>From owner-linux-kernel-outgoing@vger.rutgers.edu Tue Aug 18 09:37 EET 1998
Message-ID: <19980818012446.A1233@aikon.incarnate.net>
Date: Tue, 18 Aug 1998 01:24:46 -0500
From: Andy Sloane <andude@guildsoftware.com>
To: linux-kernel@vger.rutgers.edu
Subject: Sockets permanently in CLOSE_WAIT state.
X-Mailer: Mutt 0.94.1i
X-Orcpt: rfc822;linux-kernel@vger.rutgers.edu
Sender: owner-linux-kernel@vger.rutgers.edu
Precedence: bulk
X-Loop: majordomo@vger.rutgers.edu
> While hacking away at a project today I discovered a rather strange
> occurrence in a multithreaded environment using sockets. The jist of the
> problem is I was select()ing on a socket in one thread and at the same time
> closing it in another (yes, this is a design oversight on my part). To my
> surprise, after killing the server and all clients, the sockets remained in
> the CLOSE_WAIT state, yet no client processes were still running; I
> expected to at least see a thread stuck in a D state or something. It's
> not a 'normal' thing to do, but it's also something that an attacker could
> potentially use to run the system out of non-root sockets. Is this a known
> issue? (for the record I'm using linuxthreads.)
Same problem here. many threads, all of them in select() most of time,
anyone may try to pthread_cancel others. Select waits for 2 things:
is data coming from network or pipe. If from pipe, it goes to
infinite loop, just for letting pthread_cancel himself. Under
IRIX works fine, problems under linux. Also sometimes the
program don't die properly when i kill -9 it. And i saw two
general protection messages in dmesg...
>
> -Andy.
Lauri Jesmin <jesmin@ut.ee>
See lugu puudutas inimesi.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html