Mailing-List: contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm Sender: cygwin-owner AT sourceware DOT cygnus DOT com Delivered-To: mailing list cygwin AT sourceware DOT cygnus DOT com Message-ID: <3729BC6D.277F9A0@bgs.com> Date: Fri, 30 Apr 1999 10:21:33 -0400 From: Eugene Kanter X-Mailer: Mozilla 4.51 [en] (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 CC: cygwin AT sourceware DOT cygnus DOT com Subject: Re: tail -f locks References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Mumit Khan wrote: > > On Tue, 27 Apr 1999 thomas AT on DOT com wrote: > > > This has not been the original intention of 'tail -f': This command should > > enable the user > > to watch the contents of a file grow. If this file is locked by the tail > > command, the writing process > > will run on error. > > > > I'm not sure why you're seeing a problem, but I do the following all the > time: > > $ make > make.log 2>&1 & > $ tail -f make.log > > And in some cases, the ``tail -f make.log'' is already running in a > different window when I issue the ``make'' command. > > Is the writer explicitly locking the output file? Could you show an > example of the failure? Here is an example: bash window 1: bash window 2: bash-2.02$ touch mappfile.out bash-2.02$ tail -f mappfile.out bash $ cat > mappfile.out line1 line1 line2 line2 ^C bash $ rm mappfile.out bash $ ls mappfile.out ls: mappfile.out: No such file or directory bash $ cat > mappfile.out bash: mappfile.out: Permission denied Is this enough for you to reproduce it? Thanks, Eugene. -- Want to unsubscribe from this list? Send a message to cygwin-unsubscribe AT sourceware DOT cygnus DOT com