ethon: [BUG] Segmentation fault at ethon-0.16.0/lib/ethon/easy/operations.rb:30

I get the following seg fault:

c:0153 p:---- s:1033 e:001032 CFUNC  :easy_perform
c:0152 p:0007 s:1028 e:001027 METHOD /Users/valentino/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/ethon-0.16.0/lib/ethon/easy/operations.rb:30
c:0151 p:0013 s:1024 e:001023 METHOD /Users/valentino/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/typhoeus-1.4.0/lib/typhoeus/request/operations.rb:16
c:0150 p:0025 s:1019 e:001018 METHOD /Users/valentino/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/typhoeus-1.4.0/lib/typhoeus/request/cacheable.rb:18
c:0149 p:0018 s:1014 e:001013 METHOD /Users/valentino/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/typhoeus-1.4.0/lib/typhoeus/request/block_connection.rb:31
c:0148 p:0033 s:1010 e:001009 METHOD /Users/valentino/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/typhoeus-1.4.0/lib/typhoeus/request/stubbable.rb:25
c:0147 p:0012 s:1005 e:001004 METHOD /Users/valentino/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/typhoeus-1.4.0/lib/typhoeus/request/before.rb:26
c:0146 p:0028 s:1001 e:001000 METHOD /Users/valentino/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/typhoeus-1.4.0/lib/typhoeus/adapters/faraday.rb:82
c:0145 p:0012 s:0996 e:000995 METHOD /Users/valentino/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/typhoeus-1.4.0/lib/typhoeus/adapters/faraday.rb:72
c:0144 p:0012 s:0991 e:000990 METHOD /Users/valentino/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/faraday-1.10.3/lib/faraday/rack_builder.rb:154
c:0143 p:0040 s:0985 e:000984 METHOD /Users/valentino/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/faraday-1.10.3/lib/faraday/connection.rb:516

# -- Ruby level backtrace information ----------------------------------------
/Users/valentino/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/puma-6.0.2/lib/puma/thread_pool.rb:147:in `block in spawn_thread'
/Users/valentino/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/puma-6.0.2/lib/puma/server.rb:232:in `block in run'
/Users/valentino/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/puma-6.0.2/lib/puma/server.rb:429:in `process_client'
/Users/valentino/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/puma-6.0.2/lib/puma/request.rb:92:in `handle_request'
/Users/valentino/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/puma-6.0.2/lib/puma/thread_pool.rb:340:in `with_force_shutdown'
/Users/valentino/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/puma-6.0.2/lib/puma/request.rb:93:in `block in handle_request'
/Users/valentino/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/puma-6.0.2/lib/puma/configuration.rb:268:in `call'
/Users/valentino/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/railties-6.1.7.4/lib/rails/engine.rb:539:in `call'
/Users/valentino/.rbenv/versions/3.2.2/lib/ruby/gems/3.2.0/gems/rack-cors-1.1.1/lib/rack/cors.rb:100:in `call'

# DiagnosticReports
  "vmRegionInfo" : "0xdac11530 is not in any region.  Bytes before following region: 628992720\n      REGION TYPE                    START - END         [ VSIZE] PRT\/MAX SHRMOD  REGION DETAIL\n      UNUSED SPACE AT START\n--->  \n      __TEXT                      1003ec000-1003f0000    [   16K] r-x\/r-x SM=COW  ...s\/USER\/*\/ruby",
  "exception" : {"codes":"0x0000000000000105, 0x00000000dac11530","rawCodes":[261,3670086960],"type":"EXC_BAD_ACCESS","signal":"SIGABRT","subtype":"UNKNOWN_0x105 at 0x00000000dac11530"},
  "vmregioninfo" : "0xdac11530 is not in any region.  Bytes before following region: 628992720\n      REGION TYPE                    START - END         [ VSIZE] PRT\/MAX SHRMOD  REGION DETAIL\n      UNUSED SPACE AT START\n--->  \n      __TEXT                      1003ec000-1003f0000    [   16K] r-x\/r-x SM=COW  ...s\/USER\/*\/ruby",
  "asi" : {"CoreFoundation":["*** multi-threaded process forked ***"],"libsystem_c.dylib":["crashed on child side of fork pre-exec"]},
  "extMods" : {"caller":{"thread_create":0,"thread_set_state":0,"task_for_pid":0},"system":{"thread_create":0,"thread_set_state":0,"task_for_pid":0},"targeted":{"thread_create":0,"thread_set_state":0,"task_for_pid":0},"warnings":0},
  "faultingThread" : 7,

Chip: Apple M2 Max OS: Ventura 13.4.1

About this issue

  • Original URL
  • State: open
  • Created 3 years ago
  • Reactions: 12
  • Comments: 15

Most upvoted comments

Same here: macOS M1 Sonoma 14.1.2, Ruby 3.2.2 and Curl 8.1.2 (x86_64-apple-darwin23.0)

We we’re facing this segfault using chewy. Until a solution is found we patched the net adapter like so:

  Chewy.settings = {
    host: 'localhost:9200'
  }.tap do |config|
      config[:transport_options] = {
        proc: ->(f) { f.adapter :net_http }
      }
  end

not recommended to use it in prod.

I’m experiencing the same crash on M2 Mac using native ARM architecture when running the code in Resque, which forks the process for every job.

-- Control frame information -----------------------------------------------
c:0077 p:---- s:0408 e:000407 CFUNC  :easy_perform
c:0076 p:0062 s:0403 E:000390 METHOD /Users/timherby/.rvm/gems/ruby-2.7.7/gems/ethon-0.16.0/lib/ethon/easy/operations.rb:30

I originally thought it was due to ethon/curls/functions.rb and ethon/curls/settings.rb getting initialized, but I have confirmed it still happens even if I use require:false for both Typhoeus and Ethon and they are not loaded in the parent process (also confirmed).

I’ve also reinstalled curl, ffi, libffi, and openssl, with no luck, it’s still crashing. I’ve linked the thread above where I’m looking to abandon the use of Typhoeus and Ethon as it appears there is no energy behind fixing these segmentation faults that are plaguing the library on new macs: https://github.com/search?q=repo%3Atyphoeus%2Fethon+Segmentation+Fault&type=issues

I’d love to help troubleshoot if anyone wants to reach out and collaborate on this.

I’m also seeing the issue on macOS 13.6.1 (22G313), Ruby 3.2.2 and curl 8.4.0. This is an Intel Mac, so I don’t think this issue is exclusive to M1.

Seeing this on macOS 14.1.1 (23B81). Intel. Ruby 3.2.2, curl 8.4.0

➜ curl --version
curl 8.4.0 (x86_64-apple-darwin23.0.0) libcurl/8.4.0 (SecureTransport) OpenSSL/3.2.0 zlib/1.2.12 brotli/1.1.0 zstd/1.5.5 libidn2/2.3.4 libssh2/1.11.0 nghttp2/1.58.0 librtmp/2.3 OpenLDAP/2.6.6
Release-Date: 2023-10-11
Protocols: dict file ftp ftps gopher gophers http https imap imaps ldap ldaps mqtt pop3 pop3s rtmp rtsp scp sftp smb smbs smtp smtps telnet tftp
Features: alt-svc AsynchDNS brotli GSS-API HSTS HTTP2 HTTPS-proxy IDN IPv6 Kerberos Largefile libz MultiSSL NTLM SPNEGO SSL threadsafe TLS-SRP UnixSockets zstd

Update

Disabling Spring: DISABLE_SPRING=true solved it for me.

Not sure if this would help anyone else. I was running into segfaults in this way as well, while trying to run it from MacOS Monterey using the Apple M1 Pro chip. I discovered that if I changed my puma settings to run in standalone mode (PUMA_WORKERS=0) the application would run without issues. Clearly, not a solution to the problem, but at least a workaround for testing and debugging purposes. In my team other people told me they have build ruby using Rosetta and that has solved the issue for them as well.

I started getting this only after updating my system to macOS 13.5 (22G74) (from 13.4), M1, Ruby 3.2.2

Notably it only seems to happen (or at least I’ve only caught it) when running our test suite.

When I disable parallelization entirely (config.active_support.test_parallelization_threshold = some_arbitrarily_high_value) I do not encounter this crash, of course.

Same here; macOS 13.4.1 (22F82) M1. Ruby 3.0.3.

Seeing this on macOS 13.4 (22F66). Intel. Ruby 2.7.6 and ruby 3.2.2. curl 7.88.1