Extras build system back up
by Dan Williams
Hi,
We're back up and using the latest plague code. yum update your
plague-client and you should be good to go.
Dan
17 years, 4 months
Mock hangs - why?
by Al Pacifico
I have fifteen CPAN packages that I'm packaging as
RPMs for FC5, and one (and only one) of them seems to
hang mock. Is this an appropriate place to ask for
advice?
Thanks.
-al
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
17 years, 4 months
Mock going forward
by Clark Williams
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Ok, last unsolicited mock email of the day. Really! I promise...
We're currently shipping mock from the mock-0.6 branch of the mock CVS
repository. The HEAD (0.7) is the branch that contains the
re-architected launcher and the elevate/drop privilege logic. I'm
wondering when/how we push the new version of mock out to the world?
My preference would be to wait for FC6 to go gold, then switch over to
HEAD and deal with mock breakage at that point. Of course, the
complete lack of complaints from all you dedicated fedora buildsys
readers probably means that mock-0.7 is perfect in every way, so we
probably don't have to worry about that, eh? :)
Anyway, I'd be interested to hear what everyone thinks and I'd like to
learn what the mechanism is for implementing it. Is implementation
merely telling Jeremy or Jesse what branch to pull from? Or do we have
to do more?
Clark
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org
iD8DBQFE410cHyuj/+TTEp0RAsmsAKDDlOns80a9CorfSURSSEzjp+wslgCeLRrK
9Mcp4IziZ6O3ojXjcm64GoQ=
=IeOW
-----END PGP SIGNATURE-----
17 years, 6 months
sparc weirdness
by Dennis Gilmore
Hey all,
I have a odd situtation that im trying to work out the best way to handle.
gcc on sparc needs glibc for sparc and sparc64 but it gets built as sparc.
somehow i need to have sparc64 glibc in the buildroot only for gcc. even
though the target is sparc.
Dennis
17 years, 7 months
Mock hangs - why
by Al Pacifico
Sorry to post twice, but I need to correct what I just
wrote, as the problem **HAS** changed following the
kernel upgrade from kernel-smp.i686 2.6.17-1.2157_FC5
to kernel-smp.i686 2.6.17-1.2174_FC5:
--- Al Pacifico <adpacifico(a)yahoo.com> wrote:
> Date: Thu, 17 Aug 2006 16:59:09 -0700 (PDT)
<snip>
>
> Dan-
> I don't think that is the problem. At present, mock
> is
> hung (as described above) and 'pstree build' yields:
>
bash───mock───sh───runuser───bash───runuser───rpmbuild───sh───perl
This didn't make sense, in that if the problem
persisted after the kernel upgrade, the output should
be:
bash───mock───sh───runuser
or, perhaps:
bash───mock───sh───runuser───bash───runuser
since previously the debugging output suggests mock
has hung prior to or while starting rpmbuild.
Repeated, with --debug, it seems to hang at the same
place:
DEBUG: Executing /usr/sbin/mock-helper chroot
/var/lib/mock/fedora-5-i386-core/root /sbin/runuser -
root -c "cd /;/sbin/runuser -c 'rpmbuild --rebuild
--target i386 --nodeps
/builddir/build/SRPMS/perl-DBIx-Migration-0.05-1.fc5.src.rpm'
mockbuild"
...and after seeing that, I prematurely posted that
the problem was uncorrected with a kernel update.
After typing 'u'<ENTER>, I see:
u
Installing
/builddir/build/SRPMS/perl-DBIx-Migration-0.05-1.fc5.src.rpm
Building target platforms: i386
Building for target i386
Executing(%prep): /bin/sh -e /var/tmp/rpm-tmp.10545
<snip>
Executing(%build): /bin/sh -e /var/tmp/rpm-tmp.10545
<snip>
This module requires Module::Build to install itself.
Install Module::Build now from CPAN? [y] *** Cannot
install without Module::Build. Exiting ...
error: Bad exit status from /var/tmp/rpm-tmp.10545
<snip>
Error building package from
perl-DBIx-Migration-0.05-1.fc5.src.rpm, See build log
ending
DEBUG: Executing /usr/sbin/mock-helper umount
/var/lib/mock/fedora-5-i386-core/root/proc
DEBUG: Executing /usr/sbin/mock-helper umount
/var/lib/mock/fedora-5-i386-core/root/dev/pts
done
[build@powell SRPMS]$
I modified the .spec file, adding a BuildRequire line
for the Module::Build module and a couple more, and
now the SRPM rebuilds fine under mock. Thanks to Dan
for sharing his experience and insights!
With 'mock shell', however, the previously-described
behavior persists despite the kernel upgrade.
-al
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
17 years, 7 months
Mock hangs - why
by Al Pacifico
>
> Date: Thu, 17 Aug 2006 10:10:59 -0400
> From: Dan Williams <dcbw(a)redhat.com>
> Subject: Re: Mock hangs - why
> To: Discussion of Fedora build system
> <fedora-buildsys-list(a)redhat.com>
> Message-ID:
> <1155823859.2665.21.camel(a)localhost.localdomain>
> Content-Type: text/plain
>
> On Wed, 2006-08-16 at 17:21 -0700, Al Pacifico
> wrote:
<snip>
> > But 'mock perl-DBIx-Migration-0.05-1.fc5.src.rpm'
> > seems to hang. Adding the --debug option, the
> screen
> > output (and root log) of 'mock
> > perl-DBIx-Migration-0.05-1.fc5.src.rpm' ends with:
> >
> > Executing /usr/sbin/mock-helper chroot
> > /var/lib/mock/fedora-5-i386-core/root
> /sbin/runuser -
> > root -c "cd /;/sbin/runuser -c 'rpmbuild --rebuild
> > --target i386 --nodeps
> >
>
/builddir/build/SRPMS/perl-DBIx-Migration-0.05-1.fc5.src.rpm'
> > mockbuild"
> >
> > Then, the command prompt never comes back and when
> I
> > interrupt mock with ctrl-C (I've left my machine
> > running mock with this SRPM for over a day, so I
> don't
> > think it's an impatience thing), keystrokes are
> > retyped at the console after a long delay and many
> > keystrokes, if at all.
>
> Is some process taking up all your CPU? Is that
> process part of the
> build of your package?
>
> Install and look at the output of the 'pstree'
> command; that will show
> you exactly what mock has launched and the
> parent/child relationships
> between the build processes and mock. You can also
> try to 'strace' the
> process that seems to be hanging and perhaps get
> more information out of
> it. Sometimes attaching to the hung build process
> with gdb shows usable
> backtrace info, sometimes not.
>
> In the Extras buildsystem we've had problems with
> reproducible build
> hangs a few times, which were generally solved by
> updating the kernel.
>
> Dan
>
Dan-
I don't think that is the problem. At present, mock is
hung (as described above) and 'pstree build' yields:
bash───mock───sh───runuser───bash───runuser───rpmbuild───sh───perl
and 'uptime' yields:
16:30:20 up 7:54, 3 users, load average: 0.00,
0.00, 0.00
At present, I'm starting to work on strace... I need
to carefully review that thread on fedora-selinux list
to see how Paul Howarth limited strace to the command
in question or else I'll likely be overwhelmed by the
output of 'strace mock <culprit-SRPM>'.
<snip>
> Date: Thu, 17 Aug 2006 10:36:58 -0500
> From: Clark Williams <williams(a)redhat.com>
> Subject: Re: Mock hangs - why
> To: Discussion of Fedora build system
> <fedora-buildsys-list(a)redhat.com>
> Message-ID: <44E48D1A.2070202(a)redhat.com>
> Content-Type: text/plain; charset=ISO-8859-1
>
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Dan Williams wrote:
<snip>
> > In the Extras buildsystem we've had problems with
> reproducible build
> > hangs a few times, which were generally solved by
> updating the kernel.
> >
>
> What kernel are you running on the build system?
>
Clark-
Reproducible with kernel-smp.i686 2.6.17-1.2157_FC5
and kernel-smp.i686 2.6.17-1.2174_FC5. I updated to
2174 this morning after reading what Dan wrote.
I'm still wondering if the password prompt following
'su' after executing 'mock shell' gives the same
behavior on others' systems... and if not, which
kernel they are running (might help pinpoint the
problem).
-al
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
17 years, 7 months
Mock hangs - why
by Al Pacifico
I have fifteen CPAN packages that I'm packaging as
RPMs for FC5, and one (and just one) of them seems to
hang mock (mock-0.6-4.fc5). All of them check out with
rpmlint and all build using rpmbuild. All SRPMs except
'perl-DBIx-Migration-0.05-1.fc5.src.rpm' build
flawlessly with mock.
But 'mock perl-DBIx-Migration-0.05-1.fc5.src.rpm'
seems to hang. Adding the --debug option, the screen
output (and root log) of 'mock
perl-DBIx-Migration-0.05-1.fc5.src.rpm' ends with:
Executing /usr/sbin/mock-helper chroot
/var/lib/mock/fedora-5-i386-core/root /sbin/runuser -
root -c "cd /;/sbin/runuser -c 'rpmbuild --rebuild
--target i386 --nodeps
/builddir/build/SRPMS/perl-DBIx-Migration-0.05-1.fc5.src.rpm'
mockbuild"
Then, the command prompt never comes back and when I
interrupt mock with ctrl-C (I've left my machine
running mock with this SRPM for over a day, so I don't
think it's an impatience thing), keystrokes are
retyped at the console after a long delay and many
keystrokes, if at all.
With my other 14 packages, if using the --debug
option, this command is followed by the usual rpmbuild
output at this point, the usual mock 'Results and/or
logs in: /var/lib/mock/fedora-5-i386-core/result'
message, and then the command prompt.
I solicited advice from Jesse Keating who suggested
trying the mock shell, and what I learned is that 'su
- mockbuild' prompts for a password even though I'm in
the chroot'ed shell as 'root'. What is puzzling is
that this is before trying to rebuild the SRPM, so
it's not that this package alters the behavior of mock
(as I had previously thought).
So, I tried the following (on Jesse's advice):
[build@powell SRPMS]$ mock init
init
clean
prep
This may take a while
ending
done
Finished initializing root
[build@powell SRPMS]$ mock shell
init
mock-chroot> id
uid=0(root) gid=501 groups=101,501
context=user_u:system_r:unconfined_t:s0
mock-chroot> su - mockbuild
Password:
su: incorrect password
This leads me to believe that when I try to rebuild
this SRPM using 'mock
perl-DBIx-Migration-0.05-1.fc5.src.rpm', mock is
prompting me for a password for user 'mockbuild' and
therefore appears to hang.
I have no idea why this does not happen with the other
fourteen SRPM's and just with this one or when using
'mock shell'.
Google offered a tantalizing clue on the
fedora-selinux list (last post in the thread is at
URL:
http://listman.redhat.com/archives/fedora-selinux-list/2006-August/msg000...)
which described mock hanging at the same place on
older versions of Fedora Core, related to older
versions of selinux.
Output of rpm -V mock is:
[build@powell SRPMS]$ rpm -V mock
S.5....T. c /etc/mock/fedora-5-i386-core.cfg
[build@powell SRPMS]$
which is exactly what I'd expect, and other packages
build fine under mock:
[build@powell SRPMS]$ mock
perl-Tie-RegexpHash-0.15-1.src.rpm
init
clean
prep
This may take a while
setup
build
ending
done
Results and/or logs in:
/var/lib/mock/fedora-5-i386-core/result
... so mock is not goofed up.
My /etc/mock/default.cfg points to a slightly tweaked
configuration script, which basically just changes the
repositories to local mirrors. I've attached it to
this message.
Couple of questions:
1. Can someone check on his/her box that 'su
mockbuild' from the chroot'ed shell has the same
behavior and post back?
2. Would a mock guru out there have any suggestions?
-al
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
#!/usr/bin/python -tt
import os
config_opts['root'] = 'fedora-5-i386-core'
config_opts['target_arch'] = 'i386'
config_opts['yum.conf'] = """
[main]
cachedir=/var/cache/yum
debuglevel=1
reposdir=/dev/null
logfile=/var/log/yum.log
retries=20
obsoletes=1
gpgcheck=0
assumeyes=1
# repos
[core]
name=core
#baseurl=http://download.fedora.redhat.com/pub/fedora/linux/core/$releasever/$basearch/os/
#mirrorlist=http://fedora.redhat.com/download/mirrors/fedora-core-$releasever
baseurl=http://joplin/yum/Fedora/Core/5/base/i386/
enabled=1
#following line for "priorities" yum plug-in.... see http://wiki.centos.org/PackageManagement/Yum/Priorities
priority=1
gpgcheck=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-fedora file:///etc/pki/rpm-gpg/RPM-GPG-KEY
[updates-released]
name=updates
#baseurl=http://download.fedora.redhat.com/pub/fedora/linux/core/updates/$releasever/$basearch/
#mirrorlist=http://fedora.redhat.com/download/mirrors/updates-released-fc$releasever
baseurl=http://joplin/yum/Fedora/Core/5/updates/i386/
enabled=1
#following line for "priorities" yum plug-in.... see http://wiki.centos.org/PackageManagement/Yum/Priorities
priority=1
gpgcheck=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-fedora
[groups]
name=groups
baseurl=http://buildsys.fedoraproject.org/buildgroups/5/i386/
[extras]
name=extras
#baseurl=http://download.fedora.redhat.com/pub/fedora/linux/extras/$releasever/$basearch/
mirrorlist=http://fedora.redhat.com/download/mirrors/fedora-extras-5
enabled=1
#following line for "priorities" yum plug-in.... see http://wiki.centos.org/PackageManagement/Yum/Priorities
priority=10
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-fedora-extras
gpgcheck=1
[pacifico-testing]
name=Fedora Core Pacifico testting $releasever - $basearch
baseurl=file:///home/build/yum-repo/Fedora/Core/5/pacifico-testing/i386/
enabled=1
gpgcheck=1
gpgkey=file:///etc/pki/rpm-gpg/GPG-KEY-apacifico
# Name: RPMforge RPM Repository for Fedora Core 5 - i386
# URL: http://rpmforge.net/
[rpmforge]
name = Fedora Core 5 - i386 - RPMforge.net - dries
#baseurl = http://apt.sw.be/dries/fedora/fc5/$basearch/dries/RPMS
mirrorlist = http://apt.sw.be/dries/fedora/fc5/mirrors-rpmforge
#mirrorlist = file:///etc/yum.repos.d/mirrors-rpmforge
enabled = 1
gpgkey = file:///etc/pki/rpm-gpg/RPM-GPG-KEY-rpmforge-dries
gpgcheck = 1
"""
17 years, 7 months
updated man page
by Clark Williams
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Ok, I lied about not posting another mock message today. You can dock
my mock pay 50%...
Attached is an updated man page for mock. Please take a moment to look
it over (man ./mock.1) and see if there are any egregious errors.
Clark
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org
iD8DBQFE450OHyuj/+TTEp0RAkL6AJ99q+dJl1yjjAxYsZCsurgFa440iQCg291S
mGMG87HdfVug6x6V0YAAdEg=
=Dc9E
-----END PGP SIGNATURE-----
.TH "mock" "1" "0.4" "Seth Vidal" ""
.SH "NAME"
.LP
mock \- build SRPMs in a chroot
.SH "SYNTAX"
.LP
mock [options] [\fBrebuild\fR] \fISRPM [\fISRPM...\fR]
.LP
mock [options] \fBchroot\fR \fI<cmd>\fR
.LP
mock [options] {\fBinit\fR|\fBclean\fR|\fBshell\fR}
.SH "DESCRIPTION"
.LP
Mock is a simple chroot/rpm building program. It doesn't do anything
terribly fancy other than populate a chroot with the contents
specified by a configuration file, then build any input SRPM(s) in
that chroot.
.LP
The content of a chroot is specified by the configuration
specified with the \fB\-r\fR option. A default configuration file is
specified in /etc/mock/default.cfg, which is usually a symlink to one
of the installed configurations.
.SH "OPTIONS"
.LP
.TP
\fB\-r\fR \fICHROOT\fP
Uses specified chroot configuration as defined in
/etc/mock/<\fIchroot\fP>.cfg If none specified, uses the chroot linked
to by /etc/mock/default.cfg
.TP
\fB\-\-no\-clean\fR
Do not clean chroot before building package.
.TP
\fB\-\-arch=\fR\fIARCH\fP
Specify target build arch.
.TP
\fB\-\-debug\fR
Output copious debugging information.
.TP
\fB\-\-resultdir=\fR\fIRESULTDIR\fP
Change directory where resulting files (RPMs and build logs) are written
.TP
\fB\-\-statedir=\fR\fISTATEDIR\fP
Change directory where state information is written
.TP
\fB\-\-uniqueext=\fR\fItext\fP
Arbitrary, unique extension to append to buildroot directory name
.TP
\fB\-\-configdir=\fR\fICONFIGDIR\fP
Change directory where config files are found
.TP
\fB\-\-quiet\fR
Suppress most output
.TP
\fB\-\-autocache\fR
Turn on build-root caching
.TP
\fB\-\-rebuildcach\fR
Force rebuild of build-root cache
.TP
\fB\-\-help\fR
Show usage information and exit.
.TP
\fB\-\-version\fR
Show version number and exit.
.SH "COMMANDS"
.LP
.TP
\fBinit\fR \- initialize a chroot (install packages, setup devices, etc.)
.TP
\fBchroot\fR \- run the specified command within the chroot (which must already be initialized)
.TP
\fBclean\fR \- purge the chroot tree
.TP
\fBshell\fR \- start an interactive shell in the specified chroot
.TP
\fBrebuild\fR \- If no command is specified, rebuild is assumed. Rebuilds the specified SRPM(s)
.SH "FILES"
.LP
\fI/etc/mock/\fP \- default configuration directory
.LP
\fI/var/lib/mock\fP \- directory where chroots are created
.SH "EXAMPLES"
.LP
To rebuild test.src.rpm under the default chroot:
.LP
.RS 5
\fBmock /path/to/test.src.rpm\fR
.RE
.LP
To specify a different chroot (which must be defined in /etc/mock):
.LP
.RS 5
\fBmock \-r fedora\-4\-i386\-core /path/to/test.src.rpm\fR
.RE
.SH "BUGS"
.LP
To report bugs in mock, go to:
.LP
.RS 5
\fIhttp://bugzilla.redhat.com/bugzilla\fR.
.RE
.LP
Select the query option, using product \fBFedora Hosted Projects\fR
and component \fBmock\fR and press the \fBStart Search\fR
button. If there is a bug similar to the one you are seeing, add your
information to the comments. If not, go to the new bug page using the
same product and component and fill in the form.
.SH "AUTHORS"
.LP
Seth Vidal <skvidal(a)linux.duke.edu>
.LP
and a cast of...tens
.SH "SEE ALSO"
.LP
yum(8) rpmbuild(8)
http://fedoraproject.org/wiki/Projects/Mock
17 years, 7 months
Disttags in mock?
by Clark Williams
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Do we need to add a disttag to the mock specfile?
My initial thought was no, since it's a noarch and pretty much should
run on any python >= 2.2. But, I'm not positive that we shouldn't add
a disttag; just a nagging feeling. So I thought I'd throw it out there...
Clark
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org
iD8DBQFE41HDHyuj/+TTEp0RAi8oAKCpi9CC53V16ppZBdlzVmNd/qoZGQCcDx9D
jAmtiHxP+Zcv+W7PkfO4aEI=
=n2Xp
-----END PGP SIGNATURE-----
17 years, 7 months