9 a.m.
Hi everybody,
This years schedule does not give us too much freedom, because the
Fedora mass rebuild is scheduled for the 22nd of January, which is in a
few days. There are also some concerns with keyword arguments, but since
I have no idea how problematic this might be, I decided to give it a go.
It seems that nowadays, everybody is able to ask for side tag via
`fedpkg` so I acquired one:
~~~
$ fedpkg request-side-tag
Side tag 'f32-build-side-17977' (id 17977) created.
Use 'fedpkg build --target=f32-build-side-17977' to use it.
Use 'koji wait-repo f32-build-side-17977' to wait for the build repo to
be generated.
~~~
The name is not really nice, but ... So now, there is a opportunity for
your help.
This is the list of packages, which very likely needs rebuild:
~~~
$ dnf repoquery --disablerepo=* --enablerepo=rawhide
--enablerepo=rawhide-source --arch=src --whatrequires 'ruby-devel' |
sort | uniq
~~~
You can take the package and just fire rebuild, but please ensure that
you are using f30-ruby build target [2], i.e. the build command should
look like:
~~~
$ fedpkg build --target f32-build-side-17977
~~~
Please be careful, because if you, by a chance, omit the
f32-build-side-17977 target, you'll be building against Ruby 2.6 which
is not what you want.
If you won't do it by yourself, I'll be rebuilding all packages after I
am finished with my packages. I'll be using fermig [1] to help me with
that. If you don't want me to touch your packages for whatever reason,
please let me know.
You can follow the progress at:
https://koji.fedoraproject.org/koji/builds?inherited=0&tagID=17977&am...
or using:
~~~
$ koji list-tagged f32-build-side-17977
~~~
As always, any help/testing/feedback is welcome.
Vít
[1] https://github.com/fedora-ruby/fermig
[2] https://bugs.ruby-lang.org/issues/16431
[3]
https://src.fedoraproject.org/rpms/ruby-ldap/c/9a16c79b85525ef9636aa50e17...
2:45 p.m.
There are also some concerns with keyword arguments, but since I have
no
idea how problematic this might be, I decided to give it a go.
For what it's worth, I've tested Ruby 2.7 on a fairly large number of
packages (a couple hundred) and I only encountered one warning regarding
keyword arguments. So it's not a huge issue as far as I'm concerned.
Here's what the warning look like:
~/projects/asciidoctor/kramdown-asciidoc/.bundle/gems/ruby/2.7.0/gems/deep-cover-core-0.7.7/lib/deep_cover/analyser/stats.rb:51:
warning: Using the last argument as keyword parameters is deprecated; maybe
** should be added to the call
~/projects/asciidoctor/kramdown-asciidoc/.bundle/gems/ruby/2.7.0/gems/deep-cover-core-0.7.7/lib/deep_cover/analyser/stats.rb:17:
warning: The called method `initialize' is defined here
Both warnings pertain to the same method definition, one at the call site
and one at the definition.
Here's the definition:
def initialize(executed: 0, not_executed: 0, not_executable: 0,
ignored: 0)
@executed = executed
@not_executed = not_executed
@not_executable = not_executable
@ignored = ignored
freeze
end
Here's the call site:
Analyser::StatsBase.new(to_h.transform_values { |v| (100 *
v).fdiv(total).round(DECIMALS) })
So a hash is being passed off as keyword arguments. Ruby wants you to make
the conversion explicit. That's the essence of this change.
Best,
-Dan
--
Dan Allen | @mojavelinux | https://twitter.com/mojavelinux
12:33 a.m.
Vít Ondruch wrote on 2020/01/17 0:00:
Hi everybody,
This years schedule does not give us too much freedom, because the
Fedora mass rebuild is scheduled for the 22nd of January, which is in a
few days. There are also some concerns with keyword arguments, but since
I have no idea how problematic this might be, I decided to give it a go.
It seems that nowadays, everybody is able to ask for side tag via
`fedpkg` so I acquired one:
<snip>
Thank you (and thanks to other people) for excellent works to bring
ruby 2.7 into Fedora 32. As Vít says, the rebuilt packages can be seen on:
https://koji.fedoraproject.org/koji/builds?inherited=0&tagID=17977&am...
92 packages were already rebuilt.
Current rebuild leftovers (with regards to library dependency) are:
$ dnf repoquery --disablerepo=\* --enablerepo=koji-ruby27 --qf "%{sourcerpm}"
--whatrequires "libruby.so.2.6()(64bit)" | sort | uniq | cat -n
1 graphviz-2.42.2-4.fc32.src.rpm
https://koji.fedoraproject.org/koji/taskinfo?taskID=40688927
gv_ruby.cpp: In function 'void SWIG_RubyInitializeTrackings()':
gv_ruby.cpp:1267:85: error: call of overloaded 'rb_define_virtual_variable(const char
[21], VALUE (&)(...), NULL)' is ambiguous
1267 | rb_define_virtual_variable("SWIG_TRACKINGS_COUNT",
swig_ruby_trackings_count, NULL);
|
^
In file included from /usr/include/ruby/ruby.h:2863,
from /usr/include/ruby.h:33,
from gv_ruby.cpp:880:
/usr/include/ruby/backward/cxxanyargs.hpp:59:1: note: candidate: 'void
ruby::backward::cxxanyargs::rb_define_virtual_variable(const char*, VALUE (*)(...), void
(*)(...))'
59 | rb_define_virtual_variable(const char *q, type *w, void_type *e)
| ^~~~~~~~~~~~~~~~~~~~~~~~~~
/usr/include/ruby/backward/cxxanyargs.hpp:90:1: note: candidate: 'void
ruby::backward::cxxanyargs::rb_define_virtual_variable(const char*, VALUE (*)(...), void
(*)(VALUE, ID, VALUE*))'
90 | rb_define_virtual_variable(const char *q, type *w, rb_gvar_setter_t *e)
| ^~~~~~~~~~~~~~~~~~~~~~~~~~
Looks like this is below and it seems swig needs patching.
https://github.com/swig/swig/issues/1689
https://github.com/swig/swig/pull/1692/commits/00e291b319bd6b58bf061feee3...
2 libprelude-5.1.1-2.fc32.src.rpm
https://koji.fedoraproject.org/koji/taskinfo?taskID=40688760
Looks like the same issue as graphviz.
Prelude.cxx:1263:85: error: call of overloaded 'rb_define_virtual_variable(const char
[21], VALUE (&)(...), NULL)' is ambiguous
3 libsbml-5.18.0-9.fc32.src.rpm
https://koji.fedoraproject.org/koji/taskinfo?taskID=40688622
Looks like the same issue as graphviz.
/builddir/build/BUILD/libSBML-5.18.0-Source/build/src/bindings/ruby/libsbml_wrap.cpp:1268:85:
error: call of overloaded 'rb_define_virtual_variable(const char [21], VALUE
(&)(...), NULL)' is ambiguous
4 libsedml-0.4.4-7.fc32.src.rpm
https://koji.fedoraproject.org/koji/taskinfo?taskID=40688515
Looks like the same issue as graphviz.
/builddir/build/BUILD/libSEDML-0.4.4/build/bindings/ruby/libsedml_wrap.cpp:1268:85: error:
call of overloaded 'rb_define_virtual_variable(const char [21], VALUE (&)(...),
NULL)' is ambiguous
5 libyui-bindings-1.1.2-19.fc32.src.rpm
https://koji.fedoraproject.org/koji/taskinfo?taskID=40685537
Looks like the same issue as graphviz.
/builddir/build/BUILD/libyui-bindings-1.1.2/build-x86_64-redhat-linux-gnu/swig/ruby/yui_ruby.cxx:1267:85:
error: call of overloaded 'rb_define_virtual_variable(const char [21], VALUE
(&)(...), NULL)' is ambiguous
6 marisa-0.2.4-42.fc32.src.rpm
https://koji.fedoraproject.org/koji/taskinfo?taskID=40685286
Looks like the same issue as graphviz.
marisa-swig_wrap.cxx:1267:85: error: call of overloaded
'rb_define_virtual_variable(const char [21], VALUE (&)(...), NULL)' is
ambiguous
7 mlt-6.18.0-2.fc32.src.rpm
https://koji.fedoraproject.org/koji/taskinfo?taskID=40685147
nothing provides libpoppler.so.84()(64bit) needed by gdal-libs-2.3.2-14.fc32.x86_64
Looks like gdal needs rebuilding for poppler first.
8 nbdkit-1.17.5-1.fc32.src.rpm
https://koji.fedoraproject.org/koji/taskinfo?taskID=40688690
x86_64 test fails (ramdomly, perhaps), and armv7hl test seems to be hanging, out of my
hands.
9 openbabel-2.4.1-26.fc32.src.rpm
https://koji.fedoraproject.org/koji/taskinfo?taskID=40684400
Looks like the same issue as graphviz.
/builddir/build/BUILD/openbabel-openbabel-2-4-1/scripts/ruby/openbabel-ruby.cpp:1267:85:
error: call of overloaded 'rb_define_virtual_variable(const char [21], VALUE
(&)(...), NULL)' is ambiguous
10 rubygem-bootsnap-1.3.2-3.fc31.src.rpm
https://koji.fedoraproject.org/koji/taskinfo?taskID=40683765
Test suite fails.
/usr/share/rubygems/rubygems/core_ext/kernel_require.rb:72:in `require': no implicit
conversion of String into Integer (TypeError)
from /usr/share/rubygems/rubygems/core_ext/kernel_require.rb:72:in `require'
from
/builddir/build/BUILD/bootsnap-1.3.2/usr/share/gems/gems/bootsnap-1.3.2/lib/bootsnap/compile_cache/iseq.rb:1:in
`<top (required)>'
Note that build for ruby 2.6 seems successful:
https://koji.fedoraproject.org/koji/taskinfo?taskID=40691394
11 rubygem-curb-0.9.10-1.fc32.src.rpm
https://koji.fedoraproject.org/koji/taskinfo?taskID=40683311
Some test failure.
12 rubygem-ffi-1.10.0-3.fc31.src.rpm
https://koji.fedoraproject.org/koji/taskinfo?taskID=40683095
Failures:
1) String tests MemoryPointer#get_string returns a tainted string
Failure/Error: expect(str.tainted?).to be true
expected true
got false
# ./spec/ffi/string_spec.rb:22:in `block (2 levels) in <top (required)>'
2) String tests String returned by a method is tainted
Failure/Error: expect(str).to be_tainted
expected `"test".tainted?` to return true, got false
# ./spec/ffi/string_spec.rb:32:in `block (2 levels) in <top (required)>'
ruby 2.7 changed behavior for security model (almost deprecated security model),
some.tainted? always returns false.
https://github.com/ffi/ffi/commit/7ce0a9adee3212c0beac42ca7bd6b07aa97bdb90
By the way I tried the latest 1.12.1, however some test seems to fail:
https://koji.fedoraproject.org/koji/taskinfo?taskID=40683810
13 rubygem-ox-2.12.0-1.fc32.src.rpm
https://koji.fedoraproject.org/koji/taskinfo?taskID=40681637
Some test failure, I have not investigated deeply yet.
14 rubygem-qpid_messaging-1.39.0-4.fc31.src.rpm
DEBUG util.py:596: No matching package to install: 'qpid-cpp-client-devel'
DEBUG util.py:596: Not all dependencies satisfied
Looks line qpid-cpp is retired, someone needs taking care of depending package
maintainership.
15 rubygem-thin-1.7.2-11.fc31.src.rpm
https://koji.fedoraproject.org/koji/taskinfo?taskID=40654299
One test failure, I don't know well.
Failures:
1) Thin::Server should set lower maximum_connections size when too large
Failure/Error: @server.maximum_connections.should < 100_000
expected: < 100000
got: 100000
# ./spec/server_spec.rb:21:in `block (2 levels) in <top (required)>'
16 subversion-1.12.2-3.fc32.src.rpm
https://koji.fedoraproject.org/koji/taskinfo?taskID=40681152
Well, some linkage or macro expanding error??
/usr/share/rubygems/rubygems/core_ext/kernel_require.rb:92:in `require':
/builddir/build/BUILD/subversion-1.12.2/subversion/bindings/swig/ruby/libsvn_swig_ruby/.l
ibs/libsvn_swig_ruby-1.so.0: undefined symbol: assert -
/builddir/build/BUILD/subversion-1.12.2/subversion/bindings/swig/ruby/.ext/svn/ext/core.so
(LoadError)
17 weechat-2.4-4.fc32.src.rpm
https://koji.fedoraproject.org/koji/taskinfo?taskID=40681027
So, rebuilding srpm failed, I don't want to touch this.
DEBUG util.py:596: curl: (22) The requested URL returned error: 404 Not Found
DEBUG util.py:744: Child return code was: 22
12:37 a.m.
Mamoru TASAKA wrote on 2020/01/18 15:33:
14 rubygem-qpid_messaging-1.39.0-4.fc31.src.rpm
DEBUG util.py:596: No matching package to install: 'qpid-cpp-client-devel'
DEBUG util.py:596: Not all dependencies satisfied
Looks line qpid-cpp is retired, someone needs taking care of depending package
maintainership.
Link:
https://koji.fedoraproject.org/koji/taskinfo?taskID=40681069
Regards,
Mamoru
5:37 a.m.
Mamoru TASAKA wrote on 2020/01/18 15:33:
Thank you (and thanks to other people) for excellent works to bring
ruby 2.7 into Fedora 32. As Vít says, the rebuilt packages can be seen on:
https://koji.fedoraproject.org/koji/builds?inherited=0&tagID=17977&am...
92 packages were already rebuilt.
Current rebuild leftovers (with regards to library dependency) are:
Status update: currently leftovers are:
$ dnf repoquery --disablerepo=\* --enablerepo=koji-ruby27 --qf "%{sourcerpm}"
--whatrequires "libruby.so.2.6()(64bit)" | sort | uniq | cat -n
1 nbdkit-1.17.5-1.fc32.src.rpm
2 rubygem-curb-0.9.10-1.fc32.src.rpm
3 rubygem-ffi-1.10.0-3.fc31.src.rpm
4 rubygem-qpid_messaging-1.39.0-4.fc31.src.rpm
5 rubygem-thin-1.7.2-11.fc31.src.rpm
6 subversion-1.12.2-3.fc32.src.rpm
7 weechat-2.4-4.fc32.src.rpm
As rubygem-qpid_messaging (dependency retired) and weechat (source broken) cannot be
fixed
due to the issues completely unrelated to ruby27 change, I think now it is already
good time to get the rebuilt package merged into f32 buildroot.
Regards,
Mamoru
7:02 a.m.
Hi Mamoru,
Dne 20. 01. 20 v 12:37 Mamoru TASAKA napsal(a):
Mamoru TASAKA wrote on 2020/01/18 15:33:
>
> Thank you (and thanks to other people) for excellent works to bring
> ruby 2.7 into Fedora 32. As Vít says, the rebuilt packages can be
> seen on:
>
https://koji.fedoraproject.org/koji/builds?inherited=0&tagID=17977&am...
>
>
> 92 packages were already rebuilt.
>
> Current rebuild leftovers (with regards to library dependency) are:
>
Status update: currently leftovers are:
$ dnf repoquery --disablerepo=\* --enablerepo=koji-ruby27 --qf
"%{sourcerpm}" --whatrequires "libruby.so.2.6()(64bit)" | sort |
uniq
| cat -n
1 nbdkit-1.17.5-1.fc32.src.rpm
2 rubygem-curb-0.9.10-1.fc32.src.rpm
3 rubygem-ffi-1.10.0-3.fc31.src.rpm
4 rubygem-qpid_messaging-1.39.0-4.fc31.src.rpm
5 rubygem-thin-1.7.2-11.fc31.src.rpm
6 subversion-1.12.2-3.fc32.src.rpm
7 weechat-2.4-4.fc32.src.rpm
As rubygem-qpid_messaging (dependency retired) and weechat (source
broken) cannot be fixed
due to the issues completely unrelated to ruby27 change, I think now
it is already
good time to get the rebuilt package merged into f32 buildroot.
Thank you for the heavy lifting and of course all others who helped with
the rebuild. I have asked relengs to merge the side tag back to master:
https://pagure.io/releng/issue/9179
Vít
7:24 a.m.
Dne 20. 01. 20 v 14:02 Vít Ondruch napsal(a):
Hi Mamoru,
Dne 20. 01. 20 v 12:37 Mamoru TASAKA napsal(a):
> Mamoru TASAKA wrote on 2020/01/18 15:33:
>> Thank you (and thanks to other people) for excellent works to bring
>> ruby 2.7 into Fedora 32. As Vít says, the rebuilt packages can be
>> seen on:
>>
https://koji.fedoraproject.org/koji/builds?inherited=0&tagID=17977&am...
>>
>>
>> 92 packages were already rebuilt.
>>
>> Current rebuild leftovers (with regards to library dependency) are:
>>
> Status update: currently leftovers are:
>
> $ dnf repoquery --disablerepo=\* --enablerepo=koji-ruby27 --qf
> "%{sourcerpm}" --whatrequires "libruby.so.2.6()(64bit)" | sort |
uniq
> | cat -n
> 1 nbdkit-1.17.5-1.fc32.src.rpm
> 2 rubygem-curb-0.9.10-1.fc32.src.rpm
> 3 rubygem-ffi-1.10.0-3.fc31.src.rpm
> 4 rubygem-qpid_messaging-1.39.0-4.fc31.src.rpm
> 5 rubygem-thin-1.7.2-11.fc31.src.rpm
> 6 subversion-1.12.2-3.fc32.src.rpm
> 7 weechat-2.4-4.fc32.src.rpm
>
> As rubygem-qpid_messaging (dependency retired) and weechat (source
> broken) cannot be fixed
> due to the issues completely unrelated to ruby27 change, I think now
> it is already
> good time to get the rebuilt package merged into f32 buildroot.
Thank you for the heavy lifting and of course all others who helped with
the rebuild. I have asked relengs to merge the side tag back to master:
https://pagure.io/releng/issue/9179
I was told to merge the side tag via Bodhi, so here is the update:
https://bodhi.fedoraproject.org/updates/FEDORA-2020-223644200d
However I am not sure how it really gets merged, because it is in
Pending state now, whatever it means ...
Vít
Vít
_______________________________________________
ruby-sig mailing list -- ruby-sig(a)lists.fedoraproject.org
To unsubscribe send an email to ruby-sig-leave(a)lists.fedoraproject.org
Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives:
https://lists.fedoraproject.org/archives/list/ruby-sig@lists.fedoraprojec...
5:29 p.m.
----- Original Message -----
From: "Mamoru TASAKA" <mtasaka(a)fedoraproject.org>
To: ruby-sig(a)lists.fedoraproject.org
Sent: Monday, January 20, 2020 12:37:18 PM
Subject: Re: Ruby 2.7 - Mass rebuild
Mamoru TASAKA wrote on 2020/01/18 15:33:
>
> Thank you (and thanks to other people) for excellent works to bring
> ruby 2.7 into Fedora 32. As Vít says, the rebuilt packages can be seen on:
>
https://koji.fedoraproject.org/koji/builds?inherited=0&tagID=17977&am...
>
> 92 packages were already rebuilt.
>
> Current rebuild leftovers (with regards to library dependency) are:
>
Status update: currently leftovers are:
$ dnf repoquery --disablerepo=\* --enablerepo=koji-ruby27 --qf "%{sourcerpm}"
--whatrequires "libruby.so.2.6()(64bit)" | sort | uniq | cat -n
1 nbdkit-1.17.5-1.fc32.src.rpm
2 rubygem-curb-0.9.10-1.fc32.src.rpm
3 rubygem-ffi-1.10.0-3.fc31.src.rpm
I've created a PR \w rebase:
https://src.fedoraproject.org/rpms/rubygem-ffi/pull-request/2
I'll build it when the side-tag is pushed.
4 rubygem-qpid_messaging-1.39.0-4.fc31.src.rpm
5 rubygem-thin-1.7.2-11.fc31.src.rpm
6 subversion-1.12.2-3.fc32.src.rpm
7 weechat-2.4-4.fc32.src.rpm
As rubygem-qpid_messaging (dependency retired) and weechat (source broken)
cannot be fixed
due to the issues completely unrelated to ruby27 change, I think now it is
already
good time to get the rebuilt package merged into f32 buildroot.
Regards,
Mamoru
Regards,
--
Pavel Valena
Software Engineer, Red Hat
Brno, Czech Republic
6:58 p.m.
Vít Ondruch wrote on 2020/01/20 22:24:
Dne 20. 01. 20 v 14:02 Vít Ondruch napsal(a):
> Hi Mamoru,
>
>
> Dne 20. 01. 20 v 12:37 Mamoru TASAKA napsal(a):
>> Mamoru TASAKA wrote on 2020/01/18 15:33:
>>> Thank you (and thanks to other people) for excellent works to bring
>>> ruby 2.7 into Fedora 32. As Vít says, the rebuilt packages can be
>>> seen on:
>>>
https://koji.fedoraproject.org/koji/builds?inherited=0&tagID=17977&am...
>>>
>>>
>>> 92 packages were already rebuilt.
>>>
>>> Current rebuild leftovers (with regards to library dependency) are:
>>>
>> Status update: currently leftovers are:
>>
>> $ dnf repoquery --disablerepo=\* --enablerepo=koji-ruby27 --qf
>> "%{sourcerpm}" --whatrequires "libruby.so.2.6()(64bit)" |
sort | uniq
>> | cat -n
>> 1 nbdkit-1.17.5-1.fc32.src.rpm
>> 2 rubygem-curb-0.9.10-1.fc32.src.rpm
>> 3 rubygem-ffi-1.10.0-3.fc31.src.rpm
>> 4 rubygem-qpid_messaging-1.39.0-4.fc31.src.rpm
>> 5 rubygem-thin-1.7.2-11.fc31.src.rpm
>> 6 subversion-1.12.2-3.fc32.src.rpm
>> 7 weechat-2.4-4.fc32.src.rpm
>>
>> As rubygem-qpid_messaging (dependency retired) and weechat (source
>> broken) cannot be fixed
>> due to the issues completely unrelated to ruby27 change, I think now
>> it is already
>> good time to get the rebuilt package merged into f32 buildroot.
>
> Thank you for the heavy lifting and of course all others who helped with
> the rebuild. I have asked relengs to merge the side tag back to master:
>
> https://pagure.io/releng/issue/9179
I was told to merge the side tag via Bodhi, so here is the update:
https://bodhi.fedoraproject.org/updates/FEDORA-2020-223644200d
However I am not sure how it really gets merged, because it is in
Pending state now, whatever it means ...
Vít
Okay, now side-builds packages are merged into f32 buildroot (and newrepo
for f32 including ruby27 finished). (I updated the packages' entry of
the bodhi update Vít created to include new hivex and libguestfs, and
now bodhi pushed packages into f32)
Again thanks to all people for bringing ruby27 into Fedora 32.
Regards,
Mamoru
2:47 a.m.
Dne 21. 01. 20 v 1:58 Mamoru TASAKA napsal(a):
Vít Ondruch wrote on 2020/01/20 22:24:
>
> Dne 20. 01. 20 v 14:02 Vít Ondruch napsal(a):
>> Hi Mamoru,
>>
>>
>> Dne 20. 01. 20 v 12:37 Mamoru TASAKA napsal(a):
>>> Mamoru TASAKA wrote on 2020/01/18 15:33:
>>>> Thank you (and thanks to other people) for excellent works to bring
>>>> ruby 2.7 into Fedora 32. As Vít says, the rebuilt packages can be
>>>> seen on:
>>>>
https://koji.fedoraproject.org/koji/builds?inherited=0&tagID=17977&am...
>>>>
>>>>
>>>>
>>>> 92 packages were already rebuilt.
>>>>
>>>> Current rebuild leftovers (with regards to library dependency) are:
>>>>
>>> Status update: currently leftovers are:
>>>
>>> $ dnf repoquery --disablerepo=\* --enablerepo=koji-ruby27 --qf
>>> "%{sourcerpm}" --whatrequires "libruby.so.2.6()(64bit)" |
sort | uniq
>>> | cat -n
>>> 1 nbdkit-1.17.5-1.fc32.src.rpm
>>> 2 rubygem-curb-0.9.10-1.fc32.src.rpm
>>> 3 rubygem-ffi-1.10.0-3.fc31.src.rpm
>>> 4 rubygem-qpid_messaging-1.39.0-4.fc31.src.rpm
>>> 5 rubygem-thin-1.7.2-11.fc31.src.rpm
>>> 6 subversion-1.12.2-3.fc32.src.rpm
>>> 7 weechat-2.4-4.fc32.src.rpm
>>>
>>> As rubygem-qpid_messaging (dependency retired) and weechat (source
>>> broken) cannot be fixed
>>> due to the issues completely unrelated to ruby27 change, I think now
>>> it is already
>>> good time to get the rebuilt package merged into f32 buildroot.
>>
>> Thank you for the heavy lifting and of course all others who helped
>> with
>> the rebuild. I have asked relengs to merge the side tag back to master:
>>
>> https://pagure.io/releng/issue/9179
>
> I was told to merge the side tag via Bodhi, so here is the update:
>
> https://bodhi.fedoraproject.org/updates/FEDORA-2020-223644200d
>
> However I am not sure how it really gets merged, because it is in
> Pending state now, whatever it means ...
> Vít
Okay, now side-builds packages are merged into f32 buildroot (and newrepo
for f32 including ruby27 finished). (I updated the packages' entry of
the bodhi update Vít created to include new hivex and libguestfs, and
now bodhi pushed packages into f32)
Thank you for finishing this. The Koji was in bad shape yesterday
evening, so I had to leave prior finishing this.
Again thanks to all people for bringing ruby27 into Fedora 32.
Thank you and everybody.
And actually thanks to infrastructure guys for providing a way for
requesting and merging side tags by ourselves :)
Vít
6:25 a.m.
Mamoru TASAKA wrote on 2020/01/18 15:33:
Vít Ondruch wrote on 2020/01/17 0:00:
> Hi everybody,
>
> This years schedule does not give us too much freedom, because the
> Fedora mass rebuild is scheduled for the 22nd of January, which is in a
> few days. There are also some concerns with keyword arguments, but since
> I have no idea how problematic this might be, I decided to give it a go.
> It seems that nowadays, everybody is able to ask for side tag via
> `fedpkg` so I acquired one:
<snip>
Thank you (and thanks to other people) for excellent works to bring
ruby 2.7 into Fedora 32. As Vít says, the rebuilt packages can be seen on:
https://koji.fedoraproject.org/koji/builds?inherited=0&tagID=17977&am...
92 packages were already rebuilt.
Current rebuild leftovers (with regards to library dependency) are:
16 subversion-1.12.2-3.fc32.src.rpm
https://koji.fedoraproject.org/koji/taskinfo?taskID=40681152
Well, some linkage or macro expanding error??
/usr/share/rubygems/rubygems/core_ext/kernel_require.rb:92:in `require':
/builddir/build/BUILD/subversion-1.12.2/subversion/bindings/swig/ruby/libsvn_swig_ruby/.l
ibs/libsvn_swig_ruby-1.so.0: undefined symbol: assert -
/builddir/build/BUILD/subversion-1.12.2/subversion/bindings/swig/ruby/.ext/svn/ext/core.so
(LoadError)
For subversion, the reason of the above error which is related to ruby2.7
is found and I pushed the (maybe-tentative) solution to git.
Unfortunately now build of subversion for rawhide fails on some test,
only on i686 and armv7hl, which I believe is unrelated to ruby2.7 change
but is related to some other thing. (Note that builds on f31 is successful.)
So anyway I pushed build on rawhide so that subversion maintainer can
investigate the current status.
https://koji.fedoraproject.org/koji/buildinfo?buildID=1431029
Regards,
Mamoru
10:09 a.m.
Dne 21. 01. 20 v 0:29 Pavel Valena napsal(a):
----- Original Message -----
> From: "Mamoru TASAKA" <mtasaka(a)fedoraproject.org>
> To: ruby-sig(a)lists.fedoraproject.org
> Sent: Monday, January 20, 2020 12:37:18 PM
> Subject: Re: Ruby 2.7 - Mass rebuild
>
> Mamoru TASAKA wrote on 2020/01/18 15:33:
>> Thank you (and thanks to other people) for excellent works to bring
>> ruby 2.7 into Fedora 32. As Vít says, the rebuilt packages can be seen on:
>>
https://koji.fedoraproject.org/koji/builds?inherited=0&tagID=17977&am...
>>
>> 92 packages were already rebuilt.
>>
>> Current rebuild leftovers (with regards to library dependency) are:
>>
> Status update: currently leftovers are:
>
> $ dnf repoquery --disablerepo=\* --enablerepo=koji-ruby27 --qf
"%{sourcerpm}"
> --whatrequires "libruby.so.2.6()(64bit)" | sort | uniq | cat -n
> 1 nbdkit-1.17.5-1.fc32.src.rpm
> 2 rubygem-curb-0.9.10-1.fc32.src.rpm
> 3 rubygem-ffi-1.10.0-3.fc31.src.rpm
I've created a PR \w rebase:
https://src.fedoraproject.org/rpms/rubygem-ffi/pull-request/2
I'll build it when the side-tag is pushed.
The rubygem-ffi is built now. Thx Pavel.
Vít
> 4 rubygem-qpid_messaging-1.39.0-4.fc31.src.rpm
> 5 rubygem-thin-1.7.2-11.fc31.src.rpm
> 6 subversion-1.12.2-3.fc32.src.rpm
> 7 weechat-2.4-4.fc32.src.rpm
>
> As rubygem-qpid_messaging (dependency retired) and weechat (source broken)
> cannot be fixed
> due to the issues completely unrelated to ruby27 change, I think now it is
> already
> good time to get the rebuilt package merged into f32 buildroot.
>
> Regards,
> Mamoru
Regards,
11:04 a.m.
Dne 18. 01. 20 v 7:33 Mamoru TASAKA napsal(a):
15 rubygem-thin-1.7.2-11.fc31.src.rpm
https://koji.fedoraproject.org/koji/taskinfo?taskID=40654299
One test failure, I don't know well.
Failures:
1) Thin::Server should set lower maximum_connections size when too
large
Failure/Error: @server.maximum_connections.should < 100_000
expected: < 100000
got: 100000
# ./spec/server_spec.rb:21:in `block (2 levels) in <top (required)>'
Looking into this, I think there is (Koji) Kernel update behind this.
The last successful rubygem-thin build [1] was build on:
~~~
DEBUG buildroot.py:488: kernel version == 5.0.6-200.fc29.x86_64
~~~
While the build failures started around mid of December according to
Koschei [2] (although the missing logs are PITA). This [3] is fresh F31
scratch build which fails the same way as Rawhide.
So looking into details, the test in Thin is doing the following:
~~~
@server.maximum_connections = 100_000
~~~
This propagates into EventMachine [4] and calls `{g,s}etrlimit
(RLIMIT_NOFILE, &rlim);`. On my machine, this is the NOFILE value:
~~~
$ prlimit | grep NOFILE
NOFILE max number of open files 1024 524288 soubory
~~~
Apparently, this is way above the limit the test case expects. My
conclusion is that this had to be changed in Kernel, although it could
be even systemd. Not sure.
Nevertheless, the test is naive and it should be tweaked IMO.
BTW the test passes with `--new-chroot` mock option :)
Vít
[1] https://koji.fedoraproject.org/koji/buildinfo?buildID=1340022
[2] https://koschei.fedoraproject.org/package/rubygem-thin
[3] https://koji.fedoraproject.org/koji/taskinfo?taskID=40873714
[4] https://github.com/eventmachine/eventmachine/blob/master/ext/em.cpp#L300
[5] https://github.com/macournoyer/thin/blob/master/spec/server_spec.rb#L21
5:45 a.m.
rubygem-thin is build now. Similar medicine will be needed for older
branches, so just in case:
https://src.fedoraproject.org/rpms/rubygem-thin/c/4cd6bb0fe7adcefd68c63c3...
https://src.fedoraproject.org/rpms/rubygem-thin/c/2ede272fa463e1d37a06a61...
Vít
Dne 22. 01. 20 v 18:04 Vít Ondruch napsal(a):
Dne 18. 01. 20 v 7:33 Mamoru TASAKA napsal(a):
>
> 15 rubygem-thin-1.7.2-11.fc31.src.rpm
> https://koji.fedoraproject.org/koji/taskinfo?taskID=40654299
> One test failure, I don't know well.
> Failures:
> 1) Thin::Server should set lower maximum_connections size when too
> large
> Failure/Error: @server.maximum_connections.should < 100_000
> expected: < 100000
> got: 100000
> # ./spec/server_spec.rb:21:in `block (2 levels) in <top
> (required)>'
>
>
Looking into this, I think there is (Koji) Kernel update behind this.
The last successful rubygem-thin build [1] was build on:
~~~
DEBUG buildroot.py:488: kernel version == 5.0.6-200.fc29.x86_64
~~~
While the build failures started around mid of December according to
Koschei [2] (although the missing logs are PITA). This [3] is fresh
F31 scratch build which fails the same way as Rawhide.
So looking into details, the test in Thin is doing the following:
~~~
@server.maximum_connections = 100_000
~~~
This propagates into EventMachine [4] and calls `{g,s}etrlimit
(RLIMIT_NOFILE, &rlim);`. On my machine, this is the NOFILE value:
~~~
$ prlimit | grep NOFILE
NOFILE max number of open files 1024 524288 soubory
~~~
Apparently, this is way above the limit the test case expects. My
conclusion is that this had to be changed in Kernel, although it could
be even systemd. Not sure.
Nevertheless, the test is naive and it should be tweaked IMO.
BTW the test passes with `--new-chroot` mock option :)
Vít
[1] https://koji.fedoraproject.org/koji/buildinfo?buildID=1340022
[2] https://koschei.fedoraproject.org/package/rubygem-thin
[3] https://koji.fedoraproject.org/koji/taskinfo?taskID=40873714
[4]
https://github.com/eventmachine/eventmachine/blob/master/ext/em.cpp#L300
[5]
https://github.com/macournoyer/thin/blob/master/spec/server_spec.rb#L21
_______________________________________________
ruby-sig mailing list -- ruby-sig(a)lists.fedoraproject.org
To unsubscribe send an email to ruby-sig-leave(a)lists.fedoraproject.org
Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives:
https://lists.fedoraproject.org/archives/list/ruby-sig@lists.fedoraprojec...
1548
days inactive
1555
days old
ruby-sig@lists.fedoraproject.org
13 comments
4 participants
participants (4)
-
Dan Allen -
Mamoru TASAKA -
Pavel Valena -
Vít Ondruch