On 11/04/17 16:42, Edward Burns wrote:
>>>>>> On Fri, 24 Mar 2017 09:51:38 +0000, Mark Thomas <markt_at_apache.org> said:
> SD> Concurrency is hard, and this is exactly the sort of concurrency issue
> SD> that Java EE is supposed to solve, we should not be punting it onto
> SD> application developers when it is trivial for us to make sure the
> SD> containers do the correct thing.
>
> MT> +1
>
> MT> I couldn't agree more with everything Stuart has written.
>
>>>>>> On Mon, 10 Apr 2017 12:09:36 -0700, Edward Burns <edward.burns_at_oracle.com> said:
>
> EB> Shing-wai looked into the GlassFish implementation and suggested we
> EB> simply leave the issue in the JIRA and look to resolve it in a future
> EB> release. I will take this option.
>
> In case this was not clear, I also agree with this resolution of leaving
> the issue in JIRA for 4.0.
I'm not sure why this is being pushed out to the next spec version.
Isn't this as simple as making the ServletConfig field in GenericServlet
volatile?
Mark