9:52 a.m.
Hello,
I have a package that builds fine in a fresh rawhide mock chroot and not in
a koji build. Is there any difference between the two?
Here is the error in Koji:
+ make -j5 LIBPATH=/usr/lib64 -f makefile docs
rm -f doc/crypt.pdf *.dvi *.log *.aux *.toc *.idx *.ilg *.ind *.out
echo "hello" > crypt.ind
latex crypt > /dev/null
kpathsea: Running mktexfmt latex.fmt
make: *** [docs] Error 1
RPM build errors:
error: Bad exit status from /var/tmp/rpm-tmp.5TWg4P (%build)
Bad exit status from /var/tmp/rpm-tmp.5TWg4P (%build)
And here is the same step in mock:
DEBUG: + make -j4 LIBPATH=/usr/lib64 -f makefile docs
DEBUG: rm -f doc/crypt.pdf *.dvi *.log *.aux *.toc *.idx *.ilg *.ind *.out
DEBUG: echo "hello" > crypt.ind
DEBUG: latex crypt > /dev/null
DEBUG: latex crypt > /dev/null
DEBUG: makeindex crypt.idx > /dev/null
DEBUG: This is makeindex, version 2.15 [TeX Live 2013] (kpathsea + Thai
support).
DEBUG: Scanning input file crypt.idx....done (345 entries accepted, 0
rejected).
DEBUG: Sorting entries......done (3023 comparisons).
DEBUG: Generating output file crypt.ind....done (396 lines written, 0
warnings).
DEBUG: Output written in crypt.ind.
[...]
Log of the failed koji build:
http://kojipkgs.fedoraproject.org//work/tasks/4779/5464779/build.log
Command used for mock:
mock -v --rebuild -r fedora-rawhide-x86_64 libtomcrypt-1.17-16.fc19.src.rpm
Thanks & regards,
--Simone
--
You cannot discover new oceans unless you have the courage to lose sight of
the shore (R. W. Emerson).
http://xkcd.com/229/
10:05 a.m.
On Tue, 4 Jun 2013 16:52:14 +0200
Simone Caronni <negativo17(a)gmail.com> wrote:
Hello,
I have a package that builds fine in a fresh rawhide mock chroot and
not in a koji build. Is there any difference between the two?
Here is the error in Koji:
+ make -j5 LIBPATH=/usr/lib64 -f makefile docs
rm -f doc/crypt.pdf *.dvi *.log *.aux *.toc *.idx *.ilg *.ind *.out
echo "hello" > crypt.ind
latex crypt > /dev/null
kpathsea: Running mktexfmt latex.fmt
make: *** [docs] Error 1
RPM build errors:
error: Bad exit status from /var/tmp/rpm-tmp.5TWg4P (%build)
Bad exit status from /var/tmp/rpm-tmp.5TWg4P (%build)
And here is the same step in mock:
DEBUG: + make -j4 LIBPATH=/usr/lib64 -f makefile docs
DEBUG: rm -f doc/crypt.pdf *.dvi *.log *.aux *.toc *.idx *.ilg *.ind
*.out DEBUG: echo "hello" > crypt.ind
DEBUG: latex crypt > /dev/null
DEBUG: latex crypt > /dev/null
DEBUG: makeindex crypt.idx > /dev/null
DEBUG: This is makeindex, version 2.15 [TeX Live 2013] (kpathsea +
Thai support).
DEBUG: Scanning input file crypt.idx....done (345 entries accepted, 0
rejected).
DEBUG: Sorting entries......done (3023 comparisons).
DEBUG: Generating output file crypt.ind....done (396 lines written, 0
warnings).
DEBUG: Output written in crypt.ind.
[...]
Log of the failed koji build:
http://kojipkgs.fedoraproject.org//work/tasks/4779/5464779/build.log
Command used for mock:
mock -v --rebuild -r fedora-rawhide-x86_64
libtomcrypt-1.17-16.fc19.src.rpm
Koji uses mock to build pkgs.
Mock in Koji is run on systems which are isolated from the network.
If there is no network connection then there shouldn't be any difference
-sv
11:04 a.m.
On Tue, 2013-06-04 at 11:05 -0400, seth vidal wrote:
On Tue, 4 Jun 2013 16:52:14 +0200
Simone Caronni <negativo17(a)gmail.com> wrote:
> Hello,
>
> I have a package that builds fine in a fresh rawhide mock chroot and
> not in a koji build. Is there any difference between the two?
>
> Here is the error in Koji:
>
> + make -j5 LIBPATH=/usr/lib64 -f makefile docs
> rm -f doc/crypt.pdf *.dvi *.log *.aux *.toc *.idx *.ilg *.ind *.out
> echo "hello" > crypt.ind
> latex crypt > /dev/null
> kpathsea: Running mktexfmt latex.fmt
> make: *** [docs] Error 1
> RPM build errors:
> error: Bad exit status from /var/tmp/rpm-tmp.5TWg4P (%build)
> Bad exit status from /var/tmp/rpm-tmp.5TWg4P (%build)
>
> And here is the same step in mock:
>
> DEBUG: + make -j4 LIBPATH=/usr/lib64 -f makefile docs
> DEBUG: rm -f doc/crypt.pdf *.dvi *.log *.aux *.toc *.idx *.ilg *.ind
> *.out DEBUG: echo "hello" > crypt.ind
> DEBUG: latex crypt > /dev/null
> DEBUG: latex crypt > /dev/null
> DEBUG: makeindex crypt.idx > /dev/null
> DEBUG: This is makeindex, version 2.15 [TeX Live 2013] (kpathsea +
> Thai support).
> DEBUG: Scanning input file crypt.idx....done (345 entries accepted, 0
> rejected).
> DEBUG: Sorting entries......done (3023 comparisons).
> DEBUG: Generating output file crypt.ind....done (396 lines written, 0
> warnings).
> DEBUG: Output written in crypt.ind.
> [...]
>
> Log of the failed koji build:
>
> http://kojipkgs.fedoraproject.org//work/tasks/4779/5464779/build.log
>
> Command used for mock:
>
> mock -v --rebuild -r fedora-rawhide-x86_64
> libtomcrypt-1.17-16.fc19.src.rpm
>
Koji uses mock to build pkgs.
Mock in Koji is run on systems which are isolated from the network.
If there is no network connection then there shouldn't be any difference
I note 'make -j5' vs 'make -j4': Simone has more parallelization
capability than mock, it seems. Simone, you could fiddle with your spec
file so it doesn't do a parallel build at all, and see if the build
starts working then. If so, report to upstream that their build process
is not entirely safe for parallelization.
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Twitter: AdamW_Fedora | identi.ca: adamwfedora
http://www.happyassassin.net
1:52 a.m.
On 4 June 2013 18:04, Adam Williamson <awilliam(a)redhat.com> wrote:
I note 'make -j5' vs 'make -j4': Simone has more
parallelization
capability than mock, it seems. Simone, you could fiddle with your spec
file so it doesn't do a parallel build at all, and see if the build
starts working then. If so, report to upstream that their build process
is not entirely safe for parallelization.
Well, apparently the problem was related to some misaligning with rawhide
mirrors and the koji instance.
The problem appeared also on Mock builds on other machines.
Regards,
--Simone
--
You cannot discover new oceans unless you have the courage to lose sight of
the shore (R. W. Emerson).
http://xkcd.com/229/
2:33 p.m.
On Tue, Jun 04, 2013 at 04:52:14PM +0200, Simone Caronni wrote:
Hello,
I have a package that builds fine in a fresh rawhide mock chroot and not in
a koji build. Is there any difference between the two?
The other thing they don't tell you ... and it's an endless source of
"fun" ... is that Koji runs the whole Fedora userspace on top of a
RHEL 6 kernel.
This causes some extremely obscure bugs.
Rich.
--
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
virt-df lists disk usage of guests without needing to install any
software inside the virtual machine. Supports Linux and Windows.
http://people.redhat.com/~rjones/virt-df/
2:38 p.m.
On Wed, 5 Jun 2013 20:33:14 +0100
"Richard W.M. Jones" <rjones(a)redhat.com> wrote:
On Tue, Jun 04, 2013 at 04:52:14PM +0200, Simone Caronni wrote:
> Hello,
>
> I have a package that builds fine in a fresh rawhide mock chroot
> and not in a koji build. Is there any difference between the two?
The other thing they don't tell you ... and it's an endless source of
"fun" ... is that Koji runs the whole Fedora userspace on top of a
RHEL 6 kernel.
This causes some extremely obscure bugs.
I'm not sure how you intended this, but it reads like there's some
cabal hiding setup details from you. This is not at all the case. ;)
I'm happy to gather information about or relate information about our
setup to anyone who asks.
If you can think of places we could document things better, please do
add the details there.
There are some oddities due to running RHEL on builders, but I would
call it a pretty corner set of cases, not something to panic about.
kevin
5:08 p.m.
On Wed, Jun 05, 2013 at 01:38:33PM -0600, Kevin Fenzi wrote:
On Wed, 5 Jun 2013 20:33:14 +0100
"Richard W.M. Jones" <rjones(a)redhat.com> wrote:
> On Tue, Jun 04, 2013 at 04:52:14PM +0200, Simone Caronni wrote:
> > Hello,
> >
> > I have a package that builds fine in a fresh rawhide mock chroot
> > and not in a koji build. Is there any difference between the two?
>
> The other thing they don't tell you ... and it's an endless source of
> "fun" ... is that Koji runs the whole Fedora userspace on top of a
> RHEL 6 kernel.
>
> This causes some extremely obscure bugs.
I'm not sure how you intended this, but it reads like there's some
cabal hiding setup details from you. This is not at all the case. ;)
I'm happy to gather information about or relate information about our
setup to anyone who asks.
If you can think of places we could document things better, please do
add the details there.
Sorry I didn't mean to imply this was hidden.
There are some oddities due to running RHEL on builders, but I would
call it a pretty corner set of cases, not something to panic about.
Actually the worse case is EPEL 6 / RHEL 6 which seem to use a RHEL 5
kernel. I have to use an LD_PRELOAD hack to get things to to run
there [attached]. The issue wasn't with the LD_PRELOAD hack itself
which is quite small, but with the amount of days I spent debugging
this.
Rich.
--
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Fedora Windows cross-compiler. Compile Windows programs, test, and
build Windows installers. Over 100 libraries supported.
http://fedoraproject.org/wiki/MinGW
5:16 p.m.
On Wed, 5 Jun 2013 23:08:01 +0100
"Richard W.M. Jones" <rjones(a)redhat.com> wrote:
Sorry I didn't mean to imply this was hidden.
No worries, just wanted to clarify. ;)
> There are some oddities due to running RHEL on builders, but I
would
> call it a pretty corner set of cases, not something to panic about.
Actually the worse case is EPEL 6 / RHEL 6 which seem to use a RHEL 5
kernel. I have to use an LD_PRELOAD hack to get things to to run
there [attached]. The issue wasn't with the LD_PRELOAD hack itself
which is quite small, but with the amount of days I spent debugging
this.
Uh... thats... not likely.
We have currently 0 RHEL5 hosts, so I am at a loss as to how a RHEL5
kernel would get used anywhere. ;(
Builders are mostly RHEL6/x86_64, with 2 RHEL6 ppc64 hosts and a bunch
of Fedora18 arm SOCs doing things like notifications and tag moves (they
don't yet do any builds at all).
Do you have a link to a task that shows this?
kevin
3992
days inactive
3994
days old
7 comments
5 participants
participants (5)
-
Adam Williamson -
Kevin Fenzi -
Richard W.M. Jones -
Seth Vidal -
Simone Caronni