Re: Balancers (file based SHMs) and restart issue (Windows only?)

2015-09-04 Thread Jim Jagielski
Instead of changing the default for all OSs, sounds like some Windows specific changes may need to be done. Has anyone using Windows actually seen this or confirmed this is possible?? > On Sep 2, 2015, at 5:08 AM, Yann Ylavic wrote: > > Re PR 58024. > > AIUI, balancers

httpd/trunk -Werror and warning free

2015-09-04 Thread Stefan Eissing
Below you find a patch that would make httpd/trunk - set -std=c89 on gcc in maintainer mode - abort compilation on every warning (-Werror) in maintainer-mode - fix all warnings that are currently there Please consider over the weekend. I would very much like to have that applied so all my C89

Re: httpd/trunk -Werror and warning free

2015-09-04 Thread Ruediger Pluem
AP_DECLARE_MODULE_LAST is no good alternative to AP_DECLARE_MODULE as it will break per module loglevels. We should not introduce this API. Why is this case op_Var: removed in util_expr_eval.c? Regards Rüdiger On 09/04/2015 03:26 PM, Stefan Eissing wrote: > Below you find a patch that would

Re: [RFC] Enable OCSP Stapling by default in httpd trunk

2015-09-04 Thread Kaspar Brand
On 02.09.2015 01:54, Jeff Trawick wrote: > On 08/30/2015 02:30 AM, Kaspar Brand wrote: >> today's situation, because this assessment misses the fact that with the >> current RFC-6066-based implementation, stapling can't fully achieve the >> goal of obviating OCSP queries for the clients - all

Re: [RFC] Enable OCSP Stapling by default in httpd trunk

2015-09-04 Thread Rob Stradling
On 04/09/15 15:59, Kaspar Brand wrote: > On 02.09.2015 01:54, Jeff Trawick wrote: >> On 08/30/2015 02:30 AM, Kaspar Brand wrote: >>> today's situation, because this assessment misses the fact that with the >>> current RFC-6066-based implementation, stapling can't fully achieve the >>> goal of

Re: httpd/trunk -Werror and warning free

2015-09-04 Thread Yann Ylavic
On Fri, Sep 4, 2015 at 4:24 PM, Ruediger Pluem wrote: > AP_DECLARE_MODULE_LAST is no good alternative to AP_DECLARE_MODULE as it will > break per module loglevels. > We should not introduce this API. How about: Index: include/http_log.h

AW: httpd/trunk -Werror and warning free

2015-09-04 Thread Plüm , Rüdiger , Vodafone Group
Doesn't that create another warning about aplog_module_index_fake_use being unused? Regards Rüdiger > -Ursprüngliche Nachricht- > Von: Yann Ylavic [mailto:ylavic@gmail.com] > Gesendet: Freitag, 4. September 2015 18:13 > An: dev@httpd.apache.org > Betreff: Re: httpd/trunk -Werror

Re: httpd/trunk -Werror and warning free

2015-09-04 Thread Yann Ylavic
On Fri, Sep 4, 2015 at 7:37 PM, Plüm, Rüdiger, Vodafone Group wrote: > Doesn't that create another warning about aplog_module_index_fake_use being > unused? It may not if the compiler knows about "inline"...

Re: proposed backport of mod_h2 - v4

2015-09-04 Thread Gregg Smith
Shouldn't this be ap_array_str_contains now in h2_switch.c? +while (*protos) { +/* Add all protocols we know (tls or clear) and that + * are part of the offerings (if there have been any). + */ -->+if (!offers || ap_array_contains(offers, *protos)) { +

Re: mod_lua unable to delete cookie

2015-09-04 Thread Mark Taylor
That works, thanks! -Mark On Thu, Sep 3, 2015 at 8:56 AM, Daniel Gruno wrote: > Hi Mark, > you have to give it a number different from 0, because of how the > internals work. This is an "optional integer" which falls back to 0 if > nothing is given, and if 0, then it's

Re: r1701005 httpd-trunk\server\protocols.c

2015-09-04 Thread NormW
On 4/09/2015 5:02 PM, Gregg Smith wrote: On 9/3/2015 9:23 PM, NormW wrote: On 4/09/2015 9:05 AM, Yann Ylavic wrote: On Fri, Sep 4, 2015 at 12:52 AM, NormW wrote: Hi again, Pushing passed previous problem, now arrive at: D:\Projects\svn\httpd-trunk\server>svn diff Index:

Re: r1701005 httpd-trunk\server\protocols.c

2015-09-04 Thread Gregg Smith
On 9/3/2015 9:23 PM, NormW wrote: On 4/09/2015 9:05 AM, Yann Ylavic wrote: On Fri, Sep 4, 2015 at 12:52 AM, NormW wrote: Hi again, Pushing passed previous problem, now arrive at: D:\Projects\svn\httpd-trunk\server>svn diff Index: protocol.c

Re: r1701005 httpd-trunk\server\protocols.c

2015-09-04 Thread Stefan Eissing
I think we are all tired of fixing compiler warnings on committed code. I do not see those warnings in --maintainer-mode on OS X and yet, clang is able to generate warnings about very, very obscure things. So I assume we are not using the compiler to its fullest extend and are doings this "by

Re: svn commit: r1701005 - in /httpd/httpd/trunk: docs/manual/mod/core.xml include/http_protocol.h modules/ssl/ssl_engine_kernel.c server/core.c server/protocol.c

2015-09-04 Thread Stefan Eissing
Thanks for the patch, Yann! Applied in r1701178. > Am 03.09.2015 um 17:46 schrieb Yann Ylavic : > > On Thu, Sep 3, 2015 at 2:45 PM, wrote: >> Author: icing >> Date: Thu Sep 3 12:45:26 2015 >> New Revision: 1701005 >> >> URL: