On Thu, Aug 18, 2016 at 3:53 PM, Plüm, Rüdiger, Vodafone Group
<[email protected]> wrote:
>
> From a brief view looks good, but you miss to adjust ps->keep_alive in case
> that keepalives are killed due to busy workers.

Thanks for the review, fixed in the version I'm currently testing.

> And I don't get why we need this additional num = 0.

When apr_pollset_poll() returns an error, 'num' is not really/always
reliable (c.f. recent r1755758, and corresponding backports to APR
1.5/1.6]).
We better not fall through num > 0 here...


Regards,
Yann.

Reply via email to