Whilst performing a revdep-rebuild: Checking dynamic linking consistency... broken /opt/sun-jdk-1.5.0/share/demo/jvmti/mtrace/lib/libmtrace_g.so (requires libjava_crw_demo_g.so) broken /opt/sun-jdk-1.5.0/share/demo/jvmti/heapTracker/lib/libheapTracker_g.so (requires libjava_crw_demo_g.so) done. These are the packages that I would merge, in order: Calculating dependencies ...done! [ebuild Rf ] dev-java/sun-jdk-1.5.0 Re-emerging sun-jdk-1.5.0 does not resolve this problem, the specified library file is missing. It appears this is only required by the demo apps, so is more of a nuisance than anything (hence the low priority).
yet another "same here".
*** Bug 78218 has been marked as a duplicate of this bug. ***
Same problem here on AMD64. Might be a JDK problem though which cannot be fixed in the ebuild.
upstream bug, not possible for us to fix it: http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6227549
Seeing that it is still in jdk-1.5.0_03. It seems to be fixed by linking ln -s libjava_crw_demo_g.so libjava_crw_demo.so in /opt/sun-jdk-1.5.0.03/jre/lib/i386
sorry; I had the wrong order arguments before. :( It should be, ln -s libjava_crw_demo.so libjava_crw_demo_g.so
*** Bug 97016 has been marked as a duplicate of this bug. ***
I think we should put the symlink to the ebuild because it seems to solve this problem.
This wasn't fixed in sun 5.0_04 release either and most likely wont be ever fixed. I think the ebuild should just run find -name "*_g.so" -exec rm {} \; after extracting the jdk. The sun bug mentions that the _g (debug libraries) should not even be in the distribution package.
The upstreadm sun-jdk-1.5.0_06 has fixed this problem. I just verified this by creating my own ebuild for it. This means that this bug can be closed after #114334 has been fixed.
06 is now in the tree and running ~x86 revdep-rebuild with it does not give any complaints. Thanks for everyone who participated in this bug and reopen if this issue pops up again.
