Re: svn commit: r1834671 - in /httpd/httpd/branches/2.4.x: CHANGES docs/manual/mod/mod_md.xml modules/md/md_crypt.c modules/md/md_json.c modules/md/md_version.h modules/md/mod_md.c modules/md/mod_md_c

[Christophe Jaillet]

Le 29/06/2018 à 13:53, ic...@apache.org a écrit :

Author: icing
Date: Fri Jun 29 11:53:50 2018
New Revision: 1834671

URL: http://svn.apache.org/viewvc?rev=1834671=rev
Log:
On the 2.4.x branch:

backport of current mod_md version and documentation.


Modified:
 httpd/httpd/branches/2.4.x/CHANGES
 httpd/httpd/branches/2.4.x/docs/manual/mod/mod_md.xml
 httpd/httpd/branches/2.4.x/modules/md/md_crypt.c
 httpd/httpd/branches/2.4.x/modules/md/md_json.c
 httpd/httpd/branches/2.4.x/modules/md/md_version.h
 httpd/httpd/branches/2.4.x/modules/md/mod_md.c
 httpd/httpd/branches/2.4.x/modules/md/mod_md_config.c



Hi,

according to 2.4.x/STATUS, mod_md is not CTR.
Maybe, STATUS should be updated to add an exception for mod_md? (or 
maybe even for any module marked as experimental?)

Otherwise, this should go through STATUS first.

CJ

Re: svn commit: r1834497 - /httpd/httpd/trunk/acinclude.m4

[jean-frederic clere]

On 27/06/18 19:26, William A Rowe Jr wrote:
> Does it make sense to favor the lib/ over the lib64/ installation of
> this pkg?
> Is this consistent across all directory choices?
> 
> It seems we are missing a level of abstraction and failing to utilize AC
> stock tools themselves... see my recent trunk/ work around PCRE for
> a significant simplification/refactoring (not arguing I was correct,
> actually
> asking for more eyeballs.)

There a bunch of tests to lib and lib64 to use pkgconfig that are
similar to the one I have added, I have not checked that all components
are tested for lib and lib64, that probably requires something more
generic and remove the copy + paste we have acinclude.m4

Cheers

Jean-Frederic


> 
> On Wed, Jun 27, 2018 at 7:15 AM,  > wrote:
> 
> Author: jfclere
> Date: Wed Jun 27 12:15:32 2018
> New Revision: 1834497
> 
> URL: http://svn.apache.org/viewvc?rev=1834497=rev
> 
> Log:
> Allow openssl.pc to be in lib64 (like other components).
> 
> Modified:
>     httpd/httpd/trunk/acinclude.m4
> 
> Modified: httpd/httpd/trunk/acinclude.m4
> URL:
> 
> http://svn.apache.org/viewvc/httpd/httpd/trunk/acinclude.m4?rev=1834497=1834496=1834497=diff
> 
> 
> 
> ==
> --- httpd/httpd/trunk/acinclude.m4 (original)
> +++ httpd/httpd/trunk/acinclude.m4 Wed Jun 27 12:15:32 2018
> @@ -542,12 +542,18 @@ AC_DEFUN([APACHE_CHECK_OPENSSL],[
>      dnl Before doing anything else, load in pkg-config variables
>      if test -n "$PKGCONFIG"; then
>        saved_PKG_CONFIG_PATH="$PKG_CONFIG_PATH"
> -      if test "x$ap_openssl_base" != "x" -a \
> -              -f "${ap_openssl_base}/lib/pkgconfig/openssl.pc"; then
> -        dnl Ensure that the given path is used by pkg-config too,
> otherwise
> -        dnl the system openssl.pc might be picked up instead.
> -       
> 
> PKG_CONFIG_PATH="${ap_openssl_base}/lib/pkgconfig${PKG_CONFIG_PATH+:}${PKG_CONFIG_PATH}"
> -        export PKG_CONFIG_PATH
> +      if test "x$ap_openssl_base" != "x"; then
> +        if test -f "${ap_openssl_base}/lib/pkgconfig/openssl.pc"; then
> +          dnl Ensure that the given path is used by pkg-config too,
> otherwise
> +          dnl the system openssl.pc might be picked up instead.
> +         
> 
> PKG_CONFIG_PATH="${ap_openssl_base}/lib/pkgconfig${PKG_CONFIG_PATH+:}${PKG_CONFIG_PATH}"
> +          export PKG_CONFIG_PATH
> +        elif test -f
> "${ap_openssl_base}/lib64/pkgconfig/openssl.pc"; then
> +          dnl Ensure that the given path is used by pkg-config too,
> otherwise
> +          dnl the system openssl.pc might be picked up instead.
> +         
> 
> PKG_CONFIG_PATH="${ap_openssl_base}/lib64/pkgconfig${PKG_CONFIG_PATH+:}${PKG_CONFIG_PATH}"
> +          export PKG_CONFIG_PATH
> +        fi
>        fi
>      
>  
> AC_ARG_ENABLE(ssl-staticlib-deps,APACHE_HELP_STRING(--enable-ssl-staticlib-deps,[link
> mod_ssl with dependencies of OpenSSL's static libraries (as
> indicated by "pkg-config --static"). Must be specified in addition
> to --enable-ssl.]), [
>          if test "$enableval" = "yes"; then
> 
> 
> 


-- 
Cheers

Jean-Frederic

Re: mod_md source sync

[Stefan Eissing]

Ok, I copied the MODSSL_USE_OPENSSL_PRE_1_1_API definitions into md_crypt.c and 
hope I did it right. Otherwise Bernard will probably kick me politely...

> Am 29.06.2018 um 12:00 schrieb Stefan Eissing :
> 
> While synching sources of mod_md between apache and github, I have a question 
> for the change in md_crypt.c:
> 
> 1806939  icing 
> 1833598 ylavic #if MODSSL_USE_OPENSSL_PRE_1_1_API || 
> (defined(LIBRESSL_VERSION_NUMBER) && \
> 1833598 ylavic
> LIBRESSL_VERSION_NUMBER < 0x207f)
> 1804123  icing 
> 
> How is MODSSL_USE_OPENSSL_PRE_1_1_API available from 
> modules/ssl/ssl_private.h? And if not, what is the intention of the change 
> and what needs to be different here?
> 
> Thanks!
> 
> -Stefan

mod_md source sync

[Stefan Eissing]

While synching sources of mod_md between apache and github, I have a question 
for the change in md_crypt.c:

1806939  icing 
1833598 ylavic #if MODSSL_USE_OPENSSL_PRE_1_1_API || 
(defined(LIBRESSL_VERSION_NUMBER) && \
1833598 ylavic
LIBRESSL_VERSION_NUMBER < 0x207f)
1804123  icing 

How is MODSSL_USE_OPENSSL_PRE_1_1_API available from modules/ssl/ssl_private.h? 
And if not, what is the intention of the change and what needs to be different 
here?

Thanks!

-Stefan