The problem has been reported on the ML and is stored here so that it's not forgotten:
> >> 10828 0 280 S sh -c mount /dev/fd0u1722 /tmp/tmpfs/mondo.tmp.17767/ ...
> >> 10829 0 460 D mount /dev/fd0u1722 /tmp/tmpfs/mondo.tmp.17767/mount. ...
> >>
> >> But, I don't hav a floppy. It looks, that mondorescue try to kill the mount process after the start with a delay of 15 seconds, but nothing happens. The manual kill of the >>process with -9 fails too.
I checked the code, and you're right. The mount is done unconditionaly
:-( And after 15 seconds it's killed. In your case, as the access
remains in the D state, it will never return, thus the issue.
What you can do is comment the lines:
sprintf(command, "mount /dev/fd0u1722 %s", mountpt);
sprintf(tmp,
"(sleep 15; kill `ps | grep \"%s\" | cut -d' ' -f1` 2>
/dev/null) &",
command);
mr_msg(1, "tmp = '%s'", tmp);
system(tmp);
res = run_program_and_log_output(command, FALSE);
if (res) {
sprintf(command, "mount /dev/fd0H1440 %s", mountpt);
res = run_program_and_log_output(command, FALSE);
}
in mondo-rstr-tools.c and replace it with:
res=0;
I need to find a better solution of course.
Should be fixed in rev [1496]