Comment 4 for bug 359192

I am having a real problem seeing this as a 'rm' (and, by extension, coreutils) issue: although apport failed on the retrace, the original stacktrace shows only frames from glibc and ld-linux-x86-64.so... so this happened very early in the startup, seemingly *before* control was given to 'rm'.

So... I am adding a task on glibc, and would like the input from the glibc maintainers.