Been trying to reproduce this, but I cannot, admitably I'm
using CVS
I checked out inetutils from CVS. The sigpipe still happens for me
on my machine after letting the connection silently time out -- not
100% of the time, but >90%. I'll append the backtrace, not that
it's especially illuminating. The sigpipe happens in the `fflush
(cout);' call in the `command' function.
Hmph. Can you put a break point at the fflush, and then examine the
content in gdb? You can do function calls from gdb, so right before
fflush, you can do:
p fstat (cout, ...)
and such, I think. This bug is getting far to obscure...
(BTW 1, I don't see a bootstrap script, just bootstrap.conf, but I
copied bootstrap from coreutils and that seemed to get things going.
Tss... You didn't read the README-alpha file. :-)
BTW 2, shouldn't there be Copyright 2007 Free Software Foundation (or
whatever) added to the copyright notices?)
Yeah, it is a (serious) bug that is on my todo list. Many files have
incorrect notices and some only the usual MIT-style license. Sighs.
And now with work, even less time.
I tried on another host with the same libc and kernel and it worked :(.
Guess it's something about my terribly (satellite) connection here. I
could offer to give you access but I fear it will be unusably slow
for debugging coming in from outside.
Can you try it on the host that works, but over a satellite
connection?
Thanks.
_______________________________________________
bug-inetutils mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/bug-inetutils
