Re: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting
Looks very unlikely because of these 2 changes. These changes are not in PDE build. The best way to look at it with a bug that demonstrates the problem. Regards, Vikas Eclipse Modeling/ PDE IBM Rational EGL D Block - Bangalore, India Office Phone No : +91 - 80 - 41776506 From: David M Williams david_willi...@us.ibm.com To: Cross project issues cross-project-issues-dev@eclipse.org Date: 09/09/2014 10:45 AM Subject:Re: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting Sent by:cross-project-issues-dev-boun...@eclipse.org And, purely guessing here, there were some recent changes (last 2 weeks) in pde.ui that changed error reporting a bit ... and while I'm sure the fix was necessary, may have missed a case, or something? Just thought I'd point these out incase it helps you describe/reproduce the problem? Pure guess work on my part, just looking at what's changed. http://git.eclipse.org/c/pde/eclipse.pde.ui.git/log/?h=R4_4_maintenance Bug 435872 - NullPointerException when importing a project or opening MANIFEST.MF/plugin.xml Bug 401741 - exception when creating new extension point Thanks again, From:David M Williams/Raleigh/IBM@IBMUS To:Cross project issues cross-project-issues-dev@eclipse.org, Date:09/09/2014 01:03 AM Subject:Re: [cross-project-issues-dev] Apparent Luna SR1 regression inbuildfailure reporting Sent by:cross-project-issues-dev-boun...@eclipse.org I've not seen ... but, I've not looked either. I suggest you open a blocking defect on PDE Build, with preferably an as small as possible test case that shows the problem. I do not actually see any changes in PDE Build repo for SR1 ... and have heard there's been none for p2 ... so might be related to equinox changes or something. I'm only guessing that, since, given some of the recent changes and problems with Equinox, you might try RC1 and see if the problem exists there too? That'd be helpful to know. And, of course logs showing that a process failure occurs, but is simply not reported would also be helpful. Not to mention platform, Linux/bash, Windows/cygwin? But, a bug is the way to get the proper people looking at it. Thanks, From:Konstantin Komissarchik konstantin.komissarc...@oracle.com To:'Cross project issues' cross-project-issues-dev@eclipse.org, Date:09/08/2014 11:34 PM Subject:Re: [cross-project-issues-dev] Apparent Luna SR1 regression inbuildfailure reporting Sent by:cross-project-issues-dev-boun...@eclipse.org Anyone? We are down to the wire here and it’s looking like Luna SR1 will be useless as a build driver unless there is a last-minute fix, since we can’t depend on it failing the build correctly. - Konstantin From: cross-project-issues-dev-boun...@eclipse.org [ mailto:cross-project-issues-dev-boun...@eclipse.org] On Behalf Of Konstantin Komissarchik Sent: Monday, September 08, 2014 10:01 AM To: 'Cross project issues' Subject: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting We are seeing a serious regression in Luna SR1 (tested with platform RC3) where PDE build does not report a process failure code when the build fails. This causes any invoking build process (such as Ant) to not detect a failure and continue. Just confirmed that this works correctly if Luna is used to drive the build instead of Luna SR1. I am not sure if the regression is in the launcher or in the PDE build. Is anyone else seeing this? Thanks, - Konstantin ___ cross-project-issues-dev mailing list cross-project-issues-dev@eclipse.org To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev ___ cross-project-issues-dev mailing list cross-project-issues-dev@eclipse.org To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev ___ cross-project-issues-dev mailing list cross-project-issues-dev@eclipse.org To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev ___ cross-project-issues-dev mailing list cross-project-issues-dev@eclipse.org To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev
Re: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting
This sounds similar to https://bugs.eclipse.org/bugs/show_bug.cgi?id=436073 There have been some changes to Equinox for SR1 dealing with system properties. I am wondering somehow the system property osgi.framework.useSystemProperties=true is being set before invoking the framework from the build. Tom From: Konstantin Komissarchik konstantin.komissarc...@oracle.com To: 'Cross project issues' cross-project-issues-dev@eclipse.org Date: 09/08/2014 10:34 PM Subject:Re: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting Sent by:cross-project-issues-dev-boun...@eclipse.org Anyone? We are down to the wire here and it’s looking like Luna SR1 will be useless as a build driver unless there is a last-minute fix, since we can’t depend on it failing the build correctly. - Konstantin From: cross-project-issues-dev-boun...@eclipse.org [ mailto:cross-project-issues-dev-boun...@eclipse.org] On Behalf Of Konstantin Komissarchik Sent: Monday, September 08, 2014 10:01 AM To: 'Cross project issues' Subject: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting We are seeing a serious regression in Luna SR1 (tested with platform RC3) where PDE build does not report a process failure code when the build fails. This causes any invoking build process (such as Ant) to not detect a failure and continue. Just confirmed that this works correctly if Luna is used to drive the build instead of Luna SR1. I am not sure if the regression is in the launcher or in the PDE build. Is anyone else seeing this? Thanks, - Konstantin ___ cross-project-issues-dev mailing list cross-project-issues-dev@eclipse.org To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev ___ cross-project-issues-dev mailing list cross-project-issues-dev@eclipse.org To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev
Re: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting
Konstantin I put a fix into master for the 4.5 Eclipse I-Build that should happen today (we need to respin for another issue). Would you be able to test with this I-Build once it is done to see if it fixes the issue? Then we can consider backporting to Luna. Tom From: Thomas Watson/Austin/IBM@IBMUS To: Cross project issues cross-project-issues-dev@eclipse.org Date: 09/09/2014 07:35 AM Subject:Re: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting Sent by:cross-project-issues-dev-boun...@eclipse.org This sounds similar to https://bugs.eclipse.org/bugs/show_bug.cgi?id=436073 There have been some changes to Equinox for SR1 dealing with system properties. I am wondering somehow the system property osgi.framework.useSystemProperties=true is being set before invoking the framework from the build. Tom From:Konstantin Komissarchik konstantin.komissarc...@oracle.com To:'Cross project issues' cross-project-issues-dev@eclipse.org Date:09/08/2014 10:34 PM Subject:Re: [cross-project-issues-dev] Apparent Luna SR1 regression inbuildfailure reporting Sent by:cross-project-issues-dev-boun...@eclipse.org Anyone? We are down to the wire here and it’s looking like Luna SR1 will be useless as a build driver unless there is a last-minute fix, since we can’t depend on it failing the build correctly. - Konstantin From: cross-project-issues-dev-boun...@eclipse.org [ mailto:cross-project-issues-dev-boun...@eclipse.org] On Behalf Of Konstantin Komissarchik Sent: Monday, September 08, 2014 10:01 AM To: 'Cross project issues' Subject: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting We are seeing a serious regression in Luna SR1 (tested with platform RC3) where PDE build does not report a process failure code when the build fails. This causes any invoking build process (such as Ant) to not detect a failure and continue. Just confirmed that this works correctly if Luna is used to drive the build instead of Luna SR1. I am not sure if the regression is in the launcher or in the PDE build. Is anyone else seeing this? Thanks, - Konstantin ___ cross-project-issues-dev mailing list cross-project-issues-dev@eclipse.org To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev ___ cross-project-issues-dev mailing list cross-project-issues-dev@eclipse.org To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev ___ cross-project-issues-dev mailing list cross-project-issues-dev@eclipse.org To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev
Re: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting
Here is how we are calling into PDE Build. I don’t see any references to
osgi.framework.useSystemProperties.
java classname=org.eclipse.core.launcher.Main
fork=true failonerror=true maxmemory=${java.maxmemory}
classpath
fileset dir=${bootstrap.platform}/plugins
include name=**/org.eclipse.equinox.launcher_*.jar/
/fileset
/classpath
arg line=-clean/
arg line=-application org.eclipse.ant.core.antRunner/
arg line=-data ${build.dir}/pde/workspace/
arg line=-configuration ${build.dir}/pde/configuration/
arg line=-buildfile ${.pdeBuildDir}/scripts/build.xml/
arg value=-DtopLevelElementId=@{feature.id}/
arg value=-DarchivePrefix=eclipse/
arg value=-DbaseLocation=@{eclipse.install.dir}/
arg value=-DbuildDirectory=@{root.dir}/
arg value=-Dbuilder=${build.dir}/pde/builder/
arg value=-DcollectingFolder=collecting/
arg value=-DbuildId=@{build.id}/
arg value=-DbuildType=I/
arg value=-DbuildLabel=build/
arg value=-DROOT_RELENG_DIR=@{root.releng.dir}/
arg value=-DforceContextQualifier=@{build.id}/
arg value=-DgenerateFeatureVersionSuffix=false/
arg value=-DindividualSourceBundles=true/
arg value=-DallowBinaryCycles=true /
arg value=-DcompilerArg=${pde.build.compiler.arg}/
arg value=-DJ2SE-1.4=${java.14.system.classpath}/
arg value=-DJavaSE-1.6=${java.16.system.classpath}/
arg value=-DJavaSE-1.7=${java.17.system.classpath}/
java-args/
/java
I will work on a stand-alone repro today.
- Konstantin
From: cross-project-issues-dev-boun...@eclipse.org
[mailto:cross-project-issues-dev-boun...@eclipse.org] On Behalf Of Thomas Watson
Sent: Tuesday, September 09, 2014 6:41 AM
To: Cross project issues
Subject: Re: [cross-project-issues-dev] Apparent Luna SR1 regression in build
failure reporting
Konstantin
I put a fix into master for the 4.5 Eclipse I-Build that should happen today
(we need to respin for another issue). Would you be able to test with this
I-Build once it is done to see if it fixes the issue? Then we can consider
backporting to Luna.
Tom
From:Thomas Watson/Austin/IBM@IBMUS
To:Cross project issues cross-project-issues-dev@eclipse.org
Date:09/09/2014 07:35 AM
Subject:Re: [cross-project-issues-dev] Apparent Luna SR1 regression
inbuildfailure reporting
Sent by:cross-project-issues-dev-boun...@eclipse.org
_
This sounds similar to https://bugs.eclipse.org/bugs/show_bug.cgi?id=436073
https://bugs.eclipse.org/bugs/show_bug.cgi?id=436073
There have been some changes to Equinox for SR1 dealing with system properties.
I am wondering somehow the system property
osgi.framework.useSystemProperties=true is being set before invoking the
framework from the build.
Tom
From:Konstantin Komissarchik konstantin.komissarc...@oracle.com
To:'Cross project issues' cross-project-issues-dev@eclipse.org
Date:09/08/2014 10:34 PM
Subject:Re: [cross-project-issues-dev] Apparent Luna SR1 regression in
buildfailure reporting
Sent by:cross-project-issues-dev-boun...@eclipse.org
_
Anyone? We are down to the wire here and it’s looking like Luna SR1 will be
useless as a build driver unless there is a last-minute fix, since we can’t
depend on it failing the build correctly.
- Konstantin
From: cross-project-issues-dev-boun...@eclipse.org [
mailto:cross-project-issues-dev-boun...@eclipse.org
mailto:cross-project-issues-dev-boun...@eclipse.org] On Behalf Of Konstantin
Komissarchik
Sent: Monday, September 08, 2014 10:01 AM
To: 'Cross project issues'
Subject: [cross-project-issues-dev] Apparent Luna SR1 regression in build
failure reporting
We are seeing a serious regression in Luna SR1 (tested with platform RC3) where
PDE build does not report a process failure code when the build fails. This
causes any invoking build process (such as Ant) to not detect a failure and
continue.
Just confirmed that this works correctly if Luna is used to drive the build
instead of Luna SR1.
I am not sure if the regression is in the launcher or in the PDE build.
Is anyone else seeing this?
Thanks,
- Konstantin
___
cross-project-issues-dev mailing list
cross-project-issues-dev@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from
this list, visit
https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev
https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev
___
cross-project-issues-dev mailing list
cross-project-issues-dev@eclipse.org
To change your delivery options, retrieve your password
Re: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting
OK, please track the issue and post the repro with bug
https://bugs.eclipse.org/bugs/show_bug.cgi?id=443595
Tom
From: Konstantin Komissarchik konstantin.komissarc...@oracle.com
To: 'Cross project issues' cross-project-issues-dev@eclipse.org
Date: 09/09/2014 08:57 AM
Subject:Re: [cross-project-issues-dev] Apparent LunaSR1
regression in build failure reporting
Sent by:cross-project-issues-dev-boun...@eclipse.org
Here is how we are calling into PDE Build. I don’t see any references to
osgi.framework.useSystemProperties.
java classname=org.eclipse.core.launcher.Main
fork=true failonerror=true maxmemory=${java.maxmemory}
classpath
fileset dir=${bootstrap.platform}/plugins
include name=**/org.eclipse.equinox.launcher_*.jar/
/fileset
/classpath
arg line=-clean/
arg line=-application org.eclipse.ant.core.antRunner/
arg line=-data ${build.dir}/pde/workspace/
arg line=-configuration ${build.dir}/pde/configuration/
arg line=-buildfile ${.pdeBuildDir}/scripts/build.xml/
arg value=-DtopLevelElementId=@{feature.id}/
arg value=-DarchivePrefix=eclipse/
arg value=-DbaseLocation=@{eclipse.install.dir}/
arg value=-DbuildDirectory=@{root.dir}/
arg value=-Dbuilder=${build.dir}/pde/builder/
arg value=-DcollectingFolder=collecting/
arg value=-DbuildId=@{build.id}/
arg value=-DbuildType=I/
arg value=-DbuildLabel=build/
arg value=-DROOT_RELENG_DIR=@{root.releng.dir}/
arg value=-DforceContextQualifier=@{build.id}/
arg value=-DgenerateFeatureVersionSuffix=false/
arg value=-DindividualSourceBundles=true/
arg value=-DallowBinaryCycles=true /
arg value=-DcompilerArg=${pde.build.compiler.arg}/
arg value=-DJ2SE-1.4=${java.14.system.classpath}/
arg value=-DJavaSE-1.6=${java.16.system.classpath}/
arg value=-DJavaSE-1.7=${java.17.system.classpath}/
java-args/
/java
I will work on a stand-alone repro today.
- Konstantin
From: cross-project-issues-dev-boun...@eclipse.org [
mailto:cross-project-issues-dev-boun...@eclipse.org] On Behalf Of Thomas
Watson
Sent: Tuesday, September 09, 2014 6:41 AM
To: Cross project issues
Subject: Re: [cross-project-issues-dev] Apparent Luna SR1 regression in
build failure reporting
Konstantin
I put a fix into master for the 4.5 Eclipse I-Build that should happen
today (we need to respin for another issue). Would you be able to test
with this I-Build once it is done to see if it fixes the issue? Then we
can consider backporting to Luna.
Tom
From:Thomas Watson/Austin/IBM@IBMUS
To:Cross project issues cross-project-issues-dev@eclipse.org
Date:09/09/2014 07:35 AM
Subject:Re: [cross-project-issues-dev] Apparent Luna SR1
regressioninbuildfailure reporting
Sent by:cross-project-issues-dev-boun...@eclipse.org
This sounds similar to
https://bugs.eclipse.org/bugs/show_bug.cgi?id=436073
There have been some changes to Equinox for SR1 dealing with system
properties. I am wondering somehow the system property
osgi.framework.useSystemProperties=true is being set before invoking the
framework from the build.
Tom
From:Konstantin Komissarchik konstantin.komissarc...@oracle.com
To:'Cross project issues' cross-project-issues-dev@eclipse.org
Date:09/08/2014 10:34 PM
Subject:Re: [cross-project-issues-dev] Apparent Luna SR1
regression inbuildfailure reporting
Sent by:cross-project-issues-dev-boun...@eclipse.org
Anyone? We are down to the wire here and it’s looking like Luna SR1 will
be useless as a build driver unless there is a last-minute fix, since we
can’t depend on it failing the build correctly.
- Konstantin
From: cross-project-issues-dev-boun...@eclipse.org [
mailto:cross-project-issues-dev-boun...@eclipse.org] On Behalf Of
Konstantin Komissarchik
Sent: Monday, September 08, 2014 10:01 AM
To: 'Cross project issues'
Subject: [cross-project-issues-dev] Apparent Luna SR1 regression in build
failure reporting
We are seeing a serious regression in Luna SR1 (tested with platform RC3)
where PDE build does not report a process failure code when the build
fails. This causes any invoking build process (such as Ant) to not detect
a failure and continue.
Just confirmed that this works correctly if Luna is used to drive the
build instead of Luna SR1.
I am not sure if the regression is in the launcher or in the PDE build.
Is anyone else seeing this?
Thanks,
- Konstantin
___
cross-project-issues-dev mailing list
cross-project-issues-dev@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe
from this list, visit
https://dev.eclipse.org/mailman
Re: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting
I opened a separate bug, in case this is a different issue. Assigned to
Equinox for now. Simple repro attached.
https://bugs.eclipse.org/bugs/show_bug.cgi?id=443614
Thanks,
- Konstantin
From: cross-project-issues-dev-boun...@eclipse.org
[mailto:cross-project-issues-dev-boun...@eclipse.org] On Behalf Of Thomas Watson
Sent: Tuesday, September 09, 2014 7:23 AM
To: Cross project issues
Subject: Re: [cross-project-issues-dev] Apparent Luna SR1 regression in build
failure reporting
OK, please track the issue and post the repro with bug
https://bugs.eclipse.org/bugs/show_bug.cgi?id=443595
https://bugs.eclipse.org/bugs/show_bug.cgi?id=443595
Tom
From:Konstantin Komissarchik konstantin.komissarc...@oracle.com
To:'Cross project issues' cross-project-issues-dev@eclipse.org
Date:09/09/2014 08:57 AM
Subject:Re: [cross-project-issues-dev] Apparent LunaSR1
regressioninbuildfailure reporting
Sent by:cross-project-issues-dev-boun...@eclipse.org
_
Here is how we are calling into PDE Build. I don’t see any references to
osgi.framework.useSystemProperties.
java classname=org.eclipse.core.launcher.Main
fork=true failonerror=true maxmemory=${java.maxmemory}
classpath
fileset dir=${bootstrap.platform}/plugins
include name=**/org.eclipse.equinox.launcher_*.jar/
/fileset
/classpath
arg line=-clean/
arg line=-application org.eclipse.ant.core.antRunner/
arg line=-data ${build.dir}/pde/workspace/
arg line=-configuration ${build.dir}/pde/configuration/
arg line=-buildfile ${.pdeBuildDir}/scripts/build.xml/
arg value=-DtopLevelElementId=@{feature.id}
mailto:-DtopLevelElementId=@%7bfeature.id%7d /
arg value=-DarchivePrefix=eclipse/
arg value=-DbaseLocation=@{eclipse.install.dir}
mailto:-DbaseLocation=@%7beclipse.install.dir%7d /
arg value=-DbuildDirectory=@{root.dir}
mailto:-DbuildDirectory=@%7broot.dir%7d /
arg value=-Dbuilder=${build.dir}/pde/builder/
arg value=-DcollectingFolder=collecting/
arg value=-DbuildId=@{build.id} mailto:-DbuildId=@%7bbuild.id%7d
/
arg value=-DbuildType=I/
arg value=-DbuildLabel=build/
arg value=-DROOT_RELENG_DIR=@{root.releng.dir}
mailto:-DROOT_RELENG_DIR=@%7broot.releng.dir%7d /
arg value=-DforceContextQualifier=@{build.id}
mailto:-DforceContextQualifier=@%7bbuild.id%7d /
arg value=-DgenerateFeatureVersionSuffix=false/
arg value=-DindividualSourceBundles=true/
arg value=-DallowBinaryCycles=true /
arg value=-DcompilerArg=${pde.build.compiler.arg}/
arg value=-DJ2SE-1.4=${java.14.system.classpath}/
arg value=-DJavaSE-1.6=${java.16.system.classpath}/
arg value=-DJavaSE-1.7=${java.17.system.classpath}/
java-args/
/java
I will work on a stand-alone repro today.
- Konstantin
From: cross-project-issues-dev-boun...@eclipse.org [
mailto:cross-project-issues-dev-boun...@eclipse.org
mailto:cross-project-issues-dev-boun...@eclipse.org] On Behalf Of Thomas Watson
Sent: Tuesday, September 09, 2014 6:41 AM
To: Cross project issues
Subject: Re: [cross-project-issues-dev] Apparent Luna SR1 regression in build
failure reporting
Konstantin
I put a fix into master for the 4.5 Eclipse I-Build that should happen today
(we need to respin for another issue). Would you be able to test with this
I-Build once it is done to see if it fixes the issue? Then we can consider
backporting to Luna.
Tom
From:Thomas Watson/Austin/IBM@IBMUS
To:Cross project issues mailto:cross-project-issues-dev@eclipse.org
cross-project-issues-dev@eclipse.org
Date:09/09/2014 07:35 AM
Subject:Re: [cross-project-issues-dev] Apparent Luna SR1 regression
inbuildfailure reporting
Sent by: mailto:cross-project-issues-dev-boun...@eclipse.org
cross-project-issues-dev-boun...@eclipse.org
_
This sounds similar to https://bugs.eclipse.org/bugs/show_bug.cgi?id=436073
https://bugs.eclipse.org/bugs/show_bug.cgi?id=436073
There have been some changes to Equinox for SR1 dealing with system properties.
I am wondering somehow the system property
osgi.framework.useSystemProperties=true is being set before invoking the
framework from the build.
Tom
From:Konstantin Komissarchik
mailto:konstantin.komissarc...@oracle.com konstantin.komissarc...@oracle.com
To:'Cross project issues'
mailto:cross-project-issues-dev@eclipse.org
cross-project-issues-dev@eclipse.org
Date:09/08/2014 10:34 PM
Subject:Re: [cross-project-issues-dev] Apparent Luna SR1 regression in
buildfailure reporting
Sent by: mailto:cross-project-issues-dev-boun...@eclipse.org
cross-project
[cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting
We are seeing a serious regression in Luna SR1 (tested with platform RC3) where PDE build does not report a process failure code when the build fails. This causes any invoking build process (such as Ant) to not detect a failure and continue. Just confirmed that this works correctly if Luna is used to drive the build instead of Luna SR1. I am not sure if the regression is in the launcher or in the PDE build. Is anyone else seeing this? Thanks, - Konstantin ___ cross-project-issues-dev mailing list cross-project-issues-dev@eclipse.org To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev
Re: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting
To add one more bit of info that might help. When driving the build with Luna, I see An error has occurred. See the log file in the console output. When driving the build with Luna SR1, this message is missing. - Konstantin From: cross-project-issues-dev-boun...@eclipse.org [mailto:cross-project-issues-dev-boun...@eclipse.org] On Behalf Of Konstantin Komissarchik Sent: Monday, September 08, 2014 10:01 AM To: 'Cross project issues' Subject: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting We are seeing a serious regression in Luna SR1 (tested with platform RC3) where PDE build does not report a process failure code when the build fails. This causes any invoking build process (such as Ant) to not detect a failure and continue. Just confirmed that this works correctly if Luna is used to drive the build instead of Luna SR1. I am not sure if the regression is in the launcher or in the PDE build. Is anyone else seeing this? Thanks, - Konstantin ___ cross-project-issues-dev mailing list cross-project-issues-dev@eclipse.org To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev
Re: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting
Anyone? We are down to the wire here and it's looking like Luna SR1 will be useless as a build driver unless there is a last-minute fix, since we can't depend on it failing the build correctly. - Konstantin From: cross-project-issues-dev-boun...@eclipse.org [mailto:cross-project-issues-dev-boun...@eclipse.org] On Behalf Of Konstantin Komissarchik Sent: Monday, September 08, 2014 10:01 AM To: 'Cross project issues' Subject: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting We are seeing a serious regression in Luna SR1 (tested with platform RC3) where PDE build does not report a process failure code when the build fails. This causes any invoking build process (such as Ant) to not detect a failure and continue. Just confirmed that this works correctly if Luna is used to drive the build instead of Luna SR1. I am not sure if the regression is in the launcher or in the PDE build. Is anyone else seeing this? Thanks, - Konstantin ___ cross-project-issues-dev mailing list cross-project-issues-dev@eclipse.org To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev
Re: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting
I've not seen ... but, I've not looked either. I suggest you open a blocking defect on PDE Build, with preferably an as small as possible test case that shows the problem. I do not actually see any changes in PDE Build repo for SR1 ... and have heard there's been none for p2 ... so might be related to equinox changes or something. I'm only guessing that, since, given some of the recent changes and problems with Equinox, you might try RC1 and see if the problem exists there too? That'd be helpful to know. And, of course logs showing that a process failure occurs, but is simply not reported would also be helpful. Not to mention platform, Linux/bash, Windows/cygwin? But, a bug is the way to get the proper people looking at it. Thanks, From: Konstantin Komissarchik konstantin.komissarc...@oracle.com To: 'Cross project issues' cross-project-issues-dev@eclipse.org, Date: 09/08/2014 11:34 PM Subject:Re: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting Sent by:cross-project-issues-dev-boun...@eclipse.org Anyone? We are down to the wire here and it’s looking like Luna SR1 will be useless as a build driver unless there is a last-minute fix, since we can’t depend on it failing the build correctly. - Konstantin From: cross-project-issues-dev-boun...@eclipse.org [ mailto:cross-project-issues-dev-boun...@eclipse.org] On Behalf Of Konstantin Komissarchik Sent: Monday, September 08, 2014 10:01 AM To: 'Cross project issues' Subject: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting We are seeing a serious regression in Luna SR1 (tested with platform RC3) where PDE build does not report a process failure code when the build fails. This causes any invoking build process (such as Ant) to not detect a failure and continue. Just confirmed that this works correctly if Luna is used to drive the build instead of Luna SR1. I am not sure if the regression is in the launcher or in the PDE build. Is anyone else seeing this? Thanks, - Konstantin ___ cross-project-issues-dev mailing list cross-project-issues-dev@eclipse.org To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev ___ cross-project-issues-dev mailing list cross-project-issues-dev@eclipse.org To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev
Re: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting
And, purely guessing here, there were some recent changes (last 2 weeks) in pde.ui that changed error reporting a bit ... and while I'm sure the fix was necessary, may have missed a case, or something? Just thought I'd point these out incase it helps you describe/reproduce the problem? Pure guess work on my part, just looking at what's changed. http://git.eclipse.org/c/pde/eclipse.pde.ui.git/log/?h=R4_4_maintenance Bug 435872 - NullPointerException when importing a project or opening MANIFEST.MF/plugin.xml Bug 401741 - exception when creating new extension point Thanks again, From: David M Williams/Raleigh/IBM@IBMUS To: Cross project issues cross-project-issues-dev@eclipse.org, Date: 09/09/2014 01:03 AM Subject:Re: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting Sent by:cross-project-issues-dev-boun...@eclipse.org I've not seen ... but, I've not looked either. I suggest you open a blocking defect on PDE Build, with preferably an as small as possible test case that shows the problem. I do not actually see any changes in PDE Build repo for SR1 ... and have heard there's been none for p2 ... so might be related to equinox changes or something. I'm only guessing that, since, given some of the recent changes and problems with Equinox, you might try RC1 and see if the problem exists there too? That'd be helpful to know. And, of course logs showing that a process failure occurs, but is simply not reported would also be helpful. Not to mention platform, Linux/bash, Windows/cygwin? But, a bug is the way to get the proper people looking at it. Thanks, From:Konstantin Komissarchik konstantin.komissarc...@oracle.com To:'Cross project issues' cross-project-issues-dev@eclipse.org, Date:09/08/2014 11:34 PM Subject:Re: [cross-project-issues-dev] Apparent Luna SR1 regression inbuildfailure reporting Sent by:cross-project-issues-dev-boun...@eclipse.org Anyone? We are down to the wire here and it’s looking like Luna SR1 will be useless as a build driver unless there is a last-minute fix, since we can’t depend on it failing the build correctly. - Konstantin From: cross-project-issues-dev-boun...@eclipse.org [ mailto:cross-project-issues-dev-boun...@eclipse.org] On Behalf Of Konstantin Komissarchik Sent: Monday, September 08, 2014 10:01 AM To: 'Cross project issues' Subject: [cross-project-issues-dev] Apparent Luna SR1 regression in build failure reporting We are seeing a serious regression in Luna SR1 (tested with platform RC3) where PDE build does not report a process failure code when the build fails. This causes any invoking build process (such as Ant) to not detect a failure and continue. Just confirmed that this works correctly if Luna is used to drive the build instead of Luna SR1. I am not sure if the regression is in the launcher or in the PDE build. Is anyone else seeing this? Thanks, - Konstantin ___ cross-project-issues-dev mailing list cross-project-issues-dev@eclipse.org To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev ___ cross-project-issues-dev mailing list cross-project-issues-dev@eclipse.org To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev ___ cross-project-issues-dev mailing list cross-project-issues-dev@eclipse.org To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev
