Good morning everyone,
I'm having some problems rebuilding some RHEL6 packages from:
http://ftp.redhat.com/pub/redhat/linux/enterprise/6/en/source/SRPMS/
in particular, mock building python-2.6.5 results in:
Checking for unpackaged file(s): /usr/lib/rpm/check-files /builddir/build/BUILDROOT/python-2.6.5-3.el6.x86_64 error: Installed (but unpackaged) file(s) found: /usr/lib/debug/usr/lib64/libpython2.6.so.1.0.debug-gdb.py /usr/lib/debug/usr/lib64/libpython2.6.so.1.0.debug-gdb.pyc /usr/lib/debug/usr/lib64/libpython2.6.so.1.0.debug-gdb.pyo RPM build errors: Installed (but unpackaged) file(s) found: /usr/lib/debug/usr/lib64/libpython2.6.so.1.0.debug-gdb.py /usr/lib/debug/usr/lib64/libpython2.6.so.1.0.debug-gdb.pyc /usr/lib/debug/usr/lib64/libpython2.6.so.1.0.debug-gdb.pyo Child returncode was: 1
My build-env is a RedHat Workstation 6, beta2 release, mock 1.1.5-1.
I'm looking for suggestions here, as this is my first foray into building recently released upstream RHEL packages:
1. Looks like a bug in the source RPM, but I might be wrong. If so, is it Ok to report it to bugzilla.redhat.com? 2. Something weird in my env being it beta2?
My mock profile config attached.
Thanks.
On 11/12/2010 10:42 AM, Sergio Rubio wrote:
Good morning everyone,
I'm having some problems rebuilding some RHEL6 packages from:
http://ftp.redhat.com/pub/redhat/linux/enterprise/6/en/source/SRPMS/
First round of builds completed here with no issues.
- KB
On Fri, Nov 12, 2010 at 1:14 PM, Karanbir Singh mail-lists@karan.orgwrote:
On 11/12/2010 10:42 AM, Sergio Rubio wrote:
Good morning everyone,
I'm having some problems rebuilding some RHEL6 packages from:
http://ftp.redhat.com/pub/redhat/linux/enterprise/6/en/source/SRPMS/
First round of builds completed here with no issues.
Alright, I'll have a closer look at my build-env then...
Thanks.
- KB
CentOS-devel mailing list CentOS-devel@centos.org http://lists.centos.org/mailman/listinfo/centos-devel
On Fri, Nov 12, 2010 at 1:19 PM, Sergio Rubio sergio@rubio.name wrote:
On Fri, Nov 12, 2010 at 1:14 PM, Karanbir Singh mail-lists@karan.orgwrote:
On 11/12/2010 10:42 AM, Sergio Rubio wrote:
Good morning everyone,
I'm having some problems rebuilding some RHEL6 packages from:
http://ftp.redhat.com/pub/redhat/linux/enterprise/6/en/source/SRPMS/
First round of builds completed here with no issues.
Alright, I'll have a closer look at my build-env then...
Fixed. Some missing deps in config_opts['chroot_setup_cmd'].
Thanks for the input Karanbir.
Thanks.
- KB
CentOS-devel mailing list CentOS-devel@centos.org http://lists.centos.org/mailman/listinfo/centos-devel
Sure, my aplologies. Mock config attached.
On Fri, Nov 12, 2010 at 2:10 PM, David Hrbáč hrbac.conf@seznam.cz wrote:
Dne 12.11.2010 13:50, Sergio Rubio napsal(a):
Fixed. Some missing deps in config_opts['chroot_setup_cmd'].
Thanks for the input Karanbir.
It might be interesting for others to send the resolution, I mean the packages... DH _______________________________________________ CentOS-devel mailing list CentOS-devel@centos.org http://lists.centos.org/mailman/listinfo/centos-devel
On Fri, Nov 12, 2010 at 13:14, Karanbir Singh mail-lists@karan.org wrote:
On 11/12/2010 10:42 AM, Sergio Rubio wrote:
Good morning everyone,
I'm having some problems rebuilding some RHEL6 packages from:
http://ftp.redhat.com/pub/redhat/linux/enterprise/6/en/source/SRPMS/
First round of builds completed here with no issues.
can you tell us: - what os, version do you use on build hosts? - which version of mock and ccache do you use? - and what mock config do you use? since it's well known that there are some packages (like bash, nss) which can not be build in mock on rhel-6: https://bugzilla.redhat.com/show_bug.cgi?id=613392 https://bugzilla.redhat.com/show_bug.cgi?id=609201 so it's rather strange that build complete without issues.
or is these information are top secret?
thanks in advance. regards.
On 11/13/2010 08:25 AM, Farkas Levente wrote:
- what os, version do you use on build hosts?
CentOS-5.5
- which version of mock and ccache do you use?
no ccache, mock-0.6.3
- and what mock config do you use?
nothing interesting in the mock configs.
since it's well known that there are some packages (like bash, nss)
because you have those problems, does not mean everyone else will.
which can not be build in mock on rhel-6: https://bugzilla.redhat.com/show_bug.cgi?id=613392 https://bugzilla.redhat.com/show_bug.cgi?id=609201 so it's rather strange that build complete without issues.
Speak to the people who have the problems ? These packages have all clearly built not only for fedora, but also for red hat - and going by the changelogs, quite a few times.
- KB
On 11/13/2010 02:17 PM, Karanbir Singh wrote:
On 11/13/2010 08:25 AM, Farkas Levente wrote:
- what os, version do you use on build hosts?
CentOS-5.5
this probably not the best thing since in this case the build host's kernel version only 2.6.18, which can cause degraded glibc result. that's the reason why all fedora buildsys are rhel-6(beta) see: http://lists.fedoraproject.org/pipermail/devel/2010-July/139255.html
- which version of mock and ccache do you use?
no ccache, mock-0.6.3
which is more then 4 years old. since then many things happened...
- and what mock config do you use?
nothing interesting in the mock configs.
since it's well known that there are some packages (like bash, nss)
because you have those problems, does not mean everyone else will.
because you don't have those problems, does not mean nobody else have... and that was the reason why i wrote this letter (as i assume the difference is the build environment).
which can not be build in mock on rhel-6: https://bugzilla.redhat.com/show_bug.cgi?id=613392 https://bugzilla.redhat.com/show_bug.cgi?id=609201 so it's rather strange that build complete without issues.
Speak to the people who have the problems ? These packages have all clearly built not only for fedora, but also for red hat - and going by the changelogs, quite a few times.
i spoken and rh already verified it, but as rh official build way is rpmbuild and not mock (see: https://bugzilla.redhat.com/show_bug.cgi?id=613392#c4) they not really care about it. the reason why it's working for you probably your very old mock version and centos-5 build hosts.
2010/11/13 Karanbir Singh mail-lists@karan.org:
On 11/13/2010 08:25 AM, Farkas Levente wrote:
- what os, version do you use on build hosts?
CentOS-5.5
Why not RHEL6 or Fedora 12?
Best regards,
Morten
On 11/13/2010 02:40 PM, Farkas Levente wrote:
- which version of mock and ccache do you use?
no ccache, mock-0.6.3
which is more then 4 years old. since then many things happened...
*shrug* so ? running rpmbuild isnt rocket science.
because you don't have those problems, does not mean nobody else have... and that was the reason why i wrote this letter (as i assume the difference is the build environment).
This is the CentOS devel list, were talking about what goes into the CentOS build right ?
i spoken and rh already verified it, but as rh official build way is rpmbuild and not mock (see: https://bugzilla.redhat.com/show_bug.cgi?id=613392#c4) they not really care about it. the reason why it's working for you probably your very old mock version and centos-5 build hosts.
You say old, I say working. Our emphasis is on creating a stable and sustainable setup, I dont care if a build will take 5% longer to run; but i care *lots* if that build isn't going to give me the same result every time. So the aim isnt to break something that already works, but to use what works towards the end result : the builds.
At the moment the centos3/4/5 and centos6 build services are completely segregated. But in the next few months we will need to merge things in, how and what we do at that time will depend on testing that we do at the time. Its not a problem we need to solve right away.
The thing we *do* need to get done right away is the branding identification and patching.
- KB
Dne 14.11.2010 0:11, Karanbir Singh napsal(a):
You say old, I say working. Our emphasis is on creating a stable and sustainable setup, I dont care if a build will take 5% longer to run; but i care *lots* if that build isn't going to give me the same result every time. So the aim isnt to break something that already works, but to use what works towards the end result : the builds.
Hi, Are there any on-line references on this issue I can read about? Thanks, DH
Hi,
Sort of falling into this thread mid way, but I had an issue failing to build the RHEL6 glibc rpm on RHEl6beta due to unpackaged debug files. Reported this as a bug: https://bugzilla.redhat.com/show_bug.cgi?id=653905 .
Adding
%_enable_debug_packages 1
to .rpmmacros fixed the packaging issue for this package. Might work for other builds that fail due to unpackaged files too.
Regards, Leonard.
On 11/16/2010 05:12 PM, David Hrbáč wrote:
You say old, I say working.
Are there any on-line references on this issue I can read about?
The important thing is - what is in place now works, and works well. I dont see anything interesting in the newer mock builds to merit an upgrade. On the other hand, was it not you who originally reported issues with the file name based deps were broken for a long time with yum-3.0 / mock > 0.7; and then there were selinux issues with mock >=8 on EL5. I remember working through some of these issues with Tuomo ( bleve on irc.f.n ).
In all fairness, since CentOS-3 is now EOL, the next step might be to re-eval the situation and only consider the components we need to build. A good time to do that would be when we merge the C-6 buildsys stuff into the regular buildsystem. For now, since the C-6 buildqueues will ( or would have ) stayed quite busy for a while, we didnt want that to cause backlogs on updates for c4/c5 going out - so we setup a parallel bsys instance.
I will make sure that a lot of noise is made when we start merging stuff in, and all help / comments would be very welcome :) the 'new' system is based around a stom-server instance with various task-threads hanging off, extremely simple and very ripe for public distribution.
Much of this is perhaps not be relevant to the topic..
- KB
On Wed, 2010-11-17 at 10:27 +0000, Karanbir Singh wrote:
On 11/16/2010 05:12 PM, David Hrbáč wrote:
In all fairness, since CentOS-3 is now EOL, the next step might be to re-eval the situation and only consider the components we need to build. A good time to do that would be when we merge the C-6 buildsys stuff into the regular buildsystem. For now, since the C-6 buildqueues will ( or would have ) stayed quite busy for a while, we didnt want that to cause backlogs on updates for c4/c5 going out - so we setup a parallel bsys instance.
No build priority preemption? Entirely possible queuing priority versus build/distro/package/importance.
I will make sure that a lot of noise is made when we start merging stuff in, and all help / comments would be very welcome :) the 'new' system is based around a stom-server instance with various task-threads hanging off, extremely simple and very ripe for public distribution.
Public Distribution? Perhaps when? Would like to give it a spin.
Much of this is perhaps not be relevant to the topic..
You could make another thread?
John
On 11/17/2010 03:01 PM, JohnS wrote:
No build priority preemption? Entirely possible queuing priority versus build/distro/package/importance.
Not really, messages can be related. Either on conditional external tests or via queue allocation / hijack / bury process. In this case, I'm working with conditionals.
I will make sure that a lot of noise is made when we start merging stuff in, and all help / comments would be very welcome :) the 'new' system is
Public Distribution? Perhaps when? Would like to give it a spin.
couple of weeks after 6 is released.
-KB
On Wed, 2010-11-17 at 15:04 +0000, Karanbir Singh wrote:
On 11/17/2010 03:01 PM, JohnS wrote:
No build priority preemption? Entirely possible queuing priority versus build/distro/package/importance.
Not really, messages can be related. Either on conditional external tests or via queue allocation / hijack / bury process. In this case, I'm working with conditionals.
I will make sure that a lot of noise is made when we start merging stuff in, and all help / comments would be very welcome :) the 'new' system is
Public Distribution? Perhaps when? Would like to give it a spin.
couple of weeks after 6 is released.
-KB
--- More than happy to Guinea Pig it around before then...
John
On Wed, 17 Nov 2010, Karanbir Singh wrote:
On 11/17/2010 03:01 PM, JohnS wrote:
No build priority preemption? Entirely possible queuing priority versus build/distro/package/importance.
Not really, messages can be related. Either on conditional external tests or via queue allocation / hijack / bury process. In this case, I'm working with conditionals.
I will make sure that a lot of noise is made when we start merging stuff in, and all help / comments would be very welcome :) the 'new' system is
Public Distribution? Perhaps when? Would like to give it a spin.
couple of weeks after 6 is released.
So, in about a week, eh?
On Thu, 18 Nov 2010 14:57:58 -0500 (EST), Charlie Brady charlieb-centos-devel@budge.apana.org.au wrote:
On Wed, 17 Nov 2010, Karanbir Singh wrote:
couple of weeks after 6 is released.
So, in about a week, eh?
Dream on. I just requested a 30 day eval. And all I can get is a Beta of RHEL6. So it seems RHEL 6 is not realy realy there yet. It almost sounds like another OS vendor ;-)
Hugo.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
thus Hugo van der Kooij spake:
On Thu, 18 Nov 2010 14:57:58 -0500 (EST), Charlie Brady charlieb-centos-devel@budge.apana.org.au wrote:
On Wed, 17 Nov 2010, Karanbir Singh wrote:
couple of weeks after 6 is released.
So, in about a week, eh?
Dream on. I just requested a 30 day eval. And all I can get is a Beta of RHEL6. So it seems RHEL 6 is not realy realy there yet. It almost sounds like another OS vendor ;-)
I got mine without any problems. It's not beta.
Timo
Hugo.
On Fri, Nov 19, 2010 at 01:08:01PM +0200, Hugo van der Kooij wrote:
couple of weeks after 6 is released.
So, in about a week, eh?
Dream on. I just requested a 30 day eval. And all I can get is a Beta of RHEL6. So it seems RHEL 6 is not realy realy there yet.
That's just because the RHN web page design sucks.
The item next to the [-] is also clickable.
On 11/19/2010 05:16 AM, Timo Schoeler wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
thus Hugo van der Kooij spake:
On Thu, 18 Nov 2010 14:57:58 -0500 (EST), Charlie Brady charlieb-centos-devel@budge.apana.org.au wrote:
On Wed, 17 Nov 2010, Karanbir Singh wrote:
couple of weeks after 6 is released.
So, in about a week, eh?
Dream on. I just requested a 30 day eval. And all I can get is a Beta of RHEL6. So it seems RHEL 6 is not realy realy there yet. It almost sounds like another OS vendor ;-)
I got mine without any problems. It's not beta.
I submitted my request last saturday, getting the automated confirmation email telling me I'd get my 2 port activation soon. Nothing yet. I think YMMV, perhaps depending on how they profile you. I would guess there is a minimal human intervention in this process, and things are triaged by your profiled importance as a customer.
I know it probably won't happen, but I would love to see early access to round1 packages or the like. I know all the reasons it won't happen, but it would be nice. beta2 I'm guessing would start causing me problems around the time I decided I really wanted to use something from epel.
-dmc
Timo
Hugo.
-----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux)
iD8DBQFM5lxyfg746kcGBOwRAjitAKCX+cDJoukZHBvKAOH2+8eXs6Y42ACgk/ZL Hdnom9cgSYC9CVk7/ZFq0i0= =ESml -----END PGP SIGNATURE----- _______________________________________________ CentOS-devel mailing list CentOS-devel@centos.org http://lists.centos.org/mailman/listinfo/centos-devel
On 11/12/2010 01:14 PM, Karanbir Singh wrote:
On 11/12/2010 10:42 AM, Sergio Rubio wrote:
Good morning everyone,
I'm having some problems rebuilding some RHEL6 packages from:
http://ftp.redhat.com/pub/redhat/linux/enterprise/6/en/source/SRPMS/
First round of builds completed here with no issues.
this statement is rather strange after having a dozen of packages which is well-known not build under mock #661775 #661784 #661804 #621089 #657567 etc. because of different missing BRs and other problems. not to mention virt-top which depend on ocaml-camomile-data which is neither on rhn nor on ftp site, so simple missing from rhel-6. #661783 ie. this src.rpm is impossible to build on rhel or centos without fedora or epel packages.
so dare i ask what do you mean by "completed here with no issues" and what packages used in your mock buildroot's yum base?