest
Maybe because
it's a Friday afternoon, but I'm really struggling with getting the
latest Sun Blueprint application -- Adventure Builder -- to build
away from the Sun Reference Implementation environment.
I was really hoping to be able to do a quick build of the application
against our latest OC4J build, and then start to add some extensions in
the JMX arena to it as a demonstration. What started out as a
nice Friday afternoon task has ended up as bugger of a job. I
thought it'd be a simple trip to the shop ... now it's looking like a
full scale expedition up Mt Everest.
For a start, I certainly don't claim
to be an authority on best practices with Ant -- I got the older
Petstore app building against OC4J really quickly -- but I can
say that I'm finding it really hard to get this build environment to work against OC4J.
It just looks and feels too
complicated as I'm wading my way through all the build components
trying to change the relevant targets to work against another J2EE
implementation. There's lots of calls in the Ant scripts to
specific functionality in the J2EE RI to create the required resources
(can't these just get put into a completely separate script!), calls
out to ws compilers for all the client code which relies on the scripts
in the RI, and about a billion properties to keep track of and try and
debug when the shit hits the fan during a build.
If
I was someone looking at J2EE and thinking I'd use this as a way to
learn a bit about it on my preferred J2EE container, then I reckon I'd
be jumping at the .NET wagon rolling past ...
I guess I'll keep going -- perhaps the onus is actually on us as a J2EE
vendor to provide a distribution of this so it'll build against our
specific J2EE implementation. I know the Sun folks can't make it
that generic so it'll build against every product, but it does look
like it's very tightly coupled to the J2EE RI -- hmm, which I think is
now actually used as the official Sun Application Server .... perhaps
there is a method in their madness after all ....
Maybe I really am being a dumb arse and should just go out for a ride. Hye, good idea. Seeya.
5:01:52 PM
|
|
 |
Friday, May 21, 2004 |
Setting up OracleAS Clustering? Check Yer' Multicast Capabilities
Our blogger-in-residence Mr Debu Panda
forwarded me an email this week from someone who was trying to setup an
OracleAS file based cluster in order to demonstrate our HttpSession
failover capabilities, and who was getting an unexplained exception.
Turns out that they didn't have a
multicast enabled environment so our cluster enabled session listeners
were throwing the underlying exceptions.
To help people check first if they do
have a multicast enabled environment, I've written a multicast client
and server program which can be used to verify multicast is actually
working, see Mutlicast Checker for more details.
4:00:40 PM
|
|
 |
Wednesday, May 12, 2004 |
The Blog's Not Dead Yet
Things have gone quiet for a while;
I've been active on a few things which have taken up most of my time
over the last few weeks.
The good news is that I've got some more things to blog about soon. I
was finally forced to confront my demons in the areas of Single Sign
On, JAAS and OID that I'd previously been able to (... studiously ;-)
ignore. So in the next few days, I hope to write a bit about things
like the cool new SSO capability we have in the 904 release called
Dynamic Directives which adds programmatic single sign on operations
(login, force login, logout) to J2EE applications, JAZN SSO user
managers, and more things like that.
Time permitting of course!
4:03:07 PM
|
|
 |
Wednesday, April 28, 2004 |
OC4J Standalone as a Windows Service
Questions about running OC4J as a Windows Service appear on our OTN
J2EE/OC4J forum from time to time. I just had a real quick look
at one solution in OC4J as a Windows Service.
11:46:31 AM
|
|
 |
Friday, April 23, 2004 |
Form Based Authentication with OC4J
A few queries for an example of form-based authentication have appeared on OTN from time to time, so here's a small writeup and a working example.
For those folks who have it working, the next common question is why
does the browser URL show "j_security_check" once a succesful logon has
been made. I'm not sure about the answer to that -- to me it certainly
should show the URL of the page that was originally requested
(and is actually shown in the browser window). The PM who works on the
Servlet aspect of OC4J is looking into it for us.
3:44:23 PM
|
|
 |
Tuesday, April 20, 2004 |
© Copyright 2004 buttso.
|
|
|
|
December 2004 |
Sun |
Mon |
Tue |
Wed |
Thu |
Fri |
Sat |
|
|
|
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 |
10 |
11 |
12 |
13 |
14 |
15 |
16 |
17 |
18 |
19 |
20 |
21 |
22 |
23 |
24 |
25 |
26 |
27 |
28 |
29 |
30 |
31 |
|
Jun Jan |
|
My Musings
|
Other Links
|
|