Java SE 6 troubleshooting

Some interesting additional java parameters: B.1 HotSpot VM Command-Line Options – Troubleshooting Guide for Java SE 6 with HotSpot VM Option to automatically make memory dump when OutOfMemory occurs sounds rather interesting. These utilities, should exist in all java deployments: jps: List PIDs of java processes jmap: Make heap of java process (jmap -dump:format=b,file=test.hprof PID)…

Read more

Modifying remote webstart launch file locally

It seems that it’s possible to save remote webstart file locally. And while still using remotely distributed jar files for the webstart, modify at least few webstart parameters in locally saved jnlp -file. However, this trick seems to work only with JDK5 version of webstart, not the one in JDK6. JDK6 seems to dutifully retrieve…

Read more

JDK 6…, improvement over JDK 5…, not

For example, following issues seem to be persistently existing: Bug ID: 6566201 JNLP ClassLoader performance degraded in 1.6 significantly for the signed(?) jar Bug ID: 6532373 xcb_xlib.c:50: xcb_xlib_unlock: Assertion ’c->xlib.lock’ failed. First issue means huge slow down in webstart application startup. When doing some profiling in webstart started application, it’s consuming huge amount of memory…

Read more

Detecting from where code is called

Detect if current piece of code is called by own classes: [code lang=”java”] /** * @return True if current execution points originates from own classes */ public static boolean isOwn() { StackTraceElement[] stack = Thread.currentThread().getStackTrace(); int len = stack.length; for (int i = 2; i < len; i++) { if (stack[i].getClassName().startsWith("org.kari.")) { return true; }...

Read more

Calculating JTextComponent -height

Problem: Let’s assume that we know the available width and we should determine what is the required height for JTextComponent within that limitation. This problem arises immediately, at least in three cases in java: Variable row height table cells Tooltip size Message dialog size In all of these cases, it is necessary sometimes to know…

Read more

Eclipse 3.3 is crappy slow

After all it turns out that crap factor has overcome over the usability factor. Basically Eclipse 3.3 contains some improvements to speed things up. However, in reality it’s just slower then 3.2.x. Reason for this strangeness is the fact that Eclipse 3.2, made big speed improvement by starting to use jar packaged plugins. This reduced…

Read more

RMI over zipped stream

If large amounts of compressable data is transferred over RMI, then valid question to ask is if compression would make any benefit there. However, before going into compression of the outputstream, first thing is to do algorithmic changes to reduce traffic. I.e. no compresion can match algorithm which avoids reduntant traffic completely. Also compression adds…

Read more