From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <3DA46D8B.5010808@mvista.com> From: Steven Dake MIME-Version: 1.0 Subject: Re: [linux-lvm] Broken compile scripts in LVM References: <3DA1C82F.5080406@mvista.com> <20021009125221.A23965@sistina.com> Content-Type: multipart/alternative; boundary="------------030000000209030305040903" Sender: linux-lvm-admin@sistina.com Errors-To: linux-lvm-admin@sistina.com Reply-To: linux-lvm@sistina.com List-Help: List-Post: List-Subscribe: , List-Unsubscribe: , List-Archive: Date: Wed Oct 9 13:40:32 2002 List-Id: To: linux-lvm@sistina.com --------------030000000209030305040903 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Heinz, Thanks Jeremy had asked me to forward the patch on. I guess he also did it himself -steve Heinz J . Mauelshagen wrote: >Steven, > >I have changed that in CVS (Branch: LVM_BRANCH_1-0) a couple of days ago >and announced that on lvm-devel (the list for patches and development >discussions). Jeremy A. Puhlman pointed it out on October 4th already. > > >On Mon, Oct 07, 2002 at 10:45:19AM -0700, Steven Dake wrote: > > >>Folks, >> >>Some of the scripts use echo "string" > /dev/stderr when instead they >>should be using echo "string" >&2. I have attached a patch that fixes >>this problem. The result is that in automated builds, the stdout gets >>stomped if you use the first technique (resulting in loss of logging data). >> >>Thanks. >> >>-steve >> >> > > > >>diff -u3 lvm-1.0.5/scripts/iopver lvm-1.0.5-fixed/scripts/iopver >>--- lvm-1.0.5/scripts/iopver Tue Jan 22 06:54:13 2002 >>+++ lvm-1.0.5-fixed/scripts/iopver Thu Oct 3 19:46:40 2002 >>@@ -48,21 +48,19 @@ >> sed -e 's/.*LVM_LIB_IOP_VERSION[[:space:]]*$[[:digit:]]*$.*/\1/'`; >> >> if [ -z "$ver1" ]; then >>- echo "Error: iopver can't determine the IOP version from $file1" > \ >> >> > > > >>- echo "$new not found" > /dev/stderr >>+ echo "$new not found" >&2 >> exit 2; >> fi >> >> >> > > > > --------------030000000209030305040903 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit Heinz,
Thanks Jeremy had asked me to forward the patch on. I guess he also did it himself

-steve

Heinz J . Mauelshagen wrote:

Steven,

I have changed that in CVS (Branch: LVM_BRANCH_1-0) a couple of days ago
and announced that on lvm-devel (the list for patches and development
discussions). Jeremy A. Puhlman pointed it out on October 4th already.


On Mon, Oct 07, 2002 at 10:45:19AM -0700, Steven Dake wrote:

Folks,

Some of the scripts use echo "string" > /dev/stderr when instead they 
should be using echo "string" >&2.  I have attached a patch that fixes 
this problem.  The result is that in automated builds, the stdout gets 
stomped if you use the first technique (resulting in loss of logging data).

Thanks.

-steve

diff -u3 lvm-1.0.5/scripts/iopver lvm-1.0.5-fixed/scripts/iopver
--- lvm-1.0.5/scripts/iopver	Tue Jan 22 06:54:13 2002
+++ lvm-1.0.5-fixed/scripts/iopver	Thu Oct  3 19:46:40 2002
@@ -48,21 +48,19 @@
 	sed -e 's/.*LVM_LIB_IOP_VERSION[[:space:]]*\([[:digit:]]*\).*/\1/'`;
 
 if [ -z "$ver1" ]; then
-	echo "Error: iopver can't determine the IOP version from $file1" > \

<SNIP>

-	echo "$new not found" > /dev/stderr
+	echo "$new not found" >&2
 	exit 2;
 fi

--------------030000000209030305040903--