Red Hat Bugzilla – Bug 643359
[abrt] make-1:3.82-2.fc14.1: raise: Process /usr/bin/make was killed by signal 6 (SIGABRT)
Last modified: 2015-05-04 21:35:41 EDT
abrt version: 1.1.13
Attached file: backtrace
reason: Process /usr/bin/make was killed by signal 6 (SIGABRT)
release: Fedora release 14 (Laughlin)
# rpm -qa gcc* make glibc*
How to reproduce
1. Get the source of android:
$ curl http://android.git.kernel.org/repo >~/bin/repo
$ chmod a+x ~/bin/repo
$ mkdir android.repo
$ cd android.repo
$ ~/bin/repo init -u git://android.git.kernel.org/platform/manifest.git
$ ~/bin/repo sync
2. Try to build the source:
3. make crashes
Created attachment 453713 [details]
OS Release: Fedora release 14 (Laughlin)
How to reproduce
1. Download android-x86 source
2. Build it.
Looks like this happens building any Android source tree. I'll have to back up to Fedora 13 to continue. The severity is higher than medium.
Proposing as F14Blocker
Does make crash when building *only* android source tree ... or when building *any* project? If this only happens when building android, I don't believe we have any release criteria that would be impacted by this failure. However, if this impacts running make in all cases, we may have a valid issue. Another consideration is whether the fix could be delivered as a day-0 update. In the case of 'make', I believe that is a valid option.
With my understanding of the impact of this bug report, I don't believe this issue qualifies as a Fedora 14 release blocker.
It's only when building android source tree (yet).
But it's possibly on more projects and the new release, but make is a really important package, with "Has a severity rating of high or greater and no reasonable workaround" (from your link abovee).
This is definetely annoying and people might not use fc14, just because they can't build some projects anymore (see #comment3).
If this is not a final blocker, at least it's nice-to-have...
I believe this is the only crashing bug that I've seen for make 3.82.
Is there a package providing javah in Fedora, or does one need Oracle's java to build the thing? BTW, cutting a deterministically-failing test case out of the monster that is android would be very welcome.
(In reply to comment #8)
> Is there a package providing javah in Fedora, or does one need Oracle's java to
> build the thing? BTW, cutting a deterministically-failing test case out of the
> monster that is android would be very welcome.
I was building with OpenJDK. Same scenario works on Fedora 13 which I'm using now.
This bug was reviewed at the 2010-10-25 QA meeting (in lieu of a blocker review meeting). Everyone present agreed it looks like it's clearly not a blocker: it doesn't meet any of our criteria.
(We didn't review it for NTH status, but I wouldn't reckon it's NTH either, as it can be fully and painlessly fixed with an update).
Fedora Bugzappers volunteer triage team
Got it reproduced. It turns out I mistakenly installed make's src rpm instead of regular rpm, and was waiting for the fail to turn up sometime during the build. Turns up it appears right at the beginning.
Created attachment 455892 [details]
This reproduces the problem for me.
Created attachment 455896 [details]
This fixes the reproducer. I'll re-try the step-by step above to see if it also fixes the original problem.
... and it seems it does. I'll push the fix.
make-3.82-3.fc14 has been submitted as an update for Fedora 14.
make-3.82-3.fc14 has been pushed to the Fedora 14 testing repository. If problems still persist, please make note of it in this bug report.
If you want to test the update, you can install it with
su -c 'yum --enablerepo=updates-testing update make'. You can provide feedback for this update here: https://admin.fedoraproject.org/updates/make-3.82-3.fc14
make-3.82-3.fc14 has been pushed to the Fedora 14 stable repository. If problems still persist, please make note of it in this bug report.