[#68478] Looking for MRI projects for Ruby Google Summer of Code 2015 — Tony Arcieri <bascule@...>
Hi ruby-core,
10 messages
2015/03/10
[#68480] Re: Looking for MRI projects for Ruby Google Summer of Code 2015
— SASADA Koichi <ko1@...>
2015/03/10
I have.
[#68549] Re: Looking for MRI projects for Ruby Google Summer of Code 2015
— SASADA Koichi <ko1@...>
2015/03/17
I sent several ideas on previous, mail, but they are seems rejected?
[#68493] [Ruby trunk - Feature #10532] [PATCH] accept_nonblock supports "exception: false" — nobu@...
Issue #10532 has been updated by Nobuyoshi Nakada.
5 messages
2015/03/11
[#68503] Re: [Ruby trunk - Feature #10532] [PATCH] accept_nonblock supports "exception: false"
— Eric Wong <normalperson@...>
2015/03/12
Committed as r49948.
[#68504] Re: [Ruby trunk - Feature #10532] [PATCH] accept_nonblock supports "exception: false"
— Nobuyoshi Nakada <nobu@...>
2015/03/12
On 2015/03/12 12:08, Eric Wong wrote:
[#68506] Seven stacks (and two questions) — Jakub Trzebiatowski <jaktrze1@...>
The Ruby Hacking Guide says that Ruby has=E2=80=A6 seven stacks. Is it =
5 messages
2015/03/12
[#68520] Possible regression in 2.1 and 2.2 in binding when combined with delegate? — Joe Swatosh <joe.swatosh@...>
# The following code
3 messages
2015/03/14
[#68604] GSOC project Cross-thread Fiber support — surya pratap singh raghuvanshi <oshosurya@...>
- *hi i am a third year computer science student interested in working
6 messages
2015/03/22
[#68606] Re: GSOC project Cross-thread Fiber support
— Tony Arcieri <bascule@...>
2015/03/22
Hi Surya,
[#68619] Re: GSOC project Cross-thread Fiber support
— surya pratap singh raghuvanshi <oshosurya@...>
2015/03/23
hi tony,
[ruby-core:68628] [Ruby trunk - Bug #10998] TestIO#test_seek fails in chroot with Linux 3.19
From:
v.ondruch@...
Date:
2015-03-24 13:28:33 UTC
List:
ruby-core #68628
Issue #10998 has been updated by Vit Ondruch.
Backport changed from 2.0.0: UNKNOWN, 2.1: UNKNOWN, 2.2: UNKNOWN to 2.0.0: UNKNOWN, 2.1: UNKNOWN, 2.2: REQUIRED
The fix is passing on Fedora builders [1]. Thank you.
[1] http://koji.fedoraproject.org/koji/taskinfo?taskID=9310557
----------------------------------------
Bug #10998: TestIO#test_seek fails in chroot with Linux 3.19
https://bugs.ruby-lang.org/issues/10998#change-51929
* Author: Vit Ondruch
* Status: Closed
* Priority: Normal
* Assignee:
* ruby -v: ruby 2.2.1p85 (2015-02-26 revision 49769) [x86_64-linux]
* Backport: 2.0.0: UNKNOWN, 2.1: UNKNOWN, 2.2: REQUIRED
----------------------------------------
Recently, Ruby builds driven by CI started to fail on Fedora builders [1] with following error [6]:
~~~
2) Error:
TestIO#test_seek:
Errno::ENOENT: No such file or directory - no matching entry
/builddir/build/BUILD/ruby-2.2.1/test/ruby/test_io.rb:1711:in `fsname'
/builddir/build/BUILD/ruby-2.2.1/test/ruby/test_io.rb:1711:in `can_seek_data'
/builddir/build/BUILD/ruby-2.2.1/test/ruby/test_io.rb:1752:in `block (2 levels) in test_seek'
/builddir/build/BUILD/ruby-2.2.1/lib/open-uri.rb:36:in `open'
/builddir/build/BUILD/ruby-2.2.1/lib/open-uri.rb:36:in `open'
/builddir/build/BUILD/ruby-2.2.1/test/ruby/test_io.rb:1751:in `block in test_seek'
/builddir/build/BUILD/ruby-2.2.1/test/ruby/test_io.rb:1470:in `make_tempfile'
/builddir/build/BUILD/ruby-2.2.1/test/ruby/test_io.rb:1728:in `test_seek'
3) Error:
TestIO#test_seek_symwhence:
Errno::ENOENT: No such file or directory - no matching entry
/builddir/build/BUILD/ruby-2.2.1/test/ruby/test_io.rb:1711:in `fsname'
/builddir/build/BUILD/ruby-2.2.1/test/ruby/test_io.rb:1711:in `can_seek_data'
/builddir/build/BUILD/ruby-2.2.1/test/ruby/test_io.rb:1800:in `block (2 levels) in test_seek_symwhence'
/builddir/build/BUILD/ruby-2.2.1/lib/open-uri.rb:36:in `open'
/builddir/build/BUILD/ruby-2.2.1/lib/open-uri.rb:36:in `open'
/builddir/build/BUILD/ruby-2.2.1/test/ruby/test_io.rb:1799:in `block in test_seek_symwhence'
/builddir/build/BUILD/ruby-2.2.1/test/ruby/test_io.rb:1470:in `make_tempfile'
/builddir/build/BUILD/ruby-2.2.1/test/ruby/test_io.rb:1781:in `test_seek_symwhence'
~~~
We tracked down the issue and we believe that it is due to upgrade of Kernel from 3.17.8 to 3.19.1. In recent kernel, this commit [2] influences content of mtab in chroot (which we are using to setup our build roots). This is content of /etc/mtab using 3.17 kernel [3] and this is with 3.19 kernel [4]. As you can see, with recent kernel, there is no trace of root directory and hence the fsname method [5] can't find the underlying filesystem and the test case fails.
Is there any chance identify the FS by different/more reliable means.
BTW As I understand after discussion with colleagues, I believe that the /etc/mtab is kept around just for backward compatibility and there are several parties which would like to see this file dead entirely. It is currently just symlink anyway. So it might be just about time to stop using it.
[1] http://koschei.cloud.fedoraproject.org/package/ruby
[2] https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=9d4d6574
[3] http://paste.fedoraproject.org/201571/42712927/
[4] http://paste.fedoraproject.org/201618/14271327/
[5] https://github.com/ruby/ruby/blob/trunk/ext/-test-/file/fs.c#L14
[6] https://kojipkgs.fedoraproject.org/work/tasks/3012/9303012/build.log
--
https://bugs.ruby-lang.org/