Bug 1235245 - rubygem-celluloid does not build with recent glibc
Summary: rubygem-celluloid does not build with recent glibc
Keywords:
Status: CLOSED RAWHIDE
Alias: None
Product: Fedora
Classification: Fedora
Component: glibc
Version: 23
Hardware: Unspecified
OS: Unspecified
unspecified
unspecified
Target Milestone: ---
Assignee: Torvald Riegel
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2015-06-24 11:50 UTC by Vít Ondruch
Modified: 2016-11-24 12:14 UTC (History)
9 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2016-11-24 12:14:18 UTC
Type: Bug
Embargoed:


Attachments (Terms of Use)
ABRT's core dump (3.57 MB, application/x-gzip)
2015-07-07 14:01 UTC, Vít Ondruch
no flags Details

Description Vít Ondruch 2015-06-24 11:50:44 UTC
Description of problem:
I am currently trying to update rubygem-celluloid [1, 2] to version 1.16, but I have hard times to pass its test suite with recent glibc, since it typically hags at some point.

For the test purposes, I downgraded my glic back to glibc-2.21.90-11.fc23.x86_64.rpm and with this older version, the test suite passed just fine every time.

I suspect that it hangs due to changes announced in [3].


Version-Release number of selected component (if applicable):
glibc-2.21.90-17.fc23

How reproducible:
Randomly


Steps to Reproduce:
1.
2.
3.

Actual results:


Expected results:


Additional info:


[1] https://admin.fedoraproject.org/pkgdb/package/rubygem-celluloid/
[2] https://github.com/celluloid/celluloid
[3] https://lists.fedoraproject.org/pipermail/devel/2015-May/210772.html

Comment 1 Vít Ondruch 2015-06-24 12:39:36 UTC
Just FTR, I needed 4 attempts to build the package for F23. Poor souls who need to use this package (it is in Ruby on Rails dependency chain, so there is chance somebody will need it).

Comment 2 Siddhesh Poyarekar 2015-07-01 16:24:46 UTC
Thanks, could you please isolate a test case (preferrably in C) that I can use for debugging?

Comment 3 Vít Ondruch 2015-07-07 10:27:21 UTC
I wish it was the simple. But probably fixing bug 1229659 could help resolve this as well.

Comment 4 Vít Ondruch 2015-07-07 10:49:36 UTC
Adding this file into the test suite:

$ cat spec/glibc_spec.rb 
require 'spec_helper'

describe "a Celluloid Actor using Fibers", actor_system: :global do
  class ExampleCrash < StandardError
    attr_accessor :foo
  end

  let(:actor_class) { ExampleActorClass.create(Celluloid, Celluloid::TaskFiber) }

  it "supports synchronous calls" do
    actor = actor_class.new "Troy McClure"
    actor.greet.should eq("Hi, I'm Troy McClure")
  end
end

and executing it as:

$ while true; do rspec2 -fd spec/glibc_spec.rb; done

This triggers sometimes:

Run options: include {:focus=>true}

All examples were filtered out; ignoring {:focus=>true}

a Celluloid Actor using Fibers
  supports synchronous calls

Finished in 0.00494 seconds
1 example, 0 failures
[BUG] pthread_cond_destroy: Device or resource busy (EBUSY)
ruby 2.2.2p95 (2015-04-13 revision 50295) [x86_64-linux]

[BUG] Segmentation fault at 0x00000000000030
ruby 2.2.2p95 (2015-04-13 revision 50295) [x86_64-linux]

Segmentation fault (core dumped)Run options: include {:focus=>true}

All examples were filtered out; ignoring {:focus=>true}

a Celluloid Actor using Fibers
  supports synchronous calls

Finished in 0.00494 seconds
1 example, 0 failures
[BUG] pthread_cond_destroy: Device or resource busy (EBUSY)
ruby 2.2.2p95 (2015-04-13 revision 50295) [x86_64-linux]

[BUG] Segmentation fault at 0x00000000000030
ruby 2.2.2p95 (2015-04-13 revision 50295) [x86_64-linux]

Segmentation fault (core dumped)


Which seems to be similar to bug 1229659, but if you let it run long enough, ignoring the errors above, it may get stuck finally.

Comment 5 Carlos O'Donell 2015-07-07 11:39:46 UTC
Torvald,

Could you have a look and see what the ruby gem is doing under the hood? I usually use strace orlltrace to see what calls the gem is making and see if the actual sequence of calls makes sense. Remember that Fedora Rawhide is basically upstream master, so you can build upstream master if you want and run ruby with it.

Comment 6 Vít Ondruch 2015-07-07 14:01:06 UTC
Created attachment 1049325 [details]
ABRT's core dump

This might be hopefully useful for analysis.

Comment 7 Jan Kurik 2015-07-15 13:55:17 UTC
This bug appears to have been reported against 'rawhide' during the Fedora 23 development cycle.
Changing version to '23'.

(As we did not run this process for some time, it could affect also pre-Fedora 23 development
cycle bugs. We are very sorry. It will help us with cleanup during Fedora 23 End Of Life. Thank you.)

More information and reason for this action is here:
https://fedoraproject.org/wiki/BugZappers/HouseKeeping/Fedora23

Comment 8 Torvald Riegel 2016-11-12 00:25:31 UTC
Current Rawhide contains a new condition variable implementation.  It would be nice if someone could test with this new implementation.  Thanks.

Comment 9 Vít Ondruch 2016-11-14 14:54:28 UTC
Testing with glibc-2.24.90-13.fc26.i686, I did 10 builds in Koji and no SEGV/hangs anymore. I run the test case from comment 4 for several minutest without issues.

Comment 10 Fedora End Of Life 2016-11-24 11:57:15 UTC
This message is a reminder that Fedora 23 is nearing its end of life.
Approximately 4 (four) weeks from now Fedora will stop maintaining
and issuing updates for Fedora 23. It is Fedora's policy to close all
bug reports from releases that are no longer maintained. At that time
this bug will be closed as EOL if it remains open with a Fedora  'version'
of '23'.

Package Maintainer: If you wish for this bug to remain open because you
plan to fix it in a currently maintained version, simply change the 'version' 
to a later Fedora version.

Thank you for reporting this issue and we are sorry that we were not 
able to fix it before Fedora 23 is end of life. If you would still like 
to see this bug fixed and are able to reproduce it against a later version 
of Fedora, you are encouraged  change the 'version' to a later Fedora 
version prior this bug is closed as described in the policy above.

Although we aim to fix as many bugs as possible during every release's 
lifetime, sometimes those efforts are overtaken by events. Often a 
more recent Fedora release includes newer upstream software that fixes 
bugs or makes them obsolete.

Comment 11 Florian Weimer 2016-11-24 12:14:18 UTC
(In reply to Vít Ondruch from comment #9)
> Testing with glibc-2.24.90-13.fc26.i686, I did 10 builds in Koji and no
> SEGV/hangs anymore. I run the test case from comment 4 for several minutest
> without issues.

Closing based on this feedback.


Note You need to log in before you can comment on or make changes to this bug.