| Summary: | TPS SRPM Failures : JAVA_INCLUDE_PATH2 | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux 6 | Reporter: | Jenny Severance <jgalipea> |
| Component: | pki-core | Assignee: | Matthew Harmsen <mharmsen> |
| Status: | CLOSED WONTFIX | QA Contact: | IDM QE LIST <seceng-idm-qe-list> |
| Severity: | unspecified | Docs Contact: | |
| Priority: | high | ||
| Version: | 6.3 | CC: | ckannan, nkinder |
| Target Milestone: | rc | ||
| Target Release: | 6.3 | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2012-03-30 17:49:10 UTC | Type: | --- |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
|
Description
Jenny Severance
2012-03-23 13:41:46 UTC
'JAVA_INCLUDE_PATH2' is defined in the 'pki-core-9.0.3-24.el6.x86_64' source tarball under the following file:
* cmake/Modules/FindJNI.cmake
which is included by the top-level 'CMakeLists.txt' file:
* # where to look first for cmake modules
# (before ${CMAKE_ROOT}/Modules/ is checked)
set(CMAKE_MODULE_PATH ${CMAKE_SOURCE_DIR}/cmake/Modules)
# add definitions
include(DefineCMakeDefaults)
include(DefinePlatformDefaults)
include(DefineCompilerFlags)
include(DefineInstallationPaths)
include(DefineOptions.cmake)
include(CPackConfig.cmake)
# disallow in-source build
include(MacroEnsureOutOfSourceBuild)
macro_ensure_out_of_source_build("${PROJECT_NAME} requires an out of source build. Please create a separate build directory and run 'cmake /path/to/${PROJECT_NAME} [options]' there.")
# add macros
include(MacroCopyFile)
include(UseJava)
The "FindJNI.cmake" file contains the following definition:
* find_path(JAVA_INCLUDE_PATH2
NAMES
jni_md.h
PATHS
${JAVA_INCLUDE_PATH}
${JAVA_INCLUDE_PATH}/win32
${JAVA_INCLUDE_PATH}/linux
${JAVA_INCLUDE_PATH}/freebsd
${JAVA_INCLUDE_PATH}/solaris
)
For RHEL 6 platforms, the 'java-1.6.0-openjdk-devel' package contains the 'jni_md.h' file:
* /usr/lib/jvm/java-1.6.0-openjdk-1.6.0.0.x86_64/include/linux/jni_md.h
I successfully tested a rebuild on my system (non-mock):
* rpmbuild --rebuild --define "dist .el6" --define "_topdir `pwd`" ./SRPMS/pki-core-9.0.3-24.el6.src.rpm
The only possible explanation I can offer is that the FindJNI file is somehow not getting included for "TPS" tests. For example:
* find . -name CMakeLists.txt | xargs grep ^include\(
./CMakeLists.txt:include(DefineCMakeDefaults)
./CMakeLists.txt:include(DefinePlatformDefaults)
./CMakeLists.txt:include(DefineCompilerFlags)
./CMakeLists.txt:include(DefineInstallationPaths)
./CMakeLists.txt:include(DefineOptions.cmake)
./CMakeLists.txt:include(CPackConfig.cmake)
./CMakeLists.txt:include(MacroEnsureOutOfSourceBuild)
./CMakeLists.txt:include(MacroCopyFile)
./CMakeLists.txt:include(UseJava)
./CMakeLists.txt:include(ConfigureChecks.cmake)
I could find no place that specifically defined inclusion of "FindJNI.cmake", but this has never been a problem before, and does not seem to cause issues outside of TPS.
Attempted to replicate this issue in a 'mock' environment (using copies of RHEL 6 'mock' configuration files for i386 and x86_64) produced success for the following cases:
* rpm -q mock
mock-1.1.21-1.fc16.noarch
* mock -v -r rhel-6-i386 --rebuild ./pki-core-9.0.3-24.el6.src.rpm
* mock -v -r rhel-6-x86_64 --rebuild ./pki-core-9.0.3-24.el6.src.rpm
As I am unable to replicate this issue, may I suggest that TPS tests be re-run to see if the same error is encountered?
If it isn't, could you please consider waiving this TPS test?
There are two arches that are still failing with this error on SRPM rebuild testing. I would also suggest we waive the failures for the 6.3 errata, ship we ship and support built rpms, not rebuild rpms. Please advise ... Go ahead and waive it. Its pretty clear we can do builds on brew that we consider is a sane build env. So not seeing it as an issue. And it builds just fine on 6Server(the most important flavor). closing bug ... wont fix |