|
Roberto Nibali <ratz@xxxxxx> wrote:
> Hi Horms,
>
>> I notices that if persistance timeout is less than 2 min
>> (the FIN_WAIT timeout) then it will expire before its
>> controlled connections. This causes ip_vs_conn_expire to
>> set the template's timeout to 2min. Which caues somewhat unexpected
>> results if the persistance timeout, like say 10s.
>
> Since I've hardcoded FIN_WAIT to 10*HZ I didn't catch this.
If you think this is safe, which I strongly suspect it is,
can you send a patch, for 2.6 (or 2.4 I can rediff it if you want,
but it won't go into that tree) that changes this. I know its
trivial but lets practice sign-off procedure. Once we finally
work out a solution to the templates problem I'll send some patches
upstream, and this can go in the batch.
--
Horms
|
