Am 21.07.2015 um 23:14 schrieb Kevin A. McGrail:
On 7/21/2015 4:57 PM, Jim Jagielski wrote:
Hmmm Wondering if this is due to:
Test::LeakTrace [build_requires]
Test::TCP [build_requires]
Anyway, this is on OSX, fwiw.
This seems dependent on the Macports port(s) of perl...
Using the syste
On 7/21/2015 4:57 PM, Jim Jagielski wrote:
Hmmm Wondering if this is due to:
Test::LeakTrace [build_requires]
Test::TCP [build_requires]
Anyway, this is on OSX, fwiw.
This seems dependent on the Macports port(s) of perl...
Using the system perl doesn't seem to have this issue...
Any po
> On Jul 21, 2015, at 2:41 PM, Jim Jagielski wrote:
>
> Hmmm Wondering if this is due to:
>
> Test::LeakTrace [build_requires]
> Test::TCP [build_requires]
>
> Anyway, this is on OSX, fwiw.
>
This seems dependent on the Macports port(s) of perl...
Using the system perl doesn't seem to ha
What system are you testing on...?
I'm able to get CentOS6 running, but it fails w/ tests 17 and 18 on h2
and then hangs :(
On OSX, like I said, I can't even get past the byterange*.t tests
> On Jul 21, 2015, at 2:42 PM, Stefan Eissing
> wrote:
>
> :(
>
> It seemed the best suitable module f
Have to check and try. Think you register at the CTX as well. Don't know how
many are opened by mod_ssl.
> Am 21.07.2015 um 19:16 schrieb Eric Covener :
>
> On Mon, Jul 20, 2015 at 10:37 AM, Stefan Eissing
> wrote:
>>> Looks reasonable here, maybe need some clarification on when a
>>> protoc
:(
It seemed the best suitable module for the h2 tests. I am not stuck in it, if
we do those differently.
> Am 21.07.2015 um 20:30 schrieb Jim Jagielski :
>
> Hmmm... this is weird. The errors and hangs only seem to occur
> if Protocol::HTTP2 and AnyEvent are added...
>
>> On Jul 21, 2015,
Hmmm Wondering if this is due to:
Test::LeakTrace [build_requires]
Test::TCP [build_requires]
Anyway, this is on OSX, fwiw.
> On Jul 21, 2015, at 2:30 PM, Jim Jagielski wrote:
>
> Hmmm... this is weird. The errors and hangs only seem to occur
> if Protocol::HTTP2 and AnyEvent are added..
Hmmm... this is weird. The errors and hangs only seem to occur
if Protocol::HTTP2 and AnyEvent are added...
> On Jul 21, 2015, at 1:15 PM, Jim Jagielski wrote:
>
> I've tried some later versions of Perl (like 5.20 and 5.22) and
> I get really strange errors... like byterange.t simply hanging.
>
On Mon, Jul 20, 2015 at 10:37 AM, Stefan Eissing
wrote:
>> Looks reasonable here, maybe need some clarification on when a
>> protocol may be registered (post_config?).
>>
> Correct. That is necessary.
Would modssl_register_alpn() remain per-connection? If you're on the
fence at all -- I am worki
I've tried some later versions of Perl (like 5.20 and 5.22) and
I get really strange errors... like byterange.t simply hanging.
I also had some issues on a setup which had Perl 5.16 but with
all CPAN modules upgraded.
Is there some recommended version of Perl to use...
Ok, read the README. Always read the README...
> Am 21.07.2015 um 17:38 schrieb Stefan Eissing :
>
> Argh. What did I do? How can I check the site build before commit?
>
>> Am 21.07.2015 um 17:14 schrieb build...@apache.org:
>>
>> The Buildbot has detected a new failure on builder httpd-site-st
Argh. What did I do? How can I check the site build before commit?
> Am 21.07.2015 um 17:14 schrieb build...@apache.org:
>
> The Buildbot has detected a new failure on builder httpd-site-staging while
> building ASF Buildbot. Full details are available at:
>http://ci.apache.org/builders/http
On Tue, Jul 21, 2015 at 2:50 PM, Plüm, Rüdiger, Vodafone Group
wrote:
>
> I would go for 2.
+1
> -Original Message-
> From: Jan Kaluža
> Sent: Dienstag, 21. Juli 2015 14:35
> To: dev@httpd.apache.org
> Subject: mod_ssl: How to react on default OpenSSL SSL_CTX_set_options?
>
> Hi,
>
> in Fedora, OpenSSL maintainers are setting SSL_OP_NO_SSLv2 and
> SSL_OP_NO_SSLv3 options by defa
Hi,
in Fedora, OpenSSL maintainers are setting SSL_OP_NO_SSLv2 and
SSL_OP_NO_SSLv3 options by default [1].
This disables both SSLv2 and SSLv3 by default in the SSLv23_method(),
which is what mod_ssl uses when more than one version is requested.
The side effect of this change in OpenSSL is t
15 matches
Mail list logo
