You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@perl.apache.org by Geoffrey Young <ge...@modperlcookbook.org> on 2005/04/11 16:05:59 UTC
[RELEASE CANDIDATE] mod_perl 2.0-RC5
the mod_perl development team is pleased to announce that we have a new
candidate for mod_perl 2.0, ready and waiting for testers.
this release, mod_perl 1.999_22, is a _very_ significant release as it
contains major API changes and is completely incompatible with any prior
release of mod_perl 1.99. in other words, you will need to modify ALL of
your old 2.0 code before it will be able to run with this latest release.
fortunately, the changes required are minor, straightforward, and simple.
*** YOU WILL NEED TO MODIFY ALL YOUR CODE AFTER UPGRADING ***
so, before you dive into this release, please take the time to read this
http://perl.apache.org/docs/2.0/rename.html
as it explains the changes, the reasons behind them, and the migration path
for your current codebase.
the release tarball is available for download from here
http://perl.apache.org/docs/2.0/rename.html
did I mention to read the document at
http://perl.apache.org/docs/2.0/rename.html
?
:)
once you've read it, please take a moment to download and test the tarball
http://people.apache.org/~geoff/mod_perl-2.0.0-RC5.tar.gz
and report back successes or failures. your help in testing is greatly
appreciated
--Geoff
Changes since RC4:
******************** IMPORTANT ********************
this version of mod_perl is completely incompatible
with prior versions of mod_perl, both 1.XX and
1.99_XX. Please read the below changes carefully.
***************************************************
remove MP_INST_APACHE2 installation option and Apache2.pm - all
mod_perl related files will now be installed so they are visible
via standard @INC. also, refuse to install over mod_perl 2 versions
less than 1.999_22. [Geoffrey Young]
s/Apache::/Apache2::/g and s/mod_perl/mod_perl2/g in all module
APIs. so, Apache::RequestRec is now Apache2::RequestRec,
Apache::compat is now Apache2::compat, and so on. [joes]
move all Apache:: constants to Apache2::Const and all APR:: constants
to APR::Const. for example, Apache:OK is now Apache2::Const::OK and
APR::SUCCESS is now APR::Const::SUCCESS. [Geoffrey Young]
add $ENV{MOD_PERL_API_VERSION} as something that clearly distinguishes
which mod_perl version is being used at request time. [Geoffrey Young]
rename Apache->request() to Apache2::RequestUtil->request(), and
Apache->server() to Apache2::ServerUtil->server()
[Geoffrey Young]
fix Apache2::Status which was bailing out on trying to load modules
with dev versions like 2.121_02 [Stas]
When parsing Makefile.PL MP_* options, handle correctly the MP_FOO=0
entries [Philip M. Gollucci <pg...@p6m7g8.com>]
init the anonsub hash for base perl and each vhost +Parent (previously
was init'ed only for the base perl) [Stas]
fix a bug when a non-threaded perl is used and anonymous sub is pushed
at the server startup (the CV wasn't surviving) [Stas]
Make sure that CPAN shell doesn't triple over usage of
$ExtUtils::MakeMaker::VERSION [Randy Kobes]
Apache2::RequestRec->new now sets $r->request_time [Stas]
remove CGI.pm and Apache::Request dependencies from Apache2::Status
since they weren't used at all [Geoffrey Young]
Fixes for Apache2::Reload's touchfile feature (return Apache2::Const::OK
instead of 1) [Chris Warren <ch...@cisco.com>]
cygwin fixes: [Nick *** <do...@abv.bg>]
- doesn't like XS wrapper starting with 'static'
- need to compile everything with -DCYGWIN
ModPerl::RegistryCooker API change: s/rewrite_shebang/shebang_to_perl/
the new API now returns the string to prepend before the rest of the
script, instead of rewriting the content, which is both faster and
doesn't mislead the perl debugger [Dominique Quatravaux
<do...@idealx.com>]
Starting from ExtUtils::MakeMaker 6.26 went back to pm_to_blib target
from pm_to_blib.ts introduced in 6.22, so needed to fix the glue_pod
target, so install will work correctly [Stas]
Syntax errors in <Perl> sections were not correctly caught and
reported. [Gozer]
when building mp2 EU::MM looks into Apache-Test/MANIFEST and complains
about the missing Apache-Test/META.yml (which is indeed not included
in the modperl package due to the PAUSE problems of dealing with more
than one META.yml. Solution: Exclude Apache-Test/MANIFEST from
mod_perl distribution package. [Stas]
ModPerl::Registry no longer checks for -x bit (we don't executed
scripts anyway), and thus works on acl-based filesystems. Also
replaced the -r check with a proper error handling when the file is
read in. [Damon Buckwalter <bu...@gmail.com>]
Apache2::RequestUtil::slurp_filename now throws an APR::Error exception
object (before it was just croaking). [Stas]
fix APR::Error's overload of '==' (it was always returning true
before), and add the corresponding '!=' [Stas]
if $r->document_root was modified, restore it at the end of request
[joes]
Apache2::ServerRec method which set the non-integer fields in the
server_rec, now copy the value from the perl scalar, so if it changes
or goes out of scope the C struct is not affected. Using internal perl
variables to preserve the value, since using the server pool to
allocate the memory will mean a memory leak [Stas]
add the escape_url entry in the ModPerl::MethodLookup knowledgebase
[Stas]
Apache2::SubProcess::spawn_proc_prog now can be called in a void
context, in which case all the communication std pipes will be closed
[Stas]
fix a bug in $r->document_root, which previously weren't copying the
new string away [Stas]
introduce a new build option MP_AP_DESTDIR to aid package builders
direct the Apache-specific files to the right place. [Cory Omand
<Co...@Sun.COM>]
Fix bug in modperl_package_clear_stash() segfaulting when
encountering declared but not yet defined subroutines.
[Steve Hay <st...@uk.radan.com>, Gozer]
win32 needs PERL_SYS_INIT3/PERL_SYS_TERM calls [Steve Hay
<st...@uk.radan.com>]
Fix broken MP_STATIC_EXTS=1 build. [Gozer]
Perl -Duse64bit fix. Pointers can't just be generically
casted from/to IVs. Use PTR2IV/INT2PTR instead. [Gozer]
Perl -Duse64bit fix. apr_size_t pointers can't just be generically
casted from/to UVs. Use PTR2UV/INT2PTR instead. [Gozer]
fix a bug in Apache2::Build::dir: If the right directory isn't found in
the for loop $dir still contains a > value, so the ||= has no
effect. [Nick Wellnhofer <we...@aevum.de>]
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: [RELEASE CANDIDATE] mod_perl 2.0-RC5
Posted by "Philippe M. Chiasson" <go...@ectoplasm.org>.
Geoffrey Young wrote:
> the mod_perl development team is pleased to announce that we have a new
> candidate for mod_perl 2.0, ready and waiting for testers.
>
> [...]
Looks like things are borked on OS X ;-(
Darwin (OS X):
Static: bus error on startup (will investigate)
Dynamic:
worker & prefork:
Program received signal EXC_BAD_ACCESS, Could not access memory.
0x87c35bd4 in S_mess_alloc ()
(gdb) bt
#0 0x87c35bd4 in S_mess_alloc ()
#1 0x87c35ec0 in Perl_vmess ()
#2 0x87c3697c in Perl_vcroak ()
#3 0x87c36e2c in Perl_croak_nocontext ()
#4 0x01016b18 in modperl_global_cleanup (data=0x0) at modperl_global.c:80
#5 0x000712ac in run_cleanups (cref=0x1806228) at apr_pools.c:1952
#6 0x00070940 in apr_pool_clear (pool=0xbffffb1c) at apr_pools.c:694
#7 0x00024568 in main (argc=11, argv=0xbffffd34) at main.c:575
I'll investigate some more and post more detailled backtraces.
--------------------------------------------------------------------------------
Philippe M. Chiasson m/gozer\@(apache|cpan|ectoplasm)\.org/ GPG KeyID : 88C3A5A5
http://gozer.ectoplasm.org/ F9BF E0C2 480E 7680 1AE5 3631 CB32 A107 88C3A5A5
Re: [RELEASE CANDIDATE] mod_perl 2.0-RC5
Posted by "Philippe M. Chiasson" <go...@ectoplasm.org>.
Geoffrey Young wrote:
> the mod_perl development team is pleased to announce that we have a new
> candidate for mod_perl 2.0, ready and waiting for testers.
>
> [...]
Looks like things are borked on OS X ;-(
Darwin (OS X):
Static: bus error on startup (will investigate)
Dynamic:
worker & prefork:
Program received signal EXC_BAD_ACCESS, Could not access memory.
0x87c35bd4 in S_mess_alloc ()
(gdb) bt
#0 0x87c35bd4 in S_mess_alloc ()
#1 0x87c35ec0 in Perl_vmess ()
#2 0x87c3697c in Perl_vcroak ()
#3 0x87c36e2c in Perl_croak_nocontext ()
#4 0x01016b18 in modperl_global_cleanup (data=0x0) at modperl_global.c:80
#5 0x000712ac in run_cleanups (cref=0x1806228) at apr_pools.c:1952
#6 0x00070940 in apr_pool_clear (pool=0xbffffb1c) at apr_pools.c:694
#7 0x00024568 in main (argc=11, argv=0xbffffd34) at main.c:575
I'll investigate some more and post more detailled backtraces.
--------------------------------------------------------------------------------
Philippe M. Chiasson m/gozer\@(apache|cpan|ectoplasm)\.org/ GPG KeyID : 88C3A5A5
http://gozer.ectoplasm.org/ F9BF E0C2 480E 7680 1AE5 3631 CB32 A107 88C3A5A5
Re: [RELEASE CANDIDATE] mod_perl 2.0-RC5 {OK on OSX 10.3.8 w/ Apache
2.1.5-dev}
Posted by OpenMacNews <Op...@speakeasy.net>.
hi all,
-- On April 11, 2005 1:44:12 PM -0400 "Philip M. Gollucci"
<pg...@p6m7g8.com> wrote:
>> 2.1 without issue, and I think there are several over on dev@ that are doing
> That'd be me... I'm using 2.1.5-dev w/ mp2
fwiw, today's HEAD looks OK on my OSX 10.3.8 dev-box ...
Server Version: Apache/2.1.5-dev (Unix) DAV/2 PHP/5.0.4 SVN/1.1.3
mod_perl/1.999.22-dev Perl/v5.8.6
Server Built: Apr 7 2005 14:57:22
Module Magic Number: 20050305:0
Hostname/port: dev.internal.testdomain.com:80
Timeouts: connection: 300 keep-alive: 300
MPM Name: Worker
MPM Information: Max Daemons: 1 Threaded: yes Forked: yes
Server Architecture: 32-bit
Server Root: /usr/local/apache2
Config File: /var/Apache2/httpd.conf
Server Built With: -D APACHE_MPM_DIR="server/mpm/worker"
-D APR_HAS_MMAP
-D APR_USE_SYSVSEM_SERIALIZE
-D SINGLE_LISTEN_UNSERIALIZED_ACCEPT
-D APR_HAS_OTHER_CHILD
-D AP_HAVE_RELIABLE_PIPED_LOGS
-D HTTPD_ROOT="/usr/local/apache2"
-D SUEXEC_BIN="/usr/local/apache2/bin/suexec"
-D DEFAULT_ERRORLOG="logs/error_log"
-D AP_TYPES_CONFIG_FILE="/var/Apache2/mime.types"
-D SERVER_CONFIG_FILE="/var/Apache2/httpd.conf"
Re: [RELEASE CANDIDATE] mod_perl 2.0-RC5
Posted by "Philip M. Gollucci" <pg...@p6m7g8.com>.
>2.1 without issue, and I think there are several over on dev@ that are doing
That'd be me... I'm using 2.1.5-dev w/ mp2
Re: [RELEASE CANDIDATE] mod_perl 2.0-RC5
Posted by Geoffrey Young <ge...@modperlcookbook.org>.
Jonathan Vanasco wrote:
>
> On Apr 11, 2005, at 10:05 AM, Geoffrey Young wrote:
>
>> *** YOU WILL NEED TO MODIFY ALL YOUR CODE AFTER UPGRADING ***
>
>
> Is there going to be another one of these for mod_perl under Apache2.1 ?
not likely, but don't confuse the issue - really, the code migration forced
by this release has nothing to do with the Apache-side of the API at all.
as it stands now, mod_perl recent 1.99 versions work with both 2.0 and 2.1
just fine so long as you stick to 2.0 features. meaning, for example, you
can't use mp2 for coding authentication providers at the moment. however,
there's a CPAN module that lets you do that, so as far as I know you can use
2.1 without issue, and I think there are several over on dev@ that are doing
just that.
--Geoff
Re: [RELEASE CANDIDATE] mod_perl 2.0-RC5
Posted by "Philip M. Gollucci" <pg...@p6m7g8.com>.
Jonathan Vanasco wrote:
>
> On Apr 11, 2005, at 10:05 AM, Geoffrey Young wrote:
>
>> *** YOU WILL NEED TO MODIFY ALL YOUR CODE AFTER UPGRADING ***
>
>
> Is there going to be another one of these for mod_perl under Apache2.1 ?
I don't think so... The most radically different thing is the auth modules
and that is all additions.
This is because of the Apache -> Apache2 and mod_perl -> mod_perl2 rename.
--
END
-----------------------------------------------------------------------------
Philip M. Gollucci
Senior Developer - Liquidity Services Inc.
Phone: 202.568.6268 (Direct)
E-Mail: pgollucci@liquidation.com
Web: http://www.liquidation.com
Re: [RELEASE CANDIDATE] mod_perl 2.0-RC5
Posted by Jonathan Vanasco <jv...@mastersofbranding.com>.
On Apr 11, 2005, at 10:05 AM, Geoffrey Young wrote:
> *** YOU WILL NEED TO MODIFY ALL YOUR CODE AFTER UPGRADING ***
Is there going to be another one of these for mod_perl under Apache2.1 ?
Re: [RELEASE CANDIDATE] mod_perl 2.0-RC5
Posted by Torsten Foertsch <to...@gmx.net>.
On Monday 11 April 2005 16:05, Geoffrey Young wrote:
> http://perl.apache.org/docs/2.0/rename.html
Does that mean that new modules working only with MP2 should better be named
Apache2::...?
I think, yes.
Torsten
Re: [RELEASE CANDIDATE] mod_perl 2.0-RC5
Posted by "Philip M. Gollucci" <pg...@p6m7g8.com>.
Geoffrey Young wrote:
> the mod_perl development team is pleased to announce that we have a new
> candidate for mod_perl 2.0, ready and waiting for testers.
>
FBSD 6-current
bleed perl w/out ithreads
http2/prefork apr is not threaded
modules/proxy.t fails for me
error log and test output attached.
Everything else works.
FYI: CGI 3.08 is not yet available via CPAN at the posted link.
I had to patch it again with mp2.patch
--
END
------------------------------------------------------
Philip M. Gollucci (pgollucci@p6m7g8.com) 301.254.5198
Consultant / http://p6m7g8.net/Resume/resume.shtml
Senior Developer / Liquidity Services, Inc.
http://www.liquidityservicesinc.com
Re: [RELEASE CANDIDATE] mod_perl 2.0-RC5
Posted by Randy Kobes <ra...@theoryx5.uwinnipeg.ca>.
On Tue, 12 Apr 2005, Steve Hay wrote:
> Geoffrey Young wrote:
>
> >the mod_perl development team is pleased to announce that we have a new
> >candidate for mod_perl 2.0, ready and waiting for testers.
> >
> No good for me on Win32 :(
>
> Using Apache 2.0.53, bleadperl (@22511), I have the testsuite crashing
> the Apache.exe server every time in perl/ithreads_args.t. Here's a
> stacktrace fwiw:
[ ... ]
> At this point, my Apache.exe process is a staggering 205 MB!!!
>
> I'm using Perl's malloc here. Randy, do you see the same with your
> ActivePerl (i.e. system malloc) setup?
No, I don't ...
--
best regards,
randy
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: [RELEASE CANDIDATE] mod_perl 2.0-RC5
Posted by Stas Bekman <st...@stason.org>.
Steve Hay wrote:
> Geoffrey Young wrote:
>
>
>>the mod_perl development team is pleased to announce that we have a new
>>candidate for mod_perl 2.0, ready and waiting for testers.
>>
>
> No good for me on Win32 :(
>
> Using Apache 2.0.53, bleadperl (@22511), I have the testsuite crashing
> the Apache.exe server every time in perl/ithreads_args.t. Here's a
> stacktrace fwiw:
>
> Perl_mg_free(interpreter * 0x06809098, sv * 0x110affc8) line 406 + 3 bytes
[...]
> Needless to say, the perl/ithreads_args.t test runs fine in isolation.
> Maybe we should drop this ithreads test from the distro too?
Done. Hopefully the situation will improve once p5p will give us perl API
to prevent incorrect cloning.
I've excluded all threads tests from the distro.
--
__________________________________________________________________
Stas Bekman JAm_pH ------> Just Another mod_perl Hacker
http://stason.org/ mod_perl Guide ---> http://perl.apache.org
mailto:stas@stason.org http://use.perl.org http://apacheweek.com
http://modperlbook.org http://apache.org http://ticketmaster.com
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: [RELEASE CANDIDATE] mod_perl 2.0-RC5
Posted by Steve Hay <st...@uk.radan.com>.
Geoffrey Young wrote:
>the mod_perl development team is pleased to announce that we have a new
>candidate for mod_perl 2.0, ready and waiting for testers.
>
No good for me on Win32 :(
Using Apache 2.0.53, bleadperl (@22511), I have the testsuite crashing
the Apache.exe server every time in perl/ithreads_args.t. Here's a
stacktrace fwiw:
Perl_mg_free(interpreter * 0x06809098, sv * 0x110affc8) line 406 + 3 bytes
Perl_sv_clear(interpreter * 0x06809098, sv * 0x110affc8) line 5958 + 13
bytes
Perl_sv_free2(interpreter * 0x06809098, sv * 0x110affc8) line 6191 + 13
bytes
Perl_sv_free(interpreter * 0x06809098, sv * 0x110affc8) line 6171 + 13 bytes
Perl_ithread_create(interpreter * 0x04d942e8, sv * 0x00000000, char *
0x090a2608, sv * 0x10722534, sv * 0x104213b4) line 476 + 13 bytes
XS_threads_new(interpreter * 0x04d942e8, cv * 0x08f13eb0) line 687 + 36
bytes
Perl_pp_entersub(interpreter * 0x04d942e8) line 2816 + 16 bytes
Perl_runops_debug(interpreter * 0x04d942e8) line 1581 + 13 bytes
S_call_body(interpreter * 0x04d942e8, const op * 0x0adefd20, char 0)
line 2211 + 13 bytes
Perl_call_sv(interpreter * 0x04d942e8, sv * 0x1042010c, long 4) line
2143 + 15 bytes
modperl_callback(interpreter * 0x04d942e8, modperl_handler_t *
0x0c3e31f8, apr_pool_t * 0x0c3dec88, request_rec * 0x0c3decc0,
server_rec * 0x0029ce70, av * 0x10718e74) line 100 + 17 bytes
modperl_callback_run_handlers(int 6, int 4, request_rec * 0x0c3decc0,
conn_rec * 0x00000000, server_rec * 0x0029ce70, apr_pool_t * 0x00000000,
apr_pool_t * 0x00000000, apr_pool_t * 0x00000000, int 1) line 261 + 35 bytes
modperl_callback_per_dir(int 6, request_rec * 0x0c3decc0, int 1) line
370 + 34 bytes
modperl_response_handler_run(request_rec * 0x0c3decc0, int 1) line 976 +
13 bytes
modperl_response_handler(request_rec * 0x0c3decc0) line 1016 + 11 bytes
ap_run_handler(request_rec * 0x0c3decc0) line 153 + 78 bytes
ap_invoke_handler(request_rec * 0x0c3decc0) line 364 + 9 bytes
ap_process_request(request_rec * 0x0c3decc0) line 249 + 9 bytes
ap_process_http_connection(conn_rec * 0x00a6b4f8) line 251 + 9 bytes
ap_run_process_connection(conn_rec * 0x00a6b4f8) line 43 + 78 bytes
ap_process_connection(conn_rec * 0x00a6b4f8, void * 0x00a6b428) line 178
worker_main(long 1) line 730
_threadstartex(void * 0x0029d568) line 212 + 13 bytes
KERNEL
At this point, my Apache.exe process is a staggering 205 MB!!!
I'm using Perl's malloc here. Randy, do you see the same with your
ActivePerl (i.e. system malloc) setup?
Needless to say, the perl/ithreads_args.t test runs fine in isolation.
Maybe we should drop this ithreads test from the distro too?
- Steve
------------------------------------------------
Radan Computational Ltd.
The information contained in this message and any files transmitted with it are confidential and intended for the addressee(s) only. If you have received this message in error or there are any problems, please notify the sender immediately. The unauthorized use, disclosure, copying or alteration of this message is strictly forbidden. Note that any views or opinions presented in this email are solely those of the author and do not necessarily represent those of Radan Computational Ltd. The recipient(s) of this message should check it and any attached files for viruses: Radan Computational will accept no liability for any damage caused by any virus transmitted by this email.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org