FastSwap Feature of WebLogic Server 10.3

I’m the first to admit that if you are looking for breaking news, I am not the source. Broken news, maybe…

Anyway, I just noticed the FastSwap feature for WLS 10.3. This allows for updating classes in a running instance without having to do a full redeploy. If you have spent anytime developing WebLogic Portal applications, you know what a boon this is.  Two of the major productivity killers in building  WLP solutions is the time it takes to redeploy the whole portal to test every little change, and eventual Out of Memory errors that will occur with repeated redeployment of the same application.

Turning on the feature is a simple matter of updating weblogic.xml and weblogic-application.xml (see the documentation for the specifics).

One note of caution: This is for development use only. If your deployment processes do not include either deployment plans or environment-specific deployment descriptors, you will need to maintain a local copy outside of your normal source control.

If you found this interesting, please share.

© Scott S. Nelson

Patch for WLS 10.0 ORA-00001 Error Messages in Logs

While the message does not impact functionality, if you have to read the logs it becomes tedious to scroll past the following over and over:

<BEA-403302> <An unexpected SQL exception occured java.sql.SQLException: ORA-00001:
unique constraint (PORTAL.PK_LEASE) violated.
java.sql.SQLException: ORA-00001: unique constraint (PORTAL.PK_LEASE) violated
at oracle.jdbc.driver.DatabaseError.throwSqlException(DatabaseError.java:112)
at oracle.jdbc.driver.T4CTTIoer.processError(T4CTTIoer.java:331)

Not to mention how the fun of having every person who hasn’t looked at your application before reporting it as a problem or insisting it is the root of a bug you are chasing (it isn’t).

If you have a support account, you can surpress this with the following WLP 10.2 patch:

VZGR | 9AS6D7D9

If you found this interesting, please share.

© Scott S. Nelson

WebLogic Workshop 10.2 Debugging

On a very long, large project where there are multiple portals I will create separate worskpaces for the different portals. Especially when developing shared libraries, to save time in unit testing and debugging. Once in a while, I would get this annoying message about the WebLogic Server not starting within the timeout period. The server does eventually start, but you can’t debug without restarting.

Later in the project I had to run all of the portals in Workshop at once. Being able to debug become the norm, which is really annoying when trying to find bugs in code you have never seen before. I did some digging and found that the timout period was made configurable in an older version of Workshop. Originally, there were several versions of Workshop. Some were versions for specific products, like WebLogic Portal. Then they combined them in all into one. Unfortunately, the ability to configure the timeout was only in the WLP version and did not make it into Workspace Studio (the actual new name for Workshop that no one seems to use).

Long story short (yeah, I know, too late), I got the feature ported forward to 10.2. It is patch WW5U | F21UMQIN. Once the patch is installed, edit workshop.ini and add the following:

-Dweblogic.deployment.timeout=3600000
-Dweblogic.startserver.timeout=3600000
-Dweblogic.stopserver.timeout=3600000

3600000 is number of milliseconds, and is the maximum setting.  You can set it for whatever you want less than that, but why would you?

If you found this interesting, please share.

© Scott S. Nelson