On Sun, Sep 25, 2005 at 12:40:03AM +0200, Jan Gorski wrote: > On Sat, 24 Sep 2005 at 15:21:54 -0400, Thomas Dickey wrote: > > If you can kill the mc process from another xterm without killing the xterm, > > it is likely that the problem lies in mc rather than xterm. > > I can kill mc, but xterm is still frozen. Similar thing happens with mutt. > BTW, > is it possible and not considered a bug, that application can leave xterm in > frozen state even after it has been killed? I cannot even raise any menu with > ctrl click. How can I investigate this?
That does sound like a bug. But (I've not updated for about a month), a quick check doesn't show it breaking for me, given the locale, etc. Since I don't recall an xterm bug that looks like this, I'm thinking it is in the X libraries (though it's still possible that it's xterm of course). If I were seeing the bug, I'd try compiling xterm and running it with its debug-trace, to see if there was some useful information - to point to an xterm bug. (That would be "configure --enable-trace --enable-wide-chars --enable-luit", and the executable produces Trace-parent.out and Trace-child.out) Other than that - attaching gdb to the xterm process and doing a bt (stack trace) should show where it is executing. -- Thomas E. Dickey http://invisible-island.net ftp://invisible-island.net
pgpTn4zCgaQjX.pgp
Description: PGP signature
