[erlang-bugs] bus error & segfaults on OS X Lion 10.7.3 after repeated crypto:md5_init

[Filipe David Manana]

It affects the R14 series as well.
There's a gdb thread backtrace dump at:

https://gist.github.com/1199903

which shows a memory access error originating from the crypto NIF (thread 13)

On Wed, Apr 4, 2012 at 3:44 PM, Dave Cottlehuber <dave@REDACTED> wrote:
> Hi,
>
> Using homebrewed Erlang/OTP R15B & R15B01 on Mac OS X Lion 10.7.3
> (11D50b) only, I'm seeing bus errors or segfaults after
> crypto:md5_init() after running this continually
> https://gist.github.com/2219748
>
> Eshell V5.9.1  (abort with ^G)
> 1> Erlang R15B01 (erts-5.9.1) [source] [64-bit] [smp:8:8]
> [async-threads:0] [hipe] [kernel-poll:false]
>
> [... many times .... ]
>
> Eshell V5.9.1  (abort with ^G)
> 1> [1]    94307 segmentation fault  erl -run erlbork
>
> In some cases it takes a couple of hours to trigger the failure.
>
> lldb stacktrace for R15B01 http://friendpaste.com/sQklBG4PuaVl4lwwkGu49
> lldb stacktrace for R15B     http://friendpaste.com/4wmvBSjatl3cemkRgSZlkp
>
> The original error is apparent during running CouchDB make check.
> Thanks to fdmanana, benoitc, davisp for helping with this on #couchdb as always.
>
> Let me know if I can help further.
>
> A+
> Dave
> _______________________________________________
> erlang-bugs mailing list
> erlang-bugs@REDACTED
> http://erlang.org/mailman/listinfo/erlang-bugs

-- 
Filipe David Manana,

"Reasonable men adapt themselves to the world.
 Unreasonable men adapt the world to themselves.
 That's why all progress depends on unreasonable men."