This Bugzilla instance is a read-only archive of historic NetBeans bug reports. To report a bug in NetBeans please follow the project's instructions for reporting issues.
Created attachment 112604 [details] gdb log This morning both the ide and gdb were very confused. I don't know what led up to this, so I started by: - remove all breakpoints - exit ide - rm -rf ~/.netbeans/dev/var/cache - rebuild binaries - restart ide - run debugger There are a number of things wrong. First is that if I disable or delete breakpoints, they come back when I run. Second is that there are two errors in the gdb log that are unreported in the ide: - 12^error,msg="mi_cmd_var_create: unable to create variable object" - ~"../../gdb/thread.c:614: internal-error: ... gdb log attached. It seems that the ide should at least try to report errors from gdb so that the user is not stuck scratching his head. Also, is there a mismatch between gdb and the ide? Is that why gdb is getting an internal error? Note that if I run gdb by hand on this binary, set breakpoints, and run, gdb works just fine. Product Version: NetBeans IDE Dev (Build 201110200600) Java: 1.6.0_23; Java HotSpot(TM) Client VM 19.0-b09 System: Linux version 2.6.35.13-92.fc14.x86_64 running on i386; UTF-8; en_US (nb) User directory: /home/toddb/.netbeans/dev Cache directory: /home/toddb/.netbeans/dev/var/cache
removing ~/.netbeans and installing the version below seems to work. Shouldn't the ide have some internal diagnostics to notice when things are going wacky? Product Version: NetBeans IDE Dev (Build 201110300600) Java: 1.6.0_23; Java HotSpot(TM) Client VM 19.0-b09 System: Linux version 2.6.35.13-92.fc14.x86_64 running on i386; UTF-8; en_US (nb) User directory: /home/toddb/.netbeans/dev Cache directory: /home/toddb/.netbeans/dev/var/cache
Todd, it looks like watches are the case, next time try to delete all watches first. Probably gdb doesn't survive when we try to create watches and start program at the same time. Diagnostics with the feedback in the IDE also needed for sure.
> next time try to delete all watches first. I did not think of the watches. Good idea.
I am back in a debugger-induced fog. At one point, I added a watch for a global/static class instance. The next time I started the debugger it simply hangs and all I can do is stop. The gdb log show that it is clearly not happy. See attached.
Created attachment 114120 [details] gdb log Note: the type="Debugger" in the log is the type of global/static instance. At the time I added it, I was able to read the variable. After restarting, gdb no longer works *and* I cannot delete the watch!!
I can delete the watch from the watch window when the debugger is not running, and as long as the debugger is not running no watches appear. However, the watch that I deleted returns every time I (try) to start the debugger.
Removing ~/.netbeans/dev/var/cache finally removes the watch. Looks like some persistent data is kept there that should be synchronized with the UI.
I can reproduce "ghost" watches issue but after IDE restart they disappear, will fix anyway
steps to reproduce: - create a number of watches - restart the IDE - delete some of watches - start debugging All watches appear in the debug session (including deleted)
"ghost" watches issue fixed in: http://hg.netbeans.org/cnd-main/rev/e43820c176f1
I've create bug 206784 about watches causing gdb to fail
Integrated into 'main-golden' Changeset: http://hg.netbeans.org/main-golden/rev/e43820c176f1 User: Egor Ushakov <gorrus@netbeans.org> Log: fixed #204460 (gdb confused)
verified in NetBeans IDE Dev (Build 201201110601)
integrated into release71_fixes: http://hg.netbeans.org/releases/rev/703ec9f9668a
verified in NetBeans IDE 7.1 (Build 201112071828)
Verified in NetBeans 7.1 with updates from 201202141941