initialize/fork crash in macOS 10.13

initialize/fork crash in macOS 10.13

[Jeffrey Carl Faden]

According to this post...
http://www.sealiesoftware.com/blog/archive/2017/6/5/Objective-C_and_fork_in_macOS_1013.html

The rules around initialize and fork() have changed in macOS 10.13, which is coming out in a month or so. I tried using Unicorn today to run a unicorn_rails instance, and got this as part of my traceback:

objc[95737]: +[__NSPlaceholderDictionary initialize] may have been in progress in another thread when fork() was called. We cannot safely call it or ignore it in the fork() child process. Crashing instead. Set a breakpoint on objc_initializeAfterForkError to debug.
E, [2017-08-04T08:35:28.148339 #95531] ERROR -- : reaped #<Process::Status: pid 95736 SIGABRT (signal 6)> worker=1

I'm not sure if this is something downstream, but I thought I'd bring it to your attention. My temporary solution is to preface my commands with OBJC_DISABLE_INITIALIZE_FORK_SAFETY=YES.

Jeffrey

Re: initialize/fork crash in macOS 10.13

[Eric Wong]

Jeffrey Carl Faden <jeffreyatw@gmail.com> wrote:
> According to this post...
> http://www.sealiesoftware.com/blog/archive/2017/6/5/Objective-C_and_fork_in_macOS_1013.html
> 
> The rules around initialize and fork() have changed in macOS
> 10.13, which is coming out in a month or so. I tried using
> Unicorn today to run a unicorn_rails instance, and got this as
> part of my traceback:

This affects other Ruby programs which fork().  I suggest
reporting to ruby-core @ https://bugs.ruby-lang.org/ and having
somebody with macOS knowledge look at this(*).

I don't even know how ObjC factors into this.

> I'm not sure if this is something downstream, but I thought
> I'd bring it to your attention. My temporary solution is to
> preface my commands with
> OBJC_DISABLE_INITIALIZE_FORK_SAFETY=YES.

Thanks for documenting this for other users, at least.
I can't support macOS since I'm not allowed to know
everything about it.



(*) As a smug GNU/Linux user, I get to know EVERYTHING about how
    my system works from the kernel and up, for free!
    And when I do have a question I can't answer on my own;
    all I need to do is email the right people and lists.
    No NDAs or registration of any sort to deal with :>

Re: initialize/fork crash in macOS 10.13

[Eric Wong]

Btw, this seems to be resolved in Ruby trunk and there's a
workaround documented for existing Rubies:

	https://bugs.ruby-lang.org/issues/14009

Obviously I can't test it but it seems to be working for some people.

In case ruby-lang goes down, you can also find it from the
public-inbox mirror of ruby-core:

	https://public-inbox.org/ruby-core/?q=%22Bug%20%2314009%22&x=t