[erlang-questions] OTP-10984

Trevor Woollacott trevorw@REDACTED
Thu Dec 5 09:41:09 CET 2013

Thanks for the explanation


From: garazdawi@REDACTED [mailto:garazdawi@REDACTED] On Behalf Of Lukas
Sent: 04 December 2013 12:24
To: Trevor Woollacott
Cc: erlang-questions
Subject: Re: [erlang-questions] OTP-10984


It was a while ago so I don't remember the exact details, but I think that
in order to recreate the bug you have to do many many smallish (10-20 bytes)
writes to a file with a small (1-2 ms) delayed_write timer. You can
theoretically also get unlucky with larger delayed_write values, but I have
yet to see that actually happen. 


Also fyi this race was introduced in the R15B release of Erlang/OTP.




On Wed, Dec 4, 2013 at 11:00 AM, Trevor Woollacott
<trevorw@REDACTED <mailto:trevorw@REDACTED> > wrote:



The release notes for Erlang R16B01 have the following fix:

   OTP-10984  Fixed a race condition when using delayed_write when writing

              to a file which would cause the same data to be written

              multiple times.


Can anyone please describe under what circumstances the above race condition
will occur when using delayed_write?


Kind regards,

erlang-questions mailing list
erlang-questions@REDACTED <mailto:erlang-questions@REDACTED> 


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://erlang.org/pipermail/erlang-questions/attachments/20131205/78e2391f/attachment.htm>

More information about the erlang-questions mailing list