X-Recipient: archive-cygwin AT delorie DOT com
X-Spam-Check-By: sourceware.org
Date: Wed, 2 Dec 2009 16:23:15 +0100
From: Corinna Vinschen <corinna-cygwin AT cygwin DOT com>
To: cygwin AT cygwin DOT com
Subject: Re: 1.7.0-67: syslog() not working properly - possible bug in     writev()  ?
Message-ID: <20091202152315.GS8059@calimero.vinschen.de>
Reply-To: cygwin AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
References: <1NFlZ1-0LCxsG0 AT fwd11 DOT aul DOT t-online DOT de>  <20091202100433 DOT GQ8059 AT calimero DOT vinschen DOT de>  <1NFn3Y-2JTP6W0 AT fwd06 DOT aul DOT t-online DOT de>  <20091202141753 DOT GR8059 AT calimero DOT vinschen DOT de>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20091202141753.GR8059@calimero.vinschen.de>
User-Agent: Mutt/1.5.20 (2009-06-14)
Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
Precedence: bulk
List-Id: <cygwin.cygwin.com>
List-Unsubscribe: <mailto:cygwin-unsubscribe-archive-cygwin=delorie DOT com AT cygwin DOT com>
List-Subscribe: <mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive: <http://sourceware.org/ml/cygwin/>
List-Post: <mailto:cygwin AT cygwin DOT com>
List-Help: <mailto:cygwin-help AT cygwin DOT com>, <http://sourceware.org/ml/#faqs>
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
Delivered-To: mailing list cygwin AT cygwin DOT com

On Dec  2 15:17, Corinna Vinschen wrote:
> On Dec  2 12:07, Christian Franke wrote:
> > Corinna Vinschen wrote:
> > > On Dec  2 10:31, Christian Franke wrote:
> > > > I presume that the root of the problem is that the
> > > > 
> > > > writev(fd, { {"< PRI >", . }, { "MSG", . } }, 2) 
> > > > used within syslog() sends "< PRI >" and "MSG" in
> > > > two separate datagrams to /dev/log.
> > > > 
> > > 
> > > Probably not.  The "kernel:" message is generated if something is
> > > available on /dev/kmsg, which shouldn't be the case.  The only message
> > > type generated there are exceptions catched by Cygwin.  I'll have a
> > > look into it.  Thanks for the report.
> > > 
> > > 
> > 
> > Another test:
> > 
> > This sends one UDP package on 1.5, but two on 1.7:
> > 
> >      int sd = socket(AF_LOCAL, SOCK_DGRAM, 0);
> > 
> >      struct sockaddr_un sa; sa.sun_family = AF_LOCAL;
> >      strcpy(sa.sun_path, "/dev/log");
> >      connect(sd, (struct sockaddr *)&sa, sizeof(sa));
> > 
> >      struct iovec iv[2] = { { "<1>", 3 }, { "test: text", 10 } };
> >      writev(sd, iv, 2);
> 
> Yes, but... why is that a problem for syslog-ng?  Shouldn't it have
> to expect that a message comes in multiple UDP packages?  Hmm.

Ok, it's definitely a problem in Cygwin.  Fortunately one which is easy
to fix.  The send() function splits messages into chunks to avoid the
problem from KB 823764.  However, it's not correc to do this for
datagram sockets, so I added code to avoid this.

The "kernel:" tag was also a Cygwin problem, apparently.  The default
facility is set to 0, which is equivalent to LOG_KERN.  The "kernel:" is
added by syslog-ng in this case.  I changed vsyslog() so that LOG_KERN
messages from user space are not possible anymore since the default
facility is always set to LOG_USER if it's not set to some other
non-LOG_KERN value already.


Thanks for the report,
Corinna

-- 
Corinna Vinschen                  Please, send mails regarding Cygwin to
Cygwin Project Co-Leader          cygwin AT cygwin DOT com
Red Hat

--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple