* automatic bootstrap failure notification
@ 2002-08-06 5:19 Robert Schiele
2002-08-06 5:56 ` Daniel Jacobowitz
2002-08-14 9:53 ` Janis Johnson
0 siblings, 2 replies; 6+ messages in thread
From: Robert Schiele @ 2002-08-06 5:19 UTC (permalink / raw)
To: gcc
[-- Attachment #1: Type: text/plain, Size: 592 bytes --]
Hello,
as some may have already noticed, I am sending the results of the
regression tests after successful bootstrap processes to
gcc-testresults from my automatic build system for some time now.
I am currently adding a feature to this build system that allows
sending automatic notification messages with some relevant information
when bootstraping fails.
Would sending such messages to gcc-testresults be welcome or would
such messages be seen as annoying?
Robert
--
Robert Schiele Tel.: +49-621-181-2517
Dipl.-Wirtsch.informatiker mailto:rschiele@uni-mannheim.de
[-- Attachment #2: Type: application/pgp-signature, Size: 524 bytes --]
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: automatic bootstrap failure notification
2002-08-06 5:19 automatic bootstrap failure notification Robert Schiele
@ 2002-08-06 5:56 ` Daniel Jacobowitz
2002-08-06 6:20 ` Robert Schiele
2002-08-14 9:53 ` Janis Johnson
1 sibling, 1 reply; 6+ messages in thread
From: Daniel Jacobowitz @ 2002-08-06 5:56 UTC (permalink / raw)
To: Robert Schiele; +Cc: gcc
On Tue, Aug 06, 2002 at 02:05:22PM +0200, Robert Schiele wrote:
> Hello,
>
> as some may have already noticed, I am sending the results of the
> regression tests after successful bootstrap processes to
> gcc-testresults from my automatic build system for some time now.
>
> I am currently adding a feature to this build system that allows
> sending automatic notification messages with some relevant information
> when bootstraping fails.
>
> Would sending such messages to gcc-testresults be welcome or would
> such messages be seen as annoying?
What are you bootstrapping on? There is already a regression tester
covering i686-pc-linux-gnu, which nags committers off-list (and the
gcc-regression list created for that purpose).
--
Daniel Jacobowitz Carnegie Mellon University
MontaVista Software Debian GNU/Linux Developer
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: automatic bootstrap failure notification
2002-08-06 5:56 ` Daniel Jacobowitz
@ 2002-08-06 6:20 ` Robert Schiele
0 siblings, 0 replies; 6+ messages in thread
From: Robert Schiele @ 2002-08-06 6:20 UTC (permalink / raw)
To: gcc
[-- Attachment #1: Type: text/plain, Size: 1807 bytes --]
On Tue, Aug 06, 2002 at 08:56:13AM -0400, Daniel Jacobowitz wrote:
> On Tue, Aug 06, 2002 at 02:05:22PM +0200, Robert Schiele wrote:
> > Hello,
> >
> > as some may have already noticed, I am sending the results of the
> > regression tests after successful bootstrap processes to
> > gcc-testresults from my automatic build system for some time now.
> >
> > I am currently adding a feature to this build system that allows
> > sending automatic notification messages with some relevant information
> > when bootstraping fails.
> >
> > Would sending such messages to gcc-testresults be welcome or would
> > such messages be seen as annoying?
>
> What are you bootstrapping on? There is already a regression tester
> covering i686-pc-linux-gnu, which nags committers off-list (and the
> gcc-regression list created for that purpose).
I am currently doing nightly builds of both gcc-3_2-branch and cvs
head for the following systems:
i486-suse-linux
sparc-suse-linux
sparc-sun-solaris2.8
sparc64-sun-solaris2.8
Although one combination (sparc64-sun-solaris2.8 on cvs head) does not
do regression tests, as dejagnu is currently not availlable on this
machine for some reasons.
The stuff I intended to send to the list was not a detailled analysis
of the regressions, but a message that the bootstrap process fails on
platform x-y-z with configure options --bar --foo, the last 100 lines
of output and a link to the full build results. Just something similar
I did manually in messages like
http://gcc.gnu.org/ml/gcc/2002-08/msg00191.html. If gcc@gcc.gnu.org or
some other list is more appropriate, I could also send those messages
there instead.
Robert
--
Robert Schiele Tel.: +49-621-181-2517
Dipl.-Wirtsch.informatiker mailto:rschiele@uni-mannheim.de
[-- Attachment #2: Type: application/pgp-signature, Size: 524 bytes --]
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: automatic bootstrap failure notification
2002-08-06 5:19 automatic bootstrap failure notification Robert Schiele
2002-08-06 5:56 ` Daniel Jacobowitz
@ 2002-08-14 9:53 ` Janis Johnson
2002-08-14 13:23 ` Robert Schiele
1 sibling, 1 reply; 6+ messages in thread
From: Janis Johnson @ 2002-08-14 9:53 UTC (permalink / raw)
To: Robert Schiele; +Cc: gcc
On Tue, Aug 06, 2002 at 02:05:22PM +0200, Robert Schiele wrote:
> Hello,
>
> as some may have already noticed, I am sending the results of the
> regression tests after successful bootstrap processes to
> gcc-testresults from my automatic build system for some time now.
>
> I am currently adding a feature to this build system that allows
> sending automatic notification messages with some relevant information
> when bootstraping fails.
>
> Would sending such messages to gcc-testresults be welcome or would
> such messages be seen as annoying?
I think this is a good idea. It would be useful if everyone who has
automated builds regularly sending test results to gcc-testresults
also sent mail there about bootstrap failures, particularly if they
used the same (or similar) scripts to generate those messages so they
had similar information.
Janis
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: automatic bootstrap failure notification
2002-08-14 9:53 ` Janis Johnson
@ 2002-08-14 13:23 ` Robert Schiele
2002-08-14 20:38 ` Robert Schiele
0 siblings, 1 reply; 6+ messages in thread
From: Robert Schiele @ 2002-08-14 13:23 UTC (permalink / raw)
To: Janis Johnson; +Cc: gcc
On Wed, Aug 14, 2002 at 09:55:08AM -0700, Janis Johnson wrote:
> On Tue, Aug 06, 2002 at 02:05:22PM +0200, Robert Schiele wrote:
> > Hello,
> >
> > as some may have already noticed, I am sending the results of the
> > regression tests after successful bootstrap processes to
> > gcc-testresults from my automatic build system for some time now.
> >
> > I am currently adding a feature to this build system that allows
> > sending automatic notification messages with some relevant information
> > when bootstraping fails.
> >
> > Would sending such messages to gcc-testresults be welcome or would
> > such messages be seen as annoying?
>
> I think this is a good idea. It would be useful if everyone who has
> automated builds regularly sending test results to gcc-testresults
> also sent mail there about bootstrap failures, particularly if they
> used the same (or similar) scripts to generate those messages so they
> had similar information.
Ok, I have simulated a bootstrap failure by killing make during
bootstrap. This made the system to send the message
http://gcc.gnu.org/ml/gcc-testresults/2002-08/msg00451.html to
gcc-testresults.
Feel free to send comments about the message contents.
To be complete I am attaching the script used for this purpose. But
be warned: This is currently just a fast and ugly bash hack. You may
do with the files what you like as long as it conforms with the GPL
and you remove my email address from the NOTIFY_FAILURE variable
_please_. But it might make sense to do a somewhat cleaner
implementation for general usage. But comments for the current script
are also welcome.
The file autobuild.common is the one that matters. This file should
never be called directly, but by a configuration script. I also
attached a sample of such a configuration script for reference. Up to
now I was too lazy to write real documentation. ;-)
Robert
--
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iQEVAwUBPVq8AsQAnns5HcHpAQHM7wgA0qnONIhcYNviK+A/8tQ2XaRE4TEdaAWs
HNQwimRF0YugVedBSUhHgvF6ni7gjbPYnc2CSqEt77dJb55EwFZP2wJlCRJXrbr6
RcjcuN7mqYm6LOxW3weixnpfbEnlwm9qWHeUaaLVu5b6v8VPq5HFzJVMh3dry3u4
OK2K21RiFeiRAfK87s4DTDh33t6PfdRJRgQ4g2FVp/RKUB4jBPuFvtkdJ1AMgmnE
4T4mEMh1bP3M4aVwa2+lsk1O7GFt0Or91zLgIHDoILo5M1SE+wmm+/Swt74fbXXX
92ZFeQzXW1ttBKbtIt32HQTbCPAfGoObBj1FEr2Z2LQEsGPiARAPDA==
=iJsK
-----END PGP SIGNATURE-----
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: automatic bootstrap failure notification
2002-08-14 13:23 ` Robert Schiele
@ 2002-08-14 20:38 ` Robert Schiele
0 siblings, 0 replies; 6+ messages in thread
From: Robert Schiele @ 2002-08-14 20:38 UTC (permalink / raw)
To: Janis Johnson; +Cc: gcc
Sure there was a stupid bug in the script. ;-)
But as fixing is as easy as stupidity of the bug, here the trivial
patch:
--- autobuild.common~ Wed Aug 14 22:10:44 2002
+++ autobuild.common Thu Aug 15 05:23:43 2002
@@ -15,7 +15,6 @@
LOGDIR_WEB= http://pi3.informatik.uni-mannheim.de/~schiele/gcclogs/$LOGDIR
LOGFILE_FILE=$LOGDIR_FILE/$TRIPLE$SUFFIX
LOGFILE_WEB=$LOGDIR_WEB/$TRIPLE$SUFFIX
-chmod 755 $LOGDIR_FILE
TRIPLE_CAN="`$SRCDIR/config.sub $TRIPLE`"
VERSIONSTR="`grep version_string $SRCDIR/gcc/version.c | sed -e 's/^.*"\(.*\)".*$/\1/'`"
cd $BUILDDIR || exit 1
@@ -25,6 +24,7 @@
--with-gxx-include-dir=$PREFIX/include/g++ --with-system-zlib \
--enable-shared $CXA_ATEXIT $TRIPLE"
mkdir -p $LOGDIR_FILE
+chmod 755 $LOGDIR_FILE
rm -f $LOGFILE_FILE*
if $PRENV $SRCDIR/configure $FLAGS_CONFIGURE > $LOGFILE_FILE.configure 2>&1 && \
$PRENV gmake bootstrap > $LOGFILE_FILE.make 2>&1; then
Robert
--
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iQEVAwUBPVsh08QAnns5HcHpAQFc/gf/c7vlSSQxner8KCRkPB0H7cXCgZl2YOgt
UD/H7zEpfdZhr00aXl2+bpi0tvfHy4a22fbtpj8xOBh5hpTM8P4608wmWdsqu52C
2e8fashEDYiuE1BcDVh6ENkPScCgsu6PWGJiSIRWRnWPdNhyRj9Ut1NgbZ01lnNX
90OmF2rCEY2lKGAk8KjkfXEmwVKkVgL+h3JZFvJq8H3axa2dOw4n7eAMvkXgI5pN
fzsdnm2Yrg9lXPzdIJDQtry+uUC9O489wBYL0RnerKrNfxm1pNxeaygDCsfr+rVp
AJr7G5k4xYSPmZKaFMX1IJJwjiWFxk4GwgEq0k/Sa8gs2jxf3ERp8w==
=1sA7
-----END PGP SIGNATURE-----
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2002-08-14 20:38 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2002-08-06 5:19 automatic bootstrap failure notification Robert Schiele
2002-08-06 5:56 ` Daniel Jacobowitz
2002-08-06 6:20 ` Robert Schiele
2002-08-14 9:53 ` Janis Johnson
2002-08-14 13:23 ` Robert Schiele
2002-08-14 20:38 ` Robert Schiele
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).