broken pipe (was:Re: 1.0.0.rc2 tarballs now ready...)
The problem is in our own library. The recent changes cause itto don't catch the end of the generated file and go on reading it until the read fails. In 2.1 previous tomod_cgi.c:r1.153 the return value of ap_pass_brigade() was not checked in include_cmd(), but now it is done. That change mixed with our library update got us in that state.But I guess the ball is in our camp now, which iseasier to manage... Thanks, JJ [EMAIL PROTECTED] 6/30/2004 4:41:09 PM I am getting the following error when running CGIs on the current version of NetWare (6.5 sp2): (32)Broken pipe: ap_content_length_filter: apr_bucket_read() failed I am working on tracking down the problem. Changes done to mod_cgi.c, mod_include.(h c) back in August 22, 2003 are exposinga broken pipe problem on NetWare. mod_cgi.c had a great deal of change in r1.153. I am not sure what exactly is causing that problem. It is surfacingonly now because of recent changes in the pipe code of our own library. At this point I am highly suspicious thatthe problem is in our library, but it could also be in the http/apr code. We are trying to isolate the problem and will update the list with any new info. JJ
[STATUS] (apr) Wed Jun 30 23:45:21 EDT 2004
APACHE PORTABLE RUNTIME (APR) LIBRARY STATUS: -*-text-*- Last modified at [$Date: 2004/06/30 11:29:48 $] Releases: Standalone 1.0.0 rc2 : tagged June 30th 2004 (APR_1_0_RC2) 1.0.0 rc1 : tagged June 18th 2004 (APR_1_0_RC1) 0.9.3 : tagged March 30, 2003 0.9.2 : released March 22, 2003 0.9.1 : released September 11, 2002 0.9.0 : released August 28, 2002 Bundled with httpd 2.0a9 : released December 12, 2000 2.0a8 : released November 20, 2000 2.0a7 : released October 8, 2000 2.0a6 : released August 18, 2000 2.0a5 : released August 4, 2000 2.0a4 : released June 7, 2000 2.0a3 : released April 28, 2000 2.0a2 : released March 31, 2000 2.0a1 : released March 10, 2000 RELEASE 0.9 SHOWSTOPPERS: RELEASE 1.0 SHOWSTOPPERS: CURRENT VOTES: CURRENT test/testall -v EXCEPTIONS: Please add any platform anomilies to the following exception list. * various tests fail on Unix in VPATH builds. * 'testipsub' will tickle an Solaris 8 getaddrinfo() IPv6 bug, causing the test to hang. Configure with --disable-ipv6 if using an unpatched Solaris 8 installation. * The 'testdso' tests will not work if configured with --disable-shared since the loadable modules cannot be built. * 'testdso' fails on older versions of OpenBSD due to dlsym(NULL, ...) segfaulting. * BUG: Win32 fails test in File Info: test_stat_eq_finfo apr_stat and apr_getfileinfo differ in protection ... wrowe guesses that we are checking the handle objects' permissions rather than the filesystem objects' permissions. * Win32 Not Implemented tests Pipes: set_timeout/read_write; can't timeout blocking pipes Socket Creation: tcp6_socket and udp6_socket (at least by default) Socket Options: corkable: TCP isn't corkable Users: username: Groups from apr_uid_get not implemented RELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP: * Someone needs to port testucs to Unix. Right now it only works on Windows. * The return type of a thread function (void *) is inconsistent with the type used in apr_thread_exit()/apr_thread_join() (apr_status_t). The thread function's return type should be changed to apr_status_t so that a return from the thread main function has the same effect as apr_thread_exit(). See Message-Id: [EMAIL PROTECTED] for thread discussing this. +1: BrianH, Aaron, david, jerenkrantz Status: Will Rowe was working on this. * Need some architecture/OS specific versions of the atomic operations. progress: generic, solaris Sparc, FreeBSD5, linux, and OS/390 done need: AIX, AS400, HPUX * The new lock API is a full replacement for the old API, but is not yet complete on all platforms. Components that are incomplete or missing include: Netware: apr_proc_mutex_*() (Is proc_mutex unnecessary on Netware?) * proc_mutex is not necessary on NetWare since the OS does not support processes. The proc_mutex APIs actually redirect to the thread_mutex APIs. (bnicholes) OS/2: apr_thread_cond_*(), apr_proc_mutex_*() Less critical components that we may wish to add at some point: Beos: apr_thread_rwlock_try*lock() apr_proc_mutex_trylock() Unix: apr_thread_rwlock_*() for platforms w/o rwlocks in pthread Win32: apr_thread_cond_timedwait(), apr_proc_mutex_*() (Is proc_mutex unnecessary on Win32?) * Need to contemplate apr_strftime... platforms vary. OtherBill suggested this solution (but has no time to implement): Document our list of 'supported' escapes. Run some autoconf/m4 magic against the complete list we support. Move the strftime re-implementation from time/win32 to time/unix. Add some APR_HAVE_STRFTIME magic to use the system fn, or fail over to time/unix/strftime.c. Message-ID: [EMAIL PROTECTED] * Using reentrant libraries with non-threaded APR - Anecdotal evidence exists that suggests it is bad to mix reentrant and non-reentrant libraries and therefore we should always use the reentrant versions. - Unfortunately, on some platforms (AIX 4.2.1) defining the reentrant flag (-D_THREAD_SAFE) causes builds to fail, and so one would expect --disable-threads to fix this. Although this has been fixed for that particular version of AIX, it may be useful to only enable the reentrant versions when threads are enabled. How will we deal with this issue once APR becomes a standalone library? It is perfectly legitimate to have apps needing both versions (threaded/reentrant and non-threaded/non-reentrant) on the same machine. * Pools
[STATUS] (apr-util) Wed Jun 30 23:45:34 EDT 2004
APRUTIL LIBRARY STATUS: -*-text-*- Last modified at [$Date: 2004/06/30 11:44:19 $] Release: 1.0.0 rc2 : Tagged June 30th 2004(APU_1_0_RC2) 1.0.0 rc1 : Tagged June 2004 (APU_1_0_RC1) 0.9.3 : Tagged March 30, 2002 0.9.2 : Released March 22, 2002 (alpha) 0.9.1 : Released September 11, 2002 (alpha) 0.9.0 : Not released 2.0a9 : released December 12, 2000 RELEASE SHOWSTOPPERS: RELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP: * Solaris's Sun Freeware (sfw) package has a busted gcc/ld setup. This gcc passes -L/opt/sfw/lib to /usr/ccs/bin/ld, but does not pass -R. Therefore, when trying to run the code using a library from /opt/sfw/lib (say, libdb), the run-time linker will not look in /opt/sfw/lib and the program will die. Status: Workaround is to add -R/opt/sfw/lib to LDFLAGS. Should check latest sfw package set and see if Sun may have fixed this. * GDBM usage of errno is not-thread-safe. Fix. Other bugs that need fixing: Other features that need writing: * possibly move test/testdbm* to util/dbu Justin says: Do we still want to do this? testdate is now in test. Status: Greg +1 (volunteers) Documentation that needs writing: * API documentation Status: * doc the lifetimes of apr_dbm return values Available Patches: Open Issues:
Re: apr-util RPM - I'm stuck
On Wed, Jun 30, 2004 at 05:50:21PM +0200, Graham Leggett wrote: Joe Orton wrote: Can you post them for review first? The patch is to buildconf. It creates apr.spec from apr.spec.in on platforms that support cut. buildconf change looks OK. For the spec file: - just use a %changelog which says derived from Fedora Core apr.spec or something rather than include the entire history - in %install, remove everything between make install and # Move docs...; none of that should be necessary. - NOTICE should be in the main package %doc list for a ASL2.0-licensed package. joe
Re: apr-util RPM - I'm stuck
Joe Orton wrote: buildconf change looks OK. For the spec file: - just use a %changelog which says derived from Fedora Core apr.spec or something rather than include the entire history - in %install, remove everything between make install and # Move docs...; none of that should be necessary. Cool - will commit it when I get to my dev machine later this afternoon. - NOTICE should be in the main package %doc list for a ASL2.0-licensed package. NOTICE is not installed anywhere by the make install target - in theory the RPM should install what make install installs - is it possible to get make install to install NOTICE in the same place as LICENCE? Regards, Graham --
Re: apr-util RPM - I'm stuck
On Thu, Jul 01, 2004 at 12:23:25PM +0200, Graham Leggett wrote: Joe Orton wrote: - NOTICE should be in the main package %doc list for a ASL2.0-licensed package. NOTICE is not installed anywhere by the make install target - in theory the RPM should install what make install installs - is it possible to get make install to install NOTICE in the same place as LICENCE? LICENSE isn't installed either.
Re: apr pkgconfig use (apr.pc vs. apr-1.pc)
On Wed, Jun 30, 2004 at 10:39:16AM -0400, Greg Hudson wrote: On Wed, 2004-06-30 at 04:56, Joe Orton wrote: Install apr-config as apr-$(APR_MAJOR_VERSION)-config ? Greg wrote remove rather than rename, though indeed, I guess renaming should work. Renaming is also fine. Installing ap[ru]-$N-config alongside ap[ru]-config is probably all that's going to be feasible for the 1.0.0 release, at least I don't have time to deal with the fallout of removing ap[ru]-config entirely right now. The AP?_FIND_AP? macros in find_ap?.m4 need to be adjusted to take a minimum required library version argument, there's probably more... joe
Re: apr pkgconfig use (apr.pc vs. apr-1.pc)
On Thu, 2004-07-01 at 10:30, Joe Orton wrote: Installing ap[ru]-$N-config alongside ap[ru]-config is probably all that's going to be feasible for the 1.0.0 release, at least I don't have time to deal with the fallout of removing ap[ru]-config entirely right now. Hm. That's disappointing; it means that apr 0.9 can't be co-located on a system with apr 1.0 without hacking, and that apr-0.9-using apps will be screwed by the installation of apr 1.0.
Re: apr pkgconfig use (apr.pc vs. apr-1.pc)
On Thu, 2004-07-01 at 10:39, Joe Orton wrote: This is done on HEAD: patches welcome for below to expedite removal of ap[ru]-config. The AP?_FIND_AP? macros in find_ap?.m4 need to be adjusted to take a minimum required library version argument, there's probably more... Would it be okay to make it take just a library version argument, no minimum required? It means if an application wants to work against multiple major versions of APR, it has to implement its own fallback logic.
Re: apr pkgconfig use (apr.pc vs. apr-1.pc)
On Thu, Jul 01, 2004 at 10:45:54AM -0400, Greg Hudson wrote: On Thu, 2004-07-01 at 10:39, Joe Orton wrote: This is done on HEAD: patches welcome for below to expedite removal of ap[ru]-config. The AP?_FIND_AP? macros in find_ap?.m4 need to be adjusted to take a minimum required library version argument, there's probably more... Would it be okay to make it take just a library version argument, no minimum required? It means if an application wants to work against multiple major versions of APR, it has to implement its own fallback logic. If apr-config is not renamed in the 0.9 branch (which is really required for compatibility), then I guess yes, make the argument optional, and have logic: AP?_FIND_AP?(without extra version argument) = use apr-1-config if found else apr-config if found AP?_FIND_AP?(requires version N) = use apr-N-config if found else no fallback the app can always just call the macro again with a different N. (BTW note that the apr-config script is only named apr-N-config by make install; it's still called apr-config in a build tree) joe
Re: 1.0.0.rc2 tarballs now ready...
On Wed, Jun 30, 2004 at 05:00:52PM +0100, David Reid wrote:
So, I fixed the incorrect tag and rerolled. The revised tarballs (correctly
named to keep Joe happy this time) are being uploaded into the usual place,
http://www.apache.org/~dreid/ as I write this.
I'm getting a segfault in testsockets on Solaris 9/Intel. No time to further
debug. Is anyone else seeing this? Best I can tell is that an earlier
test is failing and it's passing a NULL parameter on.
Here's the backtrace:
(dbx 1) run testsockets
Running: testall testsockets
(process id 12674)
testsockets : \signal SEGV (no mapping at the fault address) in
apr_socket_bind at line 131 in file sockets.c
131(struct sockaddr *)sa-sa, sa-salen) == -1) {
(dbx 2) where
current thread: [EMAIL PROTECTED]
=[1] apr_socket_bind(sock = 0x80fe478, sa = (nil)), line 131 in sockets.c
[2] sendto_receivefrom(tc = 0x8100d30, data = (nil)), line 141 in
testsockets.c
[3] abts_run_test(ts = 0x8100ca0, f = 0x805fac0 =
`testall`testsockets.c`sendto_receivefrom(abts_case *tc, void *data), value =
(nil)), line 171 in abts.c
[4] testsockets(suite = 0x8100ca0), line 201 in testsockets.c
[5] main(argc = 2, argv = 0x8046954), line 409 in abts.c
HTH. -- justin
Re: 1.0.0.rc2 tarballs now ready...
On Thu, Jul 01, 2004 at 08:24:18AM -0700, Justin Erenkrantz wrote: On Wed, Jun 30, 2004 at 05:00:52PM +0100, David Reid wrote: So, I fixed the incorrect tag and rerolled. The revised tarballs (correctly named to keep Joe happy this time) are being uploaded into the usual place, http://www.apache.org/~dreid/ as I write this. I'm getting a segfault in testsockets on Solaris 9/Intel. No time to further debug. Is anyone else seeing this? Best I can tell is that an earlier test is failing and it's passing a NULL parameter on. Does ./testall -v testsockets give any useful information?
Re: 1.0.0.rc2 tarballs now ready...
On Thu, Jul 01, 2004 at 04:34:49PM +0100, Joe Orton wrote: On Thu, Jul 01, 2004 at 08:24:18AM -0700, Justin Erenkrantz wrote: On Wed, Jun 30, 2004 at 05:00:52PM +0100, David Reid wrote: So, I fixed the incorrect tag and rerolled. The revised tarballs (correctly named to keep Joe happy this time) are being uploaded into the usual place, http://www.apache.org/~dreid/ as I write this. I'm getting a segfault in testsockets on Solaris 9/Intel. No time to further debug. Is anyone else seeing this? Best I can tell is that an earlier test is failing and it's passing a NULL parameter on. Does ./testall -v testsockets give any useful information? testsockets : |Line 137: expected 0, but saw 670008 \zsh: 13868 segmentation fault ./testall -v testsockets I was curious why testall doesn't abort a sub-test after a failure, but oh well, I didn't write it... -- justin
Re: 1.0.0.rc2 tarballs now ready...
On Thu, Jul 01, 2004 at 09:28:31AM -0700, Justin Erenkrantz wrote: On Thu, Jul 01, 2004 at 04:34:49PM +0100, Joe Orton wrote: On Thu, Jul 01, 2004 at 08:24:18AM -0700, Justin Erenkrantz wrote: I'm getting a segfault in testsockets on Solaris 9/Intel. No time to further debug. Is anyone else seeing this? Best I can tell is that an earlier test is failing and it's passing a NULL parameter on. Does ./testall -v testsockets give any useful information? testsockets : |Line 137: expected 0, but saw 670008 \zsh: 13868 segmentation fault ./testall -v testsockets Oh for an apr_assert_success... the AF_UNSPEC lookup for ::1 gave EAI_NONAME by my calculations which is not unexpected for a build on a Solaris system without IPv6 configured; not a serious failure. I was curious why testall doesn't abort a sub-test after a failure, but oh well, I didn't write it... -- justin It did, and now it doesn't, and stuff like this happens a lot now. joe
apr-config and apr-1-config
Hi all, I've noticed that the most recent CVS of 1.0.0 installs both /usr/bin/apr-config, and /usr/bin/apr-1-config. Is this intentional? Regards, Graham -- smime.p7s Description: S/MIME Cryptographic Signature
Re: apr-util RPM - I'm stuck
Joe Orton wrote:
buildconf change looks OK. For the spec file:
- just use a %changelog which says derived from Fedora Core apr.spec
or something rather than include the entire history
- in %install, remove everything between make install and # Move
docs...; none of that should be necessary.
In the process of committing to apr, with the above changes added.
Can you review the proposed diff to apr-util? What should I change in
the apr-util spec file to make it work with the apr file?
- NOTICE should be in the main package %doc list for a ASL2.0-licensed
package.
Will see if I can convince make install to install this.
Regards,
Graham
--
Index: buildconf
===
RCS file: /home/cvs/apr-util/buildconf,v
retrieving revision 1.12
diff -u -r1.12 buildconf
--- buildconf 21 Feb 2004 00:43:56 - 1.12
+++ buildconf 1 Jul 2004 21:15:33 -
@@ -74,3 +74,20 @@
# Remove autoconf cache again
rm -rf autom4te*.cache
+
+# Create RPM Spec file
+if [ -f `which cut` ]; then
+ echo rebuilding rpm spec file
+ ( REVISION=`build/get-version.sh all include/apu_version.h APU`
+VERSION=`echo $REVISION | cut -d- -s -f1`
+RELEASE=`echo $REVISION | cut -d- -s -f2`
+if [ x$VERSION = x ]; then
+ VERSION=$REVISION
+ RELEASE=1
+fi
+cat ./build/rpm/apr-util.spec.in | \
+sed -e s/APU_VERSION/$VERSION/ \
+-e s/APU_RELEASE/$RELEASE/ \
+ apr-util.spec )
+fi
+
%define apuver 1
Summary: Apache Portable Runtime Utility library
Name: apr-util
Version: APU_VERSION
Release: APU_RELEASE
License: Apache Software License
Group: System Environment/Libraries
URL: http://apr.apache.org/
Source0: %{name}-%{version}.tar.gz
BuildRoot: %{_tmppath}/%{name}-%{version}-%{release}-buildroot
BuildPrereq: autoconf, libtool, doxygen, apr-devel = 0:{version}-{release}
BuildPrereq: openldap-devel, db4-devel, expat-devel
Conflicts: subversion 0.20.1-2
%description
The mission of the Apache Portable Runtime (APR) is to provide a
free library of C data structures and routines. This library
contains additional utility interfaces for APR; including support
for XML, LDAP, database interfaces, URI parsing and more.
%package devel
Group: Development/Libraries
Summary: APR utility library development kit
Requires: apr-util = %{version}-%{release}, apr-devel
Requires: openldap-devel, db4-devel, expat-devel
Conflicts: subversion-devel 0.20.1-2
%description devel
This package provides the support files which can be used to
build applications using the APR utility library. The mission
of the Apache Portable Runtime (APR) is to provide a free
library of C data structures and routines.
%prep
%setup -q
%build
autoheader autoconf
%configure --with-apr=%{_prefix} \
--includedir=%{_includedir}/apr-%{apuver} \
--with-ldap --without-gdbm
make %{?_smp_mflags} make dox
%install
rm -rf $RPM_BUILD_ROOT
make install DESTDIR=$RPM_BUILD_ROOT
# Documentation
mv docs/dox/html html
# Unpackaged files
rm -f $RPM_BUILD_ROOT%{_libdir}/aprutil.exp
%check
# Run the less verbose tests
%define tests testmd5 testrmm teststrmatch testuri testxlate
cd test; make %{?_smp_mflags} %{tests} testdbm
for t in %{tests}; do ./${t} || exit 1; done
./testdbm auto tsdbm
./testdbm -tDB auto tbdb.db
%clean
rm -rf $RPM_BUILD_ROOT
%post -p /sbin/ldconfig
%postun -p /sbin/ldconfig
%files
%defattr(-,root,root,-)
%doc CHANGES LICENSE
%{_libdir}/libaprutil-%{apuver}.so.*
%files devel
%defattr(-,root,root,-)
%{_bindir}/apu-config
%{_libdir}/libaprutil-%{apuver}.*a
%{_libdir}/libaprutil-%{apuver}.so
%{_libdir}/pkgconfig/apr-util.pc
%{_includedir}/apr-%{apuver}/*.h
%doc --parents html
%changelog
* Tue Jun 22 2004 Graham Leggett [EMAIL PROTECTED] 1.0.0-1
- update to support v1.0.0 of APR
* Sat Jun 19 2004 Joe Orton [EMAIL PROTECTED] 0.9.4-16
- have -devel require matching release of apr-util
* Tue Jun 15 2004 Elliot Lee [EMAIL PROTECTED]
- rebuilt
* Thu Apr 1 2004 Joe Orton [EMAIL PROTECTED] 0.9.4-14
- fix use of SHA1 passwords (#119651)
* Tue Mar 30 2004 Joe Orton [EMAIL PROTECTED] 0.9.4-13
- remove fundamentally broken check_sbcs() from xlate code
* Fri Mar 19 2004 Joe Orton [EMAIL PROTECTED] 0.9.4-12
- tweak xlate fix
* Fri Mar 19 2004 Joe Orton [EMAIL PROTECTED] 0.9.4-11
- rebuild with xlate fixes and tests enabled
* Tue Mar 02 2004 Elliot Lee [EMAIL PROTECTED] 0.9.4-10.1
- rebuilt
* Tue Mar 2 2004 Joe Orton [EMAIL PROTECTED] 0.9.4-10
- rename sdbm_* symbols to apu__sdbm_*
* Mon Feb 16 2004 Joe Orton [EMAIL PROTECTED] 0.9.4-9
- fix sdbm apr_dbm_exists() on s390x/ppc64
* Fri Feb 13 2004 Elliot Lee [EMAIL PROTECTED] 0.9.4-8
- rebuilt
* Thu Feb 5 2004 Joe Orton [EMAIL PROTECTED] 0.9.4-7
- fix warnings from use of apr_optional*.h with gcc 3.4
* Thu Jan 29 2004 Joe Orton [EMAIL PROTECTED] 0.9.4-6
- drop gdbm support
* Thu Jan 8 2004 Joe Orton [EMAIL PROTECTED] 0.9.4-5
- fix DB
Re: apr-config and apr-1-config
On Thu, Jul 01, 2004 at 11:10:25PM +0200, Graham Leggett wrote: I've noticed that the most recent CVS of 1.0.0 installs both /usr/bin/apr-config, and /usr/bin/apr-1-config. Is this intentional? Yes, for the moment.
Re: apr-util RPM - I'm stuck
Joe Orton wrote: - NOTICE should be in the main package %doc list for a ASL2.0-licensed package. Should NOTICE, LICENSE and CHANGES appear in both the apr and the apr-devel packages separately? Or is the existence of these files in the apr package (leaving them out of apr-devel, which depends on apr) enough? Regards, Graham -- smime.p7s Description: S/MIME Cryptographic Signature
Re: apr-util RPM - I'm stuck
On Thu, Jul 01, 2004 at 11:19:01PM +0200, Graham Leggett wrote: Can you review the proposed diff to apr-util? What should I change in the apr-util spec file to make it work with the apr file? It should just work. Otherwise: - no need to run auto* in the spec file for apr-util if not including a bunch of configure.in patches ;) - again add NOTICE to %doc and remove the history from the %changelog - %check can build and run testall - %files needs to be adjust for apr-util-1.pc
Re: apr-config and apr-1-config
Joe Orton wrote: I've noticed that the most recent CVS of 1.0.0 installs both /usr/bin/apr-config, and /usr/bin/apr-1-config. Is this intentional? Yes, for the moment. How do you suggest the RPM should handle this? - at the moment installing a v0 RPM and a v1 RPM simultaneously will cause a conflict. Should we just leave it as is for now? Regards, Graham -- smime.p7s Description: S/MIME Cryptographic Signature
Re: apr-util RPM - I'm stuck
On Thu, Jul 01, 2004 at 11:43:10PM +0200, Graham Leggett wrote: Joe Orton wrote: - NOTICE should be in the main package %doc list for a ASL2.0-licensed package. Should NOTICE, LICENSE and CHANGES appear in both the apr and the apr-devel packages separately? Yes, absolutely, they are different in each. joe
Re: apr-config and apr-1-config
On Thu, Jul 01, 2004 at 11:45:44PM +0200, Graham Leggett wrote: Joe Orton wrote: I've noticed that the most recent CVS of 1.0.0 installs both /usr/bin/apr-config, and /usr/bin/apr-1-config. Is this intentional? Yes, for the moment. How do you suggest the RPM should handle this? - at the moment installing a v0 RPM and a v1 RPM simultaneously will cause a conflict. Should we just leave it as is for now? Yes, just leave it for the moment I guess. joe
Re: apr-util RPM - I'm stuck
On Thu, Jul 01, 2004 at 10:53:47PM +0100, Joe Orton wrote: On Thu, Jul 01, 2004 at 11:43:10PM +0200, Graham Leggett wrote: Should NOTICE, LICENSE and CHANGES appear in both the apr and the apr-devel packages separately? Yes, absolutely, they are different in each. Sorry, I read that as apr and apr-util; no, you had it right; those files should not be duplicated in the -devel subpackage. joe
Re: apr pkgconfig use (apr.pc vs. apr-1.pc)
Joe Orton wrote: If apr-config is not renamed in the 0.9 branch (which is really required for compatibility), then I guess yes, make the argument optional, and have logic: AP?_FIND_AP?(without extra version argument) = use apr-1-config if found else apr-config if found AP?_FIND_AP?(requires version N) = use apr-N-config if found else no fallback I may not be understanding all of the constraints. Can you tell me what goes wrong with this solution? * apr-config does not get renamed in 0.9. * We introduce a new macro APR_FIND_APR1 to find apr 1.0 by looking for apr-1-config. and similarly for apr-util and friends. httpd 2.1 changes to use APR_FIND_APR1 etc. instead of passing an argument to APR_FIND_APR. If an app wants to work against both versions of apr, it tries APR_FIND_APR1 and then tries APR_FIND_APR if apr_found is no. (BTW note that the apr-config script is only named apr-N-config by make install; it's still called apr-config in a build tree) Any reason that can't change? If it can't change, APR_FIND_APR1 can use apr-config when using a bundled apr directory. (Also, since I haven't seen other people jumping up and volunteering to make this happen, where is the boundary between what you're willing to do and where I should submit a patch?)
Re: broken pipe (was:Re: 1.0.0.rc2 tarballs now ready...)
If you toggle the APR_FILES_AS_SOCKETS constant back to 0 (zero)does this remedy the behavior you are seeing? Bill At 05:41 PM 6/30/2004, Jean-Jacques Clar wrote: I am getting the following error when running CGIs on the current version of NetWare (6.5 sp2): (32)Broken pipe: ap_content_length_filter: apr_bucket_read() failed I am working on tracking down the problem. Changes done to mod_cgi.c, mod_include.(h c) back in August 22, 2003 are exposing a broken pipe problem on NetWare. mod_cgi.c had a great deal of change in r1.153. I am not sure what exactly is causing that problem. It is surfacing only now because of recent changes in the pipe code of our own library. At this point I am highly suspicious that the problem is in our library, but it could also be in the http/apr code. We are trying to isolate the problem and will update the list with any new info. JJ
