httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Rainer Jung <rainer.j...@kippdata.de>
Subject Re: [VOTE] Release httpd-2.2.32
Date Wed, 11 Jan 2017 09:29:49 GMT
Am 09.01.2017 um 19:21 schrieb William A Rowe Jr:
> The pre-release candidate tarballs of Apache legacy httpd 2.2.32
> can be found in;
>
> http://httpd.apache.org/dev/dist/
>
> Thanks to all for patches and reviews to get us to this point.
> STATUS file is updated to reflect end of maintenance Jul 1 '17.
>
> Your votes, please?
>
> +/-1
> [X] Release 2.2.32 as legacy GA

+1 to release, thanks for RMing

Details:

- Netware and Windows binary artefacts not checked (files currently missing)
- Sigs and hashes OK
- key in KEYS file
- contents of gz and bz2 tarballs identical
- contents of svn tag and tarballs identical
   except for expected deltas
   (we could cleanup some m4 files in apr-util/xml/expat/conftools
    at the end of buildconf, no regression)

Built on

- Solaris 8+10 Sparc as 32 Bit Binaries
- SLES 11 (64 Bits)
- RHEL 6 (64 Bits)

For all platforms built

- out of tree
- with default (static) and shared modules
- with module sets most, all and default
- with MPMs prefork, worker, event (where applicable)
- against bundled APR/APU,
   plus external APR/APU 1.5.2/1.5.4 plus external apr 1.6.x r1774577 
(or r1778111) (close to head) and apu 1.5.x r1774577 (close to head).

- using libraries
   - expat bundled plus 2.2.0
   - pcre bundled plus 8.39
   - openssl 1.0.2j with patches

- Tool chain:
     - platform gcc on Linux,
       gcc 4.4.4 for Solaris 8 and 6.2.0 for Solaris 10
     - CFLAGS: -O2 -g -Wall -fno-strict-aliasing
               (plus -mpcu=v9 -D_XPG6 -D_XOPEN_SOURCE 
-D_XOPEN_SOURCE_EXTENDED=1 -D__EXTENSIONS__ on Solaris)

164 build succeeded, 1 build on SLES 11 didn't succeed due to a 
segmentation fault in ksh during "make install". Not a regression.

   - compiler warnings:

modules/core/mod_watchdog.c:437: warning: 'rv' may be used uninitialized 
in this function
modules/proxy/mod_proxy_express.c:111: warning: 'backend' may be used 
uninitialized in this function
modules/ssl/ssl_util_stapling.c:686: warning: 'ok' may be used 
uninitialized in this function
srclib/apr-util/xlate/xlate.c:120:38: warning: passing argument 2 of 
'iconv' from incompatible pointer type (only Solaris 10)

     - in bundled PCRE

srclib/pcre/maketables.c:141:3: warning: this 'if' clause does not 
guard... [-Wmisleading-indentation]

     - in bundled apr-util

srclib/apr-util/xlate/xlate.c:120:38: warning: passing argument 2 of 
'iconv' from incompatible pointer type [-Wincompatible-pointer-types]
srclib/apr-util/xlate/xlate.c:343:42: warning: passing argument 2 of 
'iconv' from incompatible pointer type [-Wincompatible-pointer-types]

     - in mod_ssl

modules/ssl/ssl_expr_parse.c:3:1: warning: 'ssl_expr_yyrcsid' defined 
but not used [-Wunused-const-variable=]

   - Not sure whether the following observation first seen for 2.2.23 
still applies, because I have workaround in place:

     Problem with configure on Linux when using external PCRE in a 
non-system path. configure puts -lpcre into the LIBS early and the later 
tests all fail, when they try to run the conftest binary, because the 
pcre lib can not be found. It's a bit unfortunate, that all libs needed 
to build httpd are added as libs during all configure tests after the 
respective --with flag. Workaround is adding these libs to 
LD_LIBRARY_PATH just during configure. Works out of the box on Solaris, 
because "pcre-config --libs" on Solaris also outputs a "-R" flag.


Tested 330 combinations, each build with log levels info and debug

- the following test failures were seen:

a Test 54 of t/modules/include.t only and always on
   Solaris.
   This is due to a bug in the test, which uses strftime()
   with a "%s" pattern that is not supported on Solaris.
   Until recently the server and the test client both returned
   verbatim "%s" and the test succeeded. After updating some
   Perl modules for the http2 tests, the perl client even
   on Solaris now supports "%s" in strftime and the test starts
   to fail. It seems we have to fix the test.
   Not a regression

b For t/apache/server_name_port.t the following
   TODO tests that actually pass: 57, 60, 81, 84

Regards,

Rainer

Mime
View raw message