Re: [ANNOUNCE] mod_perl-2.0.13

[Steve Hay]

Thanks. Yes, there was an indexing problem on CPAN. It made it this far, though:

https://cpan.metacpan.org/authors/id/S/SH/SHAY/mod_perl-2.0.13.tar.gz

Hopefully the indexing will get fixed soon.


On Sat, 21 Oct 2023 at 12:08, Ian B  wrote:
>
> Thanks for this, just a note, 
> https://metacpan.org/release/SHAY/mod_perl-2.0.13 comes up not found.
>
>
>
> On Sat, Oct 21, 2023 at 11:44 AM Steve Hay  wrote:
>>
>> We are pleased to announce the release of mod_perl 2.0.13
>>
>> mod_perl is an Apache HTTP Server module for embedding a Perl
>> interpreter in your web server, giving you super-fast dynamic content
>> by avoiding the overhead of starting an external interpreter.
>>
>> This release is now, or soon will be, available for download from a
>> mirror site near you via:
>>
>> https://perl.apache.org/download/index.html
>>
>> or in the meantime directly from:
>>
>> https://metacpan.org/release/SHAY/mod_perl-2.0.13
>>
>> Checksums for this release are:
>>
>> SHA256:
>> mod_perl-2.0.13.tar.gz: ADE3BE31 C447B844 8869FECD FCACE258 D6D587B8 C6C773C5
>> F22735F7 0D82D6DA
>>
>> SHA512:
>> mod_perl-2.0.13.tar.gz: 453C7897 43487E2D 1167D005 F21EF978 0F36488D B8830103
>> E4EA0A06 56F685BF 5EC423CD 1BAA68EE 30EFC75D 9E2B845C
>> 96564582 03B9F48D E81A315A 91D587C9
>>
>> Major changes in this release are as follows:
>>
>> Use get_server_banner() instead of deprecated get_server_version() in
>> Apache2::Status.  [Petr Písař >
>> Avoid generating APR precompiled headers. [Sam James ]
>>
>> Fix build for perl >= 5.37.1. [Jitka Plesnikova ]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[ANNOUNCE] mod_perl-2.0.13

[Steve Hay]

We are pleased to announce the release of mod_perl 2.0.13

mod_perl is an Apache HTTP Server module for embedding a Perl
interpreter in your web server, giving you super-fast dynamic content
by avoiding the overhead of starting an external interpreter.

This release is now, or soon will be, available for download from a
mirror site near you via:

https://perl.apache.org/download/index.html

or in the meantime directly from:

https://metacpan.org/release/SHAY/mod_perl-2.0.13

Checksums for this release are:

SHA256:
mod_perl-2.0.13.tar.gz: ADE3BE31 C447B844 8869FECD FCACE258 D6D587B8 C6C773C5
F22735F7 0D82D6DA

SHA512:
mod_perl-2.0.13.tar.gz: 453C7897 43487E2D 1167D005 F21EF978 0F36488D B8830103
E4EA0A06 56F685BF 5EC423CD 1BAA68EE 30EFC75D 9E2B845C
96564582 03B9F48D E81A315A 91D587C9

Major changes in this release are as follows:

Use get_server_banner() instead of deprecated get_server_version() in
Apache2::Status.  [Petr Písař ]

Fix build for perl >= 5.37.1. [Jitka Plesnikova ]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] mod_perl-2.0.13 RC1

[Steve Hay]

Fred, are you still looking into this, or shall we assume it's
Apache::Test related and proceed with this mod_perl release?

On Sat, 26 Aug 2023 at 17:20, Steve Hay  wrote:
>
> Thanks. Is it the same problem that Adam has reproduced?
>
> If so then perhaps we can go ahead with the release and deal with this
> issue afterwards since it seems to be more related to Apache::Test
> than mod_perl.
>
> On Thu, 24 Aug 2023 at 05:55, Fred Moyer  wrote:
> >
> > On Fri, Aug 18, 2023 at 2:47 AM Steve Hay  
> > wrote:
> > > Were there any build errors/warnings?
> > > Is this a clean set-up, or do either httpd or perl have an old
> > > mod_perl in them already?
> >
> > Digging into this; appears to be a problem with the perl I build mp
> > with and the system mod_perl. Will keep you posted.
> >
> >
> > >
> > >
> > > >
> > > > On Sun, Aug 6, 2023 at 5:00 AM Steve Hay  wrote:
> > > > >
> > > > > Please download, test, and report back on this mod_perl 2.0.13 release
> > > > > candidate.
> > > > >
> > > > > https://dist.apache.org/repos/dist/dev/perl/mod_perl-2.0.13-rc1.tar.gz
> > > > >
> > > > > SHA256:
> > > > > mod_perl-2.0.13-rc1.tar.gz: EC72BC99 EC01DAE5 BEB1DC8F 2F3807CB 
> > > > > A4A9BF4D
> > > > > D492E846 858DABC0 EEA8E6B7
> > > > >
> > > > > SHA512:
> > > > > mod_perl-2.0.13-rc1.tar.gz: 15EA2E30 B5BE0B76 7E49FC46 20F93AA0 
> > > > > FF4E9EBA
> > > > > 03E7EDFD 64166223 5C1E652B B3C1CB69 
> > > > > CB60DF33
> > > > > C64121C5 C74F3ABB A9ACDB6B 19E9B19A 
> > > > > 46790DFE
> > > > > 015C77C9
> > > > >
> > > > > Major changes in this release are as follows:
> > > > >
> > > > > Use get_server_banner() instead of deprecated get_server_version() in
> > > > > Apache2::Status.  [Petr Písař  > > > >
> > > > > Avoid generating APR precompiled headers. [Sam James 
> > > > > ]
> > > > >
> > > > > Fix build for perl >= 5.37.1. [Jitka Plesnikova ]
> > > > >
> > > > > -
> > > > > To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
> > > > > For additional commands, e-mail: dev-h...@perl.apache.org
> > > > >
> > > >
> > > > -
> > > > To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
> > > > For additional commands, e-mail: dev-h...@perl.apache.org
> > > >

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] mod_perl-2.0.13 RC1

[Steve Hay]

Thanks. Is it the same problem that Adam has reproduced?

If so then perhaps we can go ahead with the release and deal with this
issue afterwards since it seems to be more related to Apache::Test
than mod_perl.

On Thu, 24 Aug 2023 at 05:55, Fred Moyer  wrote:
>
> On Fri, Aug 18, 2023 at 2:47 AM Steve Hay  wrote:
> > Were there any build errors/warnings?
> > Is this a clean set-up, or do either httpd or perl have an old
> > mod_perl in them already?
>
> Digging into this; appears to be a problem with the perl I build mp
> with and the system mod_perl. Will keep you posted.
>
>
> >
> >
> > >
> > > On Sun, Aug 6, 2023 at 5:00 AM Steve Hay  wrote:
> > > >
> > > > Please download, test, and report back on this mod_perl 2.0.13 release
> > > > candidate.
> > > >
> > > > https://dist.apache.org/repos/dist/dev/perl/mod_perl-2.0.13-rc1.tar.gz
> > > >
> > > > SHA256:
> > > > mod_perl-2.0.13-rc1.tar.gz: EC72BC99 EC01DAE5 BEB1DC8F 2F3807CB A4A9BF4D
> > > > D492E846 858DABC0 EEA8E6B7
> > > >
> > > > SHA512:
> > > > mod_perl-2.0.13-rc1.tar.gz: 15EA2E30 B5BE0B76 7E49FC46 20F93AA0 FF4E9EBA
> > > > 03E7EDFD 64166223 5C1E652B B3C1CB69 CB60DF33
> > > > C64121C5 C74F3ABB A9ACDB6B 19E9B19A 46790DFE
> > > > 015C77C9
> > > >
> > > > Major changes in this release are as follows:
> > > >
> > > > Use get_server_banner() instead of deprecated get_server_version() in
> > > > Apache2::Status.  [Petr Písař  > > >
> > > > Avoid generating APR precompiled headers. [Sam James ]
> > > >
> > > > Fix build for perl >= 5.37.1. [Jitka Plesnikova ]
> > > >
> > > > -
> > > > To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
> > > > For additional commands, e-mail: dev-h...@perl.apache.org
> > > >
> > >
> > > -
> > > To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
> > > For additional commands, e-mail: dev-h...@perl.apache.org
> > >

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] mod_perl-2.0.13 RC1

[Steve Hay]

On Fri, 18 Aug 2023 at 06:11, Fred Moyer  wrote:
>
> Hitting an issue on MacOS, perl 5.36.1, httpd 2.4.57. Not sure if this
> is me out of practice or not. Do I need to build against httpd 2.2?
>
> ulimit -c unlimited; /opt/homebrew/Cellar/perl/5.36.1/bin/perl
> /Users/phred/dev/mod_perl-2.0.13-rc1/t/TEST -bugreport -verbose=0
> /opt/homebrew/opt/httpd/bin/httpd  -d
> /Users/phred/dev/mod_perl-2.0.13-rc1/t -f
> /Users/phred/dev/mod_perl-2.0.13-rc1/t/conf/httpd.conf -D APACHE2 -D
> APACHE2_4 -D PERL_USEITHREADS
> using Apache/2.4.57 (prefork MPM)
>
> waiting 300 seconds for server to start: .httpd: Syntax error on line
> 85 of /Users/phred/dev/mod_perl-2.0.13-rc1/t/conf/httpd.conf: Cannot
> load /Users/phred/dev/mod_perl-2.0.13-rc1/src/modules/perl/mod_perl.so
> into server: 
> dlopen(/Users/phred/dev/mod_perl-2.0.13-rc1/src/modules/perl/mod_perl.so,
> 0x000A): symbol not found in flat namespace
> '_modperl_handler_anon_add'
> .^C[warning]
> halting tests

Sorry, I don't know what the problem is there. It should certainly
work with htpd-2.4.

Were there any build errors/warnings?
Is this a clean set-up, or do either httpd or perl have an old
mod_perl in them already?


>
> On Sun, Aug 6, 2023 at 5:00 AM Steve Hay  wrote:
> >
> > Please download, test, and report back on this mod_perl 2.0.13 release
> > candidate.
> >
> > https://dist.apache.org/repos/dist/dev/perl/mod_perl-2.0.13-rc1.tar.gz
> >
> > SHA256:
> > mod_perl-2.0.13-rc1.tar.gz: EC72BC99 EC01DAE5 BEB1DC8F 2F3807CB A4A9BF4D
> > D492E846 858DABC0 EEA8E6B7
> >
> > SHA512:
> > mod_perl-2.0.13-rc1.tar.gz: 15EA2E30 B5BE0B76 7E49FC46 20F93AA0 FF4E9EBA
> > 03E7EDFD 64166223 5C1E652B B3C1CB69 CB60DF33
> > C64121C5 C74F3ABB A9ACDB6B 19E9B19A 46790DFE
> > 015C77C9
> >
> > Major changes in this release are as follows:
> >
> > Use get_server_banner() instead of deprecated get_server_version() in
> > Apache2::Status.  [Petr Písař  >
> > Avoid generating APR precompiled headers. [Sam James ]
> >
> > Fix build for perl >= 5.37.1. [Jitka Plesnikova ]
> >
> > -
> > To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
> > For additional commands, e-mail: dev-h...@perl.apache.org
> >
>
> -
> To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
> For additional commands, e-mail: dev-h...@perl.apache.org
>

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] mod_perl-2.0.13 RC1

[Steve Hay]

+1 on Windows with Apache 2.4.57 / Perl 5.38.0 / VC2022.

With LWP:
t\filter\in_bbs_inject_header.t   (Wstat: 0 Tests: 36 Failed: 3)
  Failed tests:  22, 26, 30
t\modperl\env.t   (Wstat: 0 Tests: 164 Failed: 17)
  Failed tests:  25, 28, 31, 43, 46, 49, 55, 58, 73, 82, 85, 88, 91,
97, 142, 145, 163
t\modules\cgi.t   (Wstat: 0 Tests: 5 Failed: 1)
  Failed test:  3
t\modules\cgi2.t  (Wstat: 0 Tests: 5 Failed: 1)
  Failed test:  3
t\modules\cgiupload.t (Wstat: 0 Tests: 2 Failed: 2)
  Failed tests:  1-2
t\modules\cgiupload2.t(Wstat: 0 Tests: 2 Failed: 2)
  Failed tests:  1-2

Without LWP:
t\modperl\env.t   (Wstat: 0 Tests: 161 Failed: 18)
  Failed tests:  25, 28, 31, 34, 46, 49, 52, 58, 61, 70, 79, 82, 85,
88, 94, 139, 142, 160

(Results are the same as 2.0.12.)

On Sun, 6 Aug 2023 at 13:00, Steve Hay  wrote:
>
> Please download, test, and report back on this mod_perl 2.0.13 release
> candidate.
>
> https://dist.apache.org/repos/dist/dev/perl/mod_perl-2.0.13-rc1.tar.gz
>
> SHA256:
> mod_perl-2.0.13-rc1.tar.gz: EC72BC99 EC01DAE5 BEB1DC8F 2F3807CB A4A9BF4D
> D492E846 858DABC0 EEA8E6B7
>
> SHA512:
> mod_perl-2.0.13-rc1.tar.gz: 15EA2E30 B5BE0B76 7E49FC46 20F93AA0 FF4E9EBA
> 03E7EDFD 64166223 5C1E652B B3C1CB69 CB60DF33
> C64121C5 C74F3ABB A9ACDB6B 19E9B19A 46790DFE
> 015C77C9
>
> Major changes in this release are as follows:
>
> Use get_server_banner() instead of deprecated get_server_version() in
> Apache2::Status.  [Petr Písař 
> Avoid generating APR precompiled headers. [Sam James ]
>
> Fix build for perl >= 5.37.1. [Jitka Plesnikova ]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[RELEASE CANDIDATE] mod_perl-2.0.13 RC1

[Steve Hay]

Please download, test, and report back on this mod_perl 2.0.13 release
candidate.

https://dist.apache.org/repos/dist/dev/perl/mod_perl-2.0.13-rc1.tar.gz

SHA256:
mod_perl-2.0.13-rc1.tar.gz: EC72BC99 EC01DAE5 BEB1DC8F 2F3807CB A4A9BF4D
D492E846 858DABC0 EEA8E6B7

SHA512:
mod_perl-2.0.13-rc1.tar.gz: 15EA2E30 B5BE0B76 7E49FC46 20F93AA0 FF4E9EBA
03E7EDFD 64166223 5C1E652B B3C1CB69 CB60DF33
C64121C5 C74F3ABB A9ACDB6B 19E9B19A 46790DFE
015C77C9

Major changes in this release are as follows:

Use get_server_banner() instead of deprecated get_server_version() in
Apache2::Status.  [Petr Písař ]

Fix build for perl >= 5.37.1. [Jitka Plesnikova ]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] Apache-SizeLimit-0.98 RC1

[Steve Hay]

Thanks.

This is not implemented on Windows so I'm unable to test it myself.

I've pushed one further minor update to avoid a (long-standing)
warning when building inside the mod_perl build (see
https://svn.apache.org/viewvc?view=revision&revision=1911480) but I
don't think it warrants another RC. It will get retested (and more
meaningfully) when testing the mod_perl RC which is coming next...

This standalone Apache::SizeLimit release doesn't go on
https://dist.apache.org/ (it's just released to CPAN) so I will crack
on with that now even though we only have one +1.

On Sat, 29 Jul 2023 at 15:24, Adam Prime  wrote:
>
> +1
>
> server localhost:8529 started
> t/apache/all.t  skipped: apache version 1 required,
> this is version 2
> t/apache2/basic.t . ok
> t/apache2/check_n_requests2.t . ok,
> t/apache2/deprecated.t  ok
> t/apache2/zzz_check_n_requests.t .. ok
> t/pod.t ... ok
> All tests successful.
> Files=6, Tests=38,  3 wallclock secs ( 0.03 usr  0.00 sys +  1.66 cusr
> 0.19 csys =  1.88 CPU)
> Result: PASS
>
>
> All tests pass on linux with apache 2.4.41 prefork and perl 5.28.0
>
> Adam
>
> On 7/21/23 18:56, Steve Hay wrote:
> > In preparation for an upcoming release of mod_perl to support Perl
> > 5.38.0, I am making a new release of Apache-SizeLimit.
> >
> > Please download, test, and report back on this Apache-SizeLimit 0.98
> > release candidate.
> >
> > https://people.apache.org/~stevehay/Apache-SizeLimit-0.98-rc1.tar.gz
> >
> > MD5  = 8acfa750da6c4f4ae3f328e117dfa42c
> > SHA256 = 4a6cd22067d93bfd2e43ae1fb5902942576e7af6b0a7c2a1c466d550b573
> >
> > (NOTE: Apache::SizeLimit is not implemented on Windows.)
> >
> > Major changes in this release are as follows:
> >
> > Overhaul $VERSION numbering scheme to align with other mod_perl
> > modules and avoid warnings from ExtUtils::MakeMaker.
> > [Steve Hay]
> >
> > Remove use of Linux::Smaps - it is slow and uses wrong statistics anyway.
> > [Zefram , CPAN RT#93757]
> >
> > Fix unshared memory values when using /proc/self/statm.
> > [Zefram , CPAN RT#73752]
> >
> > Fix unnecessary FAIL reports on CPAN Testers by checking for mod_perl
> > or mod_perl2.
> > [Andreas Koenig ; Steve Hay]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] Apache-Reload-0.14 RC1

[Steve Hay]

Thanks.

+1 here with Apache 2.4.57 / Perl 5.38.0 / VC2022 on Windows 11.

I've pushed one further minor update to avoid a (long-standing)
warning when building inside the mod_perl build (see
https://svn.apache.org/viewvc?view=revision&revision=1911479) but I
don't think it warrants another RC. It will get retested (and more
meaningfully) when testing the mod_perl RC which is coming next...

This standalone Apache::Reload release doesn't go on
https://dist.apache.org/ (it's just released to CPAN) so I will crack
on with that now even though we only have two +1s.

On Sat, 29 Jul 2023 at 15:17, Adam Prime  wrote:
>
> +1
>
> tests pass on apache 2.4.41 w/ perl 5.28.0, which is what i've got
> immediately available to test with.
>
> Adam
>
> On 7/21/23 18:56, Steve Hay wrote:
> > In preparation for an upcoming release of mod_perl to support Perl
> > 5.38.0, I am making a new release of Apache-Reload.
> >
> > Please download, test, and report back on this Apache-Reload 0.14
> > release candidate.
> >
> > https://people.apache.org/~stevehay/Apache-Reload-0.14-rc1.tar.gz
> >
> > MD5  = 53a2201f98d05b9868581b730d2032c3
> > SHA256 = 38fcc0234e9db91624b5bc704ffe93928aa3f1d044bcc26f4839787b21123769
> >
> > Major changes in this release are as follows:
> >
> > Fix unnecessary FAIL reports on CPAN Testers by checking for mod_perl
> > or mod_perl2.
> > [Andreas Koenig , CPAN RT#34316; Steve Hay]
> >
> > Fix spelling error.
> > [, CPAN RT#127004]
> >
> > Fix Makefile.PL when Apache::Test is not present.
> > [Nigel Gregoire, CPAN RT#113778]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[RELEASE CANDIDATE] Apache-SizeLimit-0.98 RC1

[Steve Hay]

In preparation for an upcoming release of mod_perl to support Perl
5.38.0, I am making a new release of Apache-SizeLimit.

Please download, test, and report back on this Apache-SizeLimit 0.98
release candidate.

https://people.apache.org/~stevehay/Apache-SizeLimit-0.98-rc1.tar.gz

MD5  = 8acfa750da6c4f4ae3f328e117dfa42c
SHA256 = 4a6cd22067d93bfd2e43ae1fb5902942576e7af6b0a7c2a1c466d550b573

(NOTE: Apache::SizeLimit is not implemented on Windows.)

Major changes in this release are as follows:

Overhaul $VERSION numbering scheme to align with other mod_perl
modules and avoid warnings from ExtUtils::MakeMaker.
[Steve Hay]

Remove use of Linux::Smaps - it is slow and uses wrong statistics anyway.
[Zefram , CPAN RT#93757]

Fix unshared memory values when using /proc/self/statm.
[Zefram , CPAN RT#73752]

Fix unnecessary FAIL reports on CPAN Testers by checking for mod_perl
or mod_perl2.
[Andreas Koenig ; Steve Hay]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[RELEASE CANDIDATE] Apache-Reload-0.14 RC1

[Steve Hay]

In preparation for an upcoming release of mod_perl to support Perl
5.38.0, I am making a new release of Apache-Reload.

Please download, test, and report back on this Apache-Reload 0.14
release candidate.

https://people.apache.org/~stevehay/Apache-Reload-0.14-rc1.tar.gz

MD5  = 53a2201f98d05b9868581b730d2032c3
SHA256 = 38fcc0234e9db91624b5bc704ffe93928aa3f1d044bcc26f4839787b21123769

Major changes in this release are as follows:

Fix unnecessary FAIL reports on CPAN Testers by checking for mod_perl
or mod_perl2.
[Andreas Koenig , CPAN RT#34316; Steve Hay]

Fix spelling error.
[, CPAN RT#127004]

Fix Makefile.PL when Apache::Test is not present.
[Nigel Gregoire, CPAN RT#113778]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: Patch to avoid mod_perl generating APR pch

[Steve Hay]

On Wed, 5 Jul 2023 at 01:37, Sam James  wrote:
>
> `cpp` seems to try generate precompiled headers (.pch/.gch files)
> as of 
> https://github.com/apache/mod_perl/commit/9dac4cf5580660720bbc18b51bcef7c9be75d658.
>
> I've attached a patch which should suppress that which we're using in
> Gentoo.
>

Many thanks. Applied here:
https://svn.apache.org/viewvc?view=revision&revision=1910794

Hopefully there will be a 2.0.13 release soon.

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RESULT: PASS] Re: [VOTE] Release libapreq2-2.17

[Steve Hay]

On Thu, 25 Aug 2022 at 09:58, Joe Orton  wrote:
>
> Thanks for testing. The release is approved:
>
> PMC votes: +1 from ylavic, jfclere, jorton
>
> I will promote the release and announce it.
>

Thanks for this release. I didn't get round to uploading it to CPAN
yet, but a bug report has come in for it anyway:

https://rt.cpan.org/Public/Bug/Display.html?id=144470

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

New Windows mod_perl2/libapreq2 build available

[Steve Hay]

For those who are interested, I have uploaded a 64-bit build of
mod_perl-2.0.12 and libapreq2-2.16 components compatible with Apache
Lounge 2.4.52 (Win64, VC16, built 20 December 2021) and Strawberry
Perl 5.32.1.1 (64bit).

You can download it from: https://home.apache.org/~stevehay/

The SHA1 digest is: 53f44d4b9c956d60b8637daf59e92c36eb2ac7f9

Regards,
Steve

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[ANNOUNCE] mod_perl-2.0.12

[Steve Hay]

We are pleased to announce the release of mod_perl 2.0.12

mod_perl is an Apache HTTP Server module for embedding a Perl
interpreter in your web server, giving you super-fast dynamic content
by avoiding the overhead of starting an external interpreter.

This release is now, or soon will be, available for download from a
mirror site near you via:

https://perl.apache.org/download/index.html

or in the meantime directly from:

https://metacpan.org/release/SHAY/mod_perl-2.0.12

Checksums for this release are:

SHA256:
f5b821b59b0fdc9670e46ed0fcf32d8911f25126189a8b68c1652f9221eee269

SHA512:
890dca0950847e32180485cabbeffbf236af2c92c7df957a233c210022b5172957eddb1db3e9281b87cd438d2fa404a05ae99c7eda098267c68d5e9262b400b0

Major changes in this release are as follows:

Add bug tracker information to README, and add CONTRIBUTING.md. [Steve Hay]

Fix detection of APR's threading support on RHEL 8. [Petr Písař
]

Fix build for perl >= 5.33.7. [Leon Timmermans ]

Fix SIGSEGV crash due to wrong use of perl_parse(). [Charles Pigott
]

Improve installation instructions for AIX. [Rainer Tammer
]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] mod_perl-2.0.12 RC2

[Steve Hay]

On Sat, 18 Dec 2021 at 11:21, Steve Hay  wrote:
>
> Please download, test, and report back on this mod_perl 2.0.12 release
> candidate.
>

Still waiting to see the necessary votes from other committers before
I can release this.

FWIW it's all good here (Windows 10) with httpd 2.4.51 / perl 5.34.0.

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: buildbot failure in on mod_perl_2_trunk

[Steve Hay]

On Sat, 18 Dec 2021 at 18:06, Edward J. Sabol  wrote:
>
> On Dec 18, 2021, at 6:15 AM, build...@apache.org wrote:
>
> The Buildbot has detected a failed build on builder mod_perl_2_trunk while 
> building . Full details are available at:
>https://ci.apache.org/builders/mod_perl_2_trunk/builds/57
>
>
> I've opened an INFRA ticket for this:
>
> https://issues.apache.org/jira/browse/INFRA-22631
>

Thanks!

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: docs-dev mailing list full of broken cron jobs

[Steve Hay]

On Sun, 28 Nov 2021 at 00:49, sebb  wrote:
>
> The docs-dev@perl mailing list has lots and lots of cron failure messages.
>
> Are the jobs really needed?
>
> It looks like at least 2 of the jobs are trying to access
> /www/perl.apache.org on minotaur.
> This directory is not present. AFAIK it was dropped several years ago.
>
> If no-one has access to minotaur to turn off (or fix?) the jobs,
> please raise an issue with JIRA.
>

I tried and failed to delete the cron jobs, and have now requested
assistance from INFRA instead:

https://issues.apache.org/jira/browse/INFRA-22630

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[RELEASE CANDIDATE] mod_perl-2.0.12 RC2

[Steve Hay]

Please download, test, and report back on this mod_perl 2.0.12 release
candidate.

https://dist.apache.org/repos/dist/dev/perl/mod_perl-2.0.12-rc2.tar.gz

SHA256:
mod_perl-2.0.12-rc2.tar.gz: 3203E37F 7A37732B DE5F171D 4322B4AF 525A447C
7AAA908E A07A9622 ADDC9FE0

SHA512:
mod_perl-2.0.12-rc2.tar.gz: 5AA7537F 7ED1425E 6D5D3C6D F0167E64 9C2B09F4
76A756F7 66446277 77D383D2 9D7ED383 603D9B29
BAC34D83 B064E5A8 BDF843B8 BE102404 316BF50E
2382FD47

Changes since RC1 are as follows:

Add bug tracker information to README, and add CONTRIBUTING.md. [Steve Hay]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] mod_perl-2.0.12 RC1

[Steve Hay]

On Mon, 8 Nov 2021 at 20:34, vincent.vey...@libremen.org
 wrote:
>
> On Sun, 7 Nov 2021 13:48:58 +0000
> Steve Hay  wrote:
>
> > Please download, test, and report back on this mod_perl 2.0.12 release
> > candidate.
> >
> > https://dist.apache.org/repos/dist/dev/perl/mod_perl-2.0.12-rc1.tar.gz
> >
>
> Hi Steve,
>
> Tested on Debian 10.10 (Buster)
>
> I had to do the following to run the test suite :
>
> Got the error
> /usr/bin/ld: cannot find -lperl
>
> Followed this advice
> https://stackoverflow.com/questions/335928/ld-cannot-find-an-existing-library
>
> and linked libperl.so to libperl.so.5.28.1 (the current library)
>
> Then got this error :
>
> ~..(13)Permission denied: AH00091: apache2: could not open error log file 
> /home/vincent/tmp/mod_perl-2.0.12-rc1/t/logs/error_log.
> AH00015: Unable to open logs
>
> So :
>
> /tmp/mod_perl-2.0.12-rc1/t# chgrp vincent logs
> ~/tmp/mod_perl-2.0.12-rc1/t# chmod g+w logs
>
> Tests runs fine, with this warning for just about all tests :
>
> t/protocol/echo_filter.t  [Mon Nov 08 21:03:49.933516 2021] 
> [core:warn] [pid 25091] AH00111: Config variable ${APACHE_RUN_DIR} is not 
> defined
> apache2: Syntax error on line 80 of /etc/apache2/httpd.conf: 
> DefaultRuntimeDir must be a valid directory, absolute or relative to 
> ServerRoot
>
>
> Here is the Test Summary Report
>
> ---
> t/apache/daemon.t (Wstat: 0 Tests: 2 Failed: 2)
>   Failed tests:  1-2
> t/filter/in_bbs_inject_header.t   (Wstat: 0 Tests: 36 Failed: 3)
>   Failed tests:  22, 26, 30
> Files=245, Tests=3505, 758 wallclock secs ( 2.95 usr  1.22 sys + 574.12 cusr 
> 163.00 csys = 741.29 CPU)
> Result: FAIL

Thanks for testing, and apologies for the slow response.

The t/filter/in_bbs_inject_header.t failure is documented in the
README. I don't recall seeing the t/apache/daemon.t failure before.
Does it happen with the current release version (2.0.11) as well?

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[RELEASE CANDIDATE] mod_perl-2.0.12 RC1

[Steve Hay]

Please download, test, and report back on this mod_perl 2.0.12 release
candidate.

https://dist.apache.org/repos/dist/dev/perl/mod_perl-2.0.12-rc1.tar.gz

SHA256:
mod_perl-2.0.12-rc1.tar.gz: 50CF6B06 6BAFF225 78F3BB77 7695F6CC 142D1F68
8E085067 A055EB2B B58425D3

SHA512:
mod_perl-2.0.12-rc1.tar.gz: 320DDC4C 896F68BE C50C54D5 03179D81 B6FFB889
C66BB568 FD5976F4 BD1224BE 617B2DFB E78D6DF0
AC8F9474 93730DD1 763A2A50 E594D407 B36336E4
69ABE24F

Major changes in this release are as follows:

Fix detection of APR's threading support on RHEL 8. [Petr Písař
]

Fix build for perl >= 5.33.7. [Leon Timmermans ]

Fix SIGSEGV crash due to wrong use of perl_parse(). [Charles Pigott
]

Improve installation instructions for AIX. [Rainer Tammer
]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] Apache-Test-1.43 RC1

[Steve Hay]

On Sat, 6 Nov 2021 at 01:24, Fred Moyer  wrote:
>
> On Fri, Nov 5, 2021 at 3:04 AM Steve Hay
>  wrote:
>
> > I will wait a few more days in case anyone is able to test on OSX, but if 
> > not then I will go ahead and release this as-is, i.e. with the !OSX since 
> > it makes sense to retain it if nobody can test on that platform.
>
> +1 (been a while since I've done that here, feels good!)
>

Thanks to everyone for testing/reviewing/voting. The release is now up
on CPAN :-)

I will make an rc1 for mod_perl shortly, and then return to the
Apache-Test issues that Ed raised above.

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[ANNOUNCE] Apache-Test-1.43

[Steve Hay]

We are pleased to announce the release of Apache-Test 1.43.

Apache::Test is a test toolkit for testing an Apache server with any
configuration. It works with Apache 1.3 and Apache 2.0/2.2/2.4 and any
of its modules, including mod_perl 1.0 and 2.0. It was originally
developed for testing mod_perl 2.0.

This release is now, or soon will be, available for download from your
favourite CPAN mirror or you can find it at:

https://metacpan.org/release/Apache-Test

Checksums for this release are:

MD5  = 1778e4b2ed4a083cbbca840780b67c72
SHA1 = 3169a9ac4c4e2d8bc3477ff495ecf965c1906556

Major changes in this release are as follows:

Our implementation of "which" should not search
for a program in the elements of PATH when program
already contains a path. [rjung]

Fix hash generation when the openssl command
was overwritten with the environment variable
"APACHE_TEST_OPENSSL_CMD". [rjung]

Add new config option limitrequestline needed
when the local directory paths are long. [rjung]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] Apache-Test-1.43 RC1

[Steve Hay]

On Fri, 5 Nov 2021 at 09:25, Rainer Jung  wrote:

> Am 05.11.2021 um 10:07 schrieb Steve Hay:
> > On Fri, 5 Nov 2021 at 01:54, Edward J. Sabol  > <mailto:edwardjsa...@gmail.com>> wrote:
> >
> > Anyway, these are pretty minor changes, but I don't think this
> > change is quite correct:
> >
> > +# No need to search PATH components
> > +# if $program already contains a path
> > +return $program if !OSX and !WINFU and
> > +$program =~ /\// and -f $program and -x $program;
> > +
> >
> > OSX uses slashes for path separators and supports "-f" and "-x". I
> > feel the "!OSX and " should be removed here, but I can't test it as
> > all of my Apache httpd systems are on Linux.
> >
> >  return $program if !WINFU and
> >  $program =~ /\// and -f $program and -x $program;
> >
> >
> > Unfortunately I can't test OSX either. Is anyone here able to?
> >
> > The change in question is this:
> >
> > https://svn.apache.org/viewvc?view=revision&revision=1880585
> > <https://svn.apache.org/viewvc?view=revision&revision=1880585>
> >
> > Rainer, did you include that "!OSX" for a specific reason? Do you know
> > if Ed is correct that it can be dropped as he suggests?
>
> I can't test OSX either and I think that was the only reason for the !OSX.
>
>

Ok, thanks.

I will wait a few more days in case anyone is able to test on OSX, but if
not then I will go ahead and release this as-is, i.e. with the !OSX since
it makes sense to retain it if nobody can test on that platform.

Re: [RELEASE CANDIDATE] Apache-Test-1.43 RC1

[Steve Hay]

On Fri, 5 Nov 2021 at 01:54, Edward J. Sabol  wrote:

> Anyway, these are pretty minor changes, but I don't think this change is
> quite correct:
>
> +# No need to search PATH components
> +# if $program already contains a path
> +return $program if !OSX and !WINFU and
> +$program =~ /\// and -f $program and -x $program;
> +
>
> OSX uses slashes for path separators and supports "-f" and "-x". I feel
> the "!OSX and " should be removed here, but I can't test it as all of my
> Apache httpd systems are on Linux.
>
> return $program if !WINFU and
> $program =~ /\// and -f $program and -x $program;
>
>

Unfortunately I can't test OSX either. Is anyone here able to?

The change in question is this:

https://svn.apache.org/viewvc?view=revision&revision=1880585

Rainer, did you include that "!OSX" for a specific reason? Do you know if
Ed is correct that it can be dropped as he suggests?

Re: [RELEASE CANDIDATE] Apache-Test-1.43 RC1

[Steve Hay]

On Wed, 3 Nov 2021 at 03:44, Edward J. Sabol  wrote:

> On Oct 26, 2021, at 3:28 AM, Steve Hay  wrote:
> > Do you get the same behaviour with Apache-Test-1.42, or are these new
> problems in 1.43-rc1?
>
> Yes, I get the same behavior with 1.42. At a minimum, I need to
>
> ln -s /path/to/my/apache/modules
>
> inside the Apache-Test-1.43-rc1 directory and create a t/conf/
> httpd.conf.in file (as suggested by the README) with a bunch of
> LoadModule commands in it to get the httpd server to start up in any kind
> of useful state.
>
> The resulting "make test" output still skips the tests of mod_cgi,
> mod_perl, etc., even though I explicitly load them in t/conf/httpd.conf.in.
> I think I've tracked down the reason for this to Apache::TestConfig's
> modules hash doesn't include all of the modules that are actually loaded. I
> haven't been able to figure out why though. I suspect it's parsing the
> wrong httpd.conf file. If I hack Apache::Test's need_module to always
> return 1, all tests pass, so these modules are loaded. This seems to be a
> long-standing problem for some (but not most) people, as evidenced by this
> thread from 2006:
>
> https://www.mail-archive.com/test-dev@perl.apache.org/msg00321.html
>
> > If the problems are not new then I'm inclined to leave them for now
> (assuming I can get some +1s from voters on releasing this) and look at
> addressing them for the next release. The purpose of this release was
> really just to update Apache-Test in mod_perl to the current version before
> making a long-overdue release of that.
>
> Do you have a link that shows the changes from 1.42 to 1.43-rc1? Or just
> send me the diffs? That's probably a better way for me to evaluate the
> changes.
>
>
Let's come back to the existing problems you've raised for the next
release, otherwise this will hold up the mod_perl release.

I've attached a diff of the changes from 1.42 to 1.43-rc1.
diff -ruN Apache-Test-1.42/Changes Apache-Test-1.43-rc1/Changes
--- Apache-Test-1.42/Changes2019-08-28 08:12:07.0 +0100
+++ Apache-Test-1.43-rc1/Changes2021-10-15 09:13:30.0 +0100
@@ -6,6 +6,19 @@
 
 =over 3
 
+=item 1.43-rc1
+
+Our implementation of "which" should not search
+for a program in the elements of PATH when program
+already contains a path. [rjung]
+
+Fix hash generation when the openssl command
+was overwritten with the environment variable
+"APACHE_TEST_OPENSSL_CMD". [rjung]
+
+Add new config option limitrequestline needed
+when the local directory paths are long. [rjung]
+
 =item 1.42 Aug 28 2019
 
 Fix loading apache_test_config.pm for recent perls in which '.' is
diff -ruN Apache-Test-1.42/META.json Apache-Test-1.43-rc1/META.json
--- Apache-Test-1.42/META.json  2019-08-28 08:12:08.0 +0100
+++ Apache-Test-1.43-rc1/META.json  2021-10-15 09:13:30.0 +0100
@@ -39,6 +39,6 @@
   }
},
"release_status" : "stable",
-   "version" : "1.42",
+   "version" : "1.43-rc1",
"x_serialization_backend" : "JSON::PP version 4.02"
 }
diff -ruN Apache-Test-1.42/META.yml Apache-Test-1.43-rc1/META.yml
--- Apache-Test-1.42/META.yml   2019-08-28 08:12:08.0 +0100
+++ Apache-Test-1.43-rc1/META.yml   2021-10-15 09:13:30.0 +0100
@@ -21,5 +21,5 @@
   Cwd: '2.06'
   File::Spec: '0.8'
   Win32::Process: '0'
-version: '1.42'
+version: 1.43-rc1
 x_serialization_backend: 'CPAN::Meta::YAML version 0.018'
diff -ruN Apache-Test-1.42/lib/Apache/Test.pm 
Apache-Test-1.43-rc1/lib/Apache/Test.pm
--- Apache-Test-1.42/lib/Apache/Test.pm 2019-08-28 08:12:08.0 +0100
+++ Apache-Test-1.43-rc1/lib/Apache/Test.pm 2021-10-15 09:13:30.0 
+0100
@@ -37,7 +37,7 @@
 
 use vars qw(@ISA @EXPORT %EXPORT_TAGS $VERSION %SubTests @SkipReasons);
 
-$VERSION = '1.42';
+$VERSION = '1.43';
 
 my @need = qw(need_lwp need_http11 need_cgi need_access need_auth
   need_module need_apache need_min_apache_version 
need_min_apache_fix
diff -ruN Apache-Test-1.42/lib/Apache/TestCommon.pm 
Apache-Test-1.43-rc1/lib/Apache/TestCommon.pm
--- Apache-Test-1.42/lib/Apache/TestCommon.pm   2019-08-28 08:12:08.0 
+0100
+++ Apache-Test-1.43-rc1/lib/Apache/TestCommon.pm   2021-10-15 
09:13:30.0 +0100
@@ -53,7 +53,7 @@
 $received += length $chunk;
 });
 
-ok t_cmp($length, $received, 'bytes in body');
+ok t_cmp($received, $length, 'bytes in body');
 }
 }
 }
diff -ruN Apache-Test-1.42/lib/Apache/TestConfig.pm 
Apache-Test-1.43-rc1/lib/Apache/TestConfig.pm
--- Apache-Test-1.42/lib/Apache/TestConfig.pm   

Re: [RELEASE CANDIDATE] Apache-Test-1.43 RC1

[Steve Hay]

On Tue, 26 Oct 2021 at 08:28, Steve Hay  wrote:

> On Fri, 15 Oct 2021 at 21:33, Edward J. Sabol 
> wrote:
>
>> On Oct 15, 2021, at 4:26 AM, Steve Hay  wrote:
>> > As a preliminary to making a new mod_perl release, I am updating
>> Apache-Test.
>> >
>> > Please download, test, and report back on this Apache-Test 1.43 release
>> candidate.
>> >
>> > https://dist.apache.org/repos/dist/dev/perl/Apache-Test-1.43-rc1.tar.gz
>>
>> INSTALL says to "perl Makefile.PL -httpd /path/to/your/apache/bin/httpd",
>> but I found I had to
>>
>> perl Makefile.PL -httpd /path/to/your/apache/bin/httpd -apxs
>> /path/to/your/apache/bin/apxs
>>
>> or else "make test" reported a lot of "APXS (/usr/sbin/apxs) query for
>> SBINDIR failed" errors.
>>
>> Obviously, /usr/sbin/apxs != /path/to/your/apache/bin/apxs for my httpd
>> installation.
>>
>> Well, after I "perl Makefile.PL -httpd /path/to/your/apache/bin/httpd
>> -apxs /path/to/your/apache/bin/apxs", I encountered the following:
>>
>> waiting 60 seconds for server to start: .AH00526: Syntax error on line 12
>> of /path/to/tests/Apache-Test-1.43-rc1/t/conf/httpd.conf:
>> Invalid command 'User', perhaps misspelled or defined by a module not
>> included in the server configuration
>> [  error]
>> server has died with status 255 (t/logs/error_log wasn't created, start
>> the server in the debug mode)
>> sh: line 1: 39154 Terminated  /path/to/perl
>> /path/to/tests/Apache-Test-1.43-rc1/t/TEST -bugreport -verbose=0
>>
>> The first non-comment lines in t/conf/httpd.conf are
>>
>> User""
>> Group""
>>
>> I'm using Apache 2.4.41. If I try to edit t/conf/httpd.conf to comment
>> out these lines or to add "LoadModule unixd_module modules/mod_unixd.so" so
>> that the httpd.conf will work, my changes just get destroyed the next time
>> I "make test". Frustrating.
>>
>> Also, It would be nice if Apache::Test would skip trying to "ulimit -c
>> unlimited" if $LOGNAME != "root" in order to avoid the following message:
>>
>> sh: line 0: ulimit: core file size: cannot modify limit: Operation not
>> permitted
>>
>> If only root should "make test", then the INSTALL instructions should say
>> to use sudo or something.
>>
>>
> Thanks for testing, and apologies for not getting back to you sooner.
>
> Unfortunately, I don't know anything about building Apache/mod_perl on
> *nix systems, and was hoping that someone else might step in with some
> answers here.
>
> Do you get the same behaviour with Apache-Test-1.42, or are these new
> problems in 1.43-rc1?
>
> If the problems are not new then I'm inclined to leave them for now
> (assuming I can get some +1s from voters on releasing this) and look at
> addressing them for the next release. The purpose of this release was
> really just to update Apache-Test in mod_perl to the current version before
> making a long-overdue release of that.
>
>
+1 from me, using VS2019 (v16.9.4) x64 on Windows 10, Perl 5.34.0, Apache
httpd 2.4.51, mod_perl from SVN (r1894283), libapreq2-2.16, with and
without LWP installed.

(For me the build works just doing "perl Makefile.PL && nmake &&
nmake test" after setting my perl.exe and httpd.exe folders in the %PATH%.)

Still looking for more +1s...

Re: [RELEASE CANDIDATE] Apache-Test-1.43 RC1

[Steve Hay]

On Fri, 15 Oct 2021 at 21:33, Edward J. Sabol 
wrote:

> On Oct 15, 2021, at 4:26 AM, Steve Hay  wrote:
> > As a preliminary to making a new mod_perl release, I am updating
> Apache-Test.
> >
> > Please download, test, and report back on this Apache-Test 1.43 release
> candidate.
> >
> > https://dist.apache.org/repos/dist/dev/perl/Apache-Test-1.43-rc1.tar.gz
>
> INSTALL says to "perl Makefile.PL -httpd /path/to/your/apache/bin/httpd",
> but I found I had to
>
> perl Makefile.PL -httpd /path/to/your/apache/bin/httpd -apxs
> /path/to/your/apache/bin/apxs
>
> or else "make test" reported a lot of "APXS (/usr/sbin/apxs) query for
> SBINDIR failed" errors.
>
> Obviously, /usr/sbin/apxs != /path/to/your/apache/bin/apxs for my httpd
> installation.
>
> Well, after I "perl Makefile.PL -httpd /path/to/your/apache/bin/httpd
> -apxs /path/to/your/apache/bin/apxs", I encountered the following:
>
> waiting 60 seconds for server to start: .AH00526: Syntax error on line 12
> of /path/to/tests/Apache-Test-1.43-rc1/t/conf/httpd.conf:
> Invalid command 'User', perhaps misspelled or defined by a module not
> included in the server configuration
> [  error]
> server has died with status 255 (t/logs/error_log wasn't created, start
> the server in the debug mode)
> sh: line 1: 39154 Terminated  /path/to/perl
> /path/to/tests/Apache-Test-1.43-rc1/t/TEST -bugreport -verbose=0
>
> The first non-comment lines in t/conf/httpd.conf are
>
> User""
> Group""
>
> I'm using Apache 2.4.41. If I try to edit t/conf/httpd.conf to comment out
> these lines or to add "LoadModule unixd_module modules/mod_unixd.so" so
> that the httpd.conf will work, my changes just get destroyed the next time
> I "make test". Frustrating.
>
> Also, It would be nice if Apache::Test would skip trying to "ulimit -c
> unlimited" if $LOGNAME != "root" in order to avoid the following message:
>
> sh: line 0: ulimit: core file size: cannot modify limit: Operation not
> permitted
>
> If only root should "make test", then the INSTALL instructions should say
> to use sudo or something.
>
>
Thanks for testing, and apologies for not getting back to you sooner.

Unfortunately, I don't know anything about building Apache/mod_perl on *nix
systems, and was hoping that someone else might step in with some answers
here.

Do you get the same behaviour with Apache-Test-1.42, or are these new
problems in 1.43-rc1?

If the problems are not new then I'm inclined to leave them for now
(assuming I can get some +1s from voters on releasing this) and look at
addressing them for the next release. The purpose of this release was
really just to update Apache-Test in mod_perl to the current version before
making a long-overdue release of that.

[RELEASE CANDIDATE] Apache-Test-1.43 RC1

[Steve Hay]

As a preliminary to making a new mod_perl release, I am updating
Apache-Test.

Please download, test, and report back on this Apache-Test 1.43 release
candidate.

https://dist.apache.org/repos/dist/dev/perl/Apache-Test-1.43-rc1.tar.gz

MD5  = 44d976ae19d5ab2879e43a26a360366b
SHA1 = 151f61214d6e04e671304f34312f3b3ab30b40f2

Major changes in this release are as follows:

Our implementation of "which" should not search
for a program in the elements of PATH when program
already contains a path. [rjung]

Fix hash generation when the openssl command
was overwritten with the environment variable
"APACHE_TEST_OPENSSL_CMD". [rjung]

Add new config option limitrequestline needed
when the local directory paths are long. [rjung]

Re: [RESULT: PASS] Re: [VOTE] Release libapreq2-2.16

[Steve Hay]

On Fri, 19 Mar 2021 at 14:37, Steve Hay  wrote:
>
> On Fri, 19 Mar 2021 at 14:33, Joe Orton  wrote:
>>
>> Thanks for testing again, the vote has passed:
>>
>> PMC votes: +1 from rpluem, ylavic, jorton
>> Community: +1 stevehay
>>
>> I'll promote the release & send the announcement mail.
>
>
> Thanks for this, and sorry for the hassle in wanting another release so soon! 
> I'll get the new version uploaded to CPAN shortly, and hopefully it will get 
> indexed correctly this time.

Success! We now have libapreq2-2.16 correctly indexed on (Meta)CPAN:

https://metacpan.org/release/SHAY/libapreq2-2.16

Thanks to Joe and all the testers/voters for getting this update done.

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RESULT: PASS] Re: [VOTE] Release libapreq2-2.16

[Steve Hay]

On Fri, 19 Mar 2021 at 14:33, Joe Orton  wrote:

> Thanks for testing again, the vote has passed:
>
> PMC votes: +1 from rpluem, ylavic, jorton
> Community: +1 stevehay
>
> I'll promote the release & send the announcement mail.
>

Thanks for this, and sorry for the hassle in wanting another release so
soon! I'll get the new version uploaded to CPAN shortly, and hopefully it
will get indexed correctly this time.

Re: [DISCUSS] The future of mod_perl

[Steve Hay]

On Wed, 17 Mar 2021 at 19:40, Sander Striker  wrote:
>
> Dear community members,
>
> As projects mature, they will naturally reach a point where activity reduces 
> to a level such that the project is no longer sustainable.  At Apache, 
> projects reach this stage when there are not at least 3 active PMC members 
> providing oversight. Projects that reach this stage are usually placed in the 
> Attic [1] or absorbed by another Apache project willing to manage its 
> releases.
>
> If you are interested in seeing mod_perl remain an active project, and are 
> able to help maintain and provide oversight, please respond in this thread 
> indicating that you are interested in performing the duties of a PMC 
> member[2].
>

I confirm again that I am still here and willing to participate and
sustain oversight of mod_perl as required.

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Does anyone know what MP_NEED_HASH_SEED_FIXUP is for?

[Steve Hay]

The reason I ask is that it is broken as of perl 5.33.7 and it looks
like it will remain that way. See:
https://github.com/Perl/perl5/issues/18617

The problem is that the assignment to PL_hash_seed done in
modperl_hash_seed_set() is no longer possible, and there is a
suggestion that it was terribly wrong of mod_perl to be doing this
anyway so the bug is really in mod_perl, not the perl change that
causes the build breakage.

If nobody knows any better than I'm inclined to apply the following
diff to simply switch off MP_NEED_HASH_SEED_FIXUP for perl 5.33.7 and
higher:

Index: src/modules/perl/modperl_perl.c
===
--- src/modules/perl/modperl_perl.c (revision 1887671)
+++ src/modules/perl/modperl_perl.c (working copy)
@@ -210,7 +210,10 @@
  }
 }

-#if !(MP_PERL_VERSION_AT_MOST(5, 8, 0)) && \
+/* The need for MP_NEED_HASH_SEED_FIXUP is unclear, and it no longer works as
+ * of Perl 5.33.7 anyway: See https://github.com/Perl/perl5/issues/18617 */
+#if (!(MP_PERL_VERSION_AT_MOST(5, 8, 0)) && \
+ MP_PERL_VERSION_AT_MOST(5, 33, 6)) && \
 (defined(USE_HASH_SEED) || defined(USE_HASH_SEED_EXPLICIT))
 #define MP_NEED_HASH_SEED_FIXUP
 #endif

Using Apache 2.4.46 and mod_perl 2.0.11 on Windows 10 with VS2019 x64
I get the following test failures from a build with perl 5.33.6:

t\filter\in_bbs_inject_header.t   (Wstat: 0 Tests: 36 Failed: 3)
  Failed tests:  22, 26, 30
t\modperl\env.t   (Wstat: 0 Tests: 164 Failed: 17)
  Failed tests:  25, 28, 31, 43, 46, 49, 55, 58, 73, 82
85, 88, 91, 97, 142, 145, 163
t\modperl\setupenv.t  (Wstat: 0 Tests: 63 Failed: 6)
  Failed tests:  8, 22, 29, 36, 50, 57
t\preconnection\note.t(Wstat: 0 Tests: 1 Failed: 1)
  Failed test:  1

As of 5.33.7, the build now fails:

modperl_perl.c(271): error C2102: '&' requires l-value

But with the above patch applied the build works again, and I actually
have fewer test failures!:

t\modperl\env.t   (Wstat: 0 Tests: 158 Failed: 17)
  Failed tests:  25, 28, 31, 43, 46, 49, 55, 58, 67, 76
79, 82, 85, 91, 136, 139, 157
t\preconnection\note.t(Wstat: 0 Tests: 1 Failed: 1)
  Failed test:  1

Please can people test this patch on other setups and report whether
it causes any breakage.

And if anyone can remember what this hash seed fix-up was intended to
do then it would be interesting to hear. I worry that something that I
haven't tested may be requiring it.

The Changes file has this to say about it:

"perl 5.8.1 randomizes the hash seed, because we precalculate the hash
values of mgv elements the hash seed has to be the same across all
perl interpreters. So mod_perl now intercepts cases where perl would
have randomize it, do the seed randomization by itself and tell perl
to use that value."

That change is rev. 70047 in SVN (30 July 2003), but the commit
comment is just the same as the Changes comment above. No tests seem
to have been added, and I don't have an example of anything that
requires this fix-up in order to work.

Steve

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [VOTE] Release libapreq2-2.16

[Steve Hay]

On Wed, 10 Mar 2021 at 16:07, Joe Orton  wrote:
>
> On Wed, Mar 10, 2021 at 04:05:16PM +, Joe Orton wrote:
> > Hi, I've prepared a candidate release tarball for libapreq2 v2.15 here:
>
> You can see I'm still an amateur at this - should read ^^ v2.16
>

Many thanks for rolling out this release.

Community vote: +1

Tested on Windows 10 with:
- VS2013 Update 5 x86, httpd 2.2.34, perl 5.32.1, mod_perl 2.0.11
- VS2019 v16.7.3 x64, httpd 2.4.46, perl 5.33.6, mod_perl 2.0.11

One failure for both:

t\apreq\cgi.t  (Wstat: 2304 Tests: 31 Failed: 0)
  Non-zero exit status: 9
  Parse errors: Bad plan.  You planned 71 tests but ran 31.

which is the same "Odd number of elements in hash assignment at
t\apreq\cgi.t line 197." failure as I saw in 2.15.

Diffing 2.15 with 2.16 looks good, in particular with the expected
differences in META.yml, which I hope will fix the MetaCPAN indexing
problem.

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

New Windows mod_perl2/libapreq2 build available

[Steve Hay]

For those who are interested, I have uploaded a 64-bit build of
mod_perl-2.0.11 and libapreq2-2.15 components compatible with Apache
Lounge 2.4.46 (Win64, VC16, built 18 February 2021) and Strawberry
Perl 5.32.1.1 (64bit).

You can download it from: https://people.apache.org/~stevehay/

The SHA1 digest is: 33e231eb91901007e25fd78eb5643901672c519b

Regards,
Steve

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[ANNOUNCE] mod_perl-2.0.11

[Steve Hay]

We are pleased to announce the release of mod_perl 2.0.11.

mod_perl is an Apache HTTP Server module for embedding a Perl
interpreter in your web server, giving you super-fast dynamic content
by avoiding the overhead of starting an external interpreter.

This release is now, or soon will be, available for download from a
mirror site near you via:

http://perl.apache.org/download/index.html

or in the meantime directly from:

http://apache.org/dist/perl/
https://metacpan.org/release/SHAY/mod_perl-2.0.11

Checksums for this release are:

MD5  = 897708961bcc079b53c91860f95a2b59
SHA1 = 7606ab4862605a72db4b51d2e498588305b45719

Major changes in this release are as follows:

Fix t/modules/apache_resource.t failures [Steve Hay]

Fix [CVE-2011-2767] Arbitrary Perl code execution in the context of
the user account via a user-owned .htaccess. Patch from
bugs.debian.org #644169. [Jan Ingvoldstad
]

Fix potential test suite hangs due to pipelined response deadlocks.
Patch from rt.cpan.org #82409. [Zefram ]

Fix t/compat/request.t failures [Steve Hay]

Fix use-after-free segfault in ap_server_config_defines seen on
start-up on OpenBSD. [Found/fixed by Sam Vaughan/Joe Orton]

Fix build with Perls earlier than 5.13.6. [Rainer Jung
]

Fix filter/in_bbs_inject_header.t test failure with Apache 2.4.25+.
[Stefan Fritsch ]

Fix apache/read.t test failure with Apache 2.4.25+. [Niko Tyni
]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] mod_perl-2.0.11 RC3

[Steve Hay]

On Thu, 3 Oct 2019 at 10:14, Pete Houston  wrote:
>
> On Wed, Oct 02, 2019 at 06:07:39PM +0100, Steve Hay wrote:
> >
> > I'm not sure about this area, but if it works for with your patch then
> > it's probably an improvement :-)
> >
> > Just out of interest, does the following alternative patch work?
> >
> > Index: lib/Apache2/Resource.pm
> > ===
> > --- lib/Apache2/Resource.pm (revision 1866274)
> > +++ lib/Apache2/Resource.pm (working copy)
> > @@ -115,8 +115,7 @@
> >"");
> >
> >  for my $res (keys %$lim) {
> > -my $val = eval "&BSD::Resource::${res}()";
> > -my ($soft, $hard) = getrlimit $val;
> > +my ($soft, $hard) = getrlimit($lim->{$res});
> >  (my $limit = $res) =~ s/^RLIMIT_//;
> >  ($soft, $hard) = ("$soft " . BM($soft), "$hard ". BM($hard))
> >  if $is_mb{$limit};
>
> Yes, that works just as well for me.
>
> > This avoids making the &BSD::Resource::${res}() calls, which are what
> > is returning undef for you in some cases. Or does this fail just the
> > same because some of the values of %$lim are also undef anyway?
>
> The %$lim values are not undef. See the attached stand-alone test and
> output for the problematic values which I see and which also shows that
> the limits for those resources are still retrievable via getrlimit().
>

Great! Thanks for confirming. I've now committed that fix, and hope to
make the final release soon: I don't think this single change warrants
an RC4.

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] mod_perl-2.0.11 RC3

[Steve Hay]

On Wed, 2 Oct 2019 at 11:03, Pete Houston  wrote:
>
> On Wed, Oct 02, 2019 at 08:13:59AM +0100, Steve Hay wrote:
> >
> > I think the last line of the log extracts above is the clue, but I'm
> > not going to be able to fix this myself. Line 119 in
> > Apache2/Resource.pm is:
> >
> > my ($soft, $hard) = getrlimit $val;
> >
> > So it looks like $val is blank. That comes from line 118:
> >
> > my $val = eval "&BSD::Resource::${res}()";
> >
> > where $res is one of the keys of the $lim hashref, which comes from line 
> > 112:
> >
> > my $lim = get_rlimits();
> >
> > Both getrlimit() and get_rlimits() are further BSD::Resource
> > functions, which I see the error log also has further references to
> > above (several uninitialized value warnings), but is not a thing that
> > I have access to on Windows. Unless anyone else can help here then
> > we'll have to let this one go for now. Hopefully it's not a
> > showstopper.
>
> Thanks for this diagnosis. Going through the resources here I do see
> some which were undef and thus causing getrlimit to complain. Attached
> is the trivial patch which now makes the test pass for me. It seemed
> reasonable to simply omit undef resources but you might have a contrary
> view.
>
> For the record the undef resources I see are:
>
> RLIMIT_NICE
> RLIMIT_SIGPENDING
> RLIMIT_RTTIME
> RLIMIT_RTPRIO
> RLIMIT_MSGQUEUE
>

I'm not sure about this area, but if it works for with your patch then
it's probably an improvement :-)

Just out of interest, does the following alternative patch work?

Index: lib/Apache2/Resource.pm
===
--- lib/Apache2/Resource.pm (revision 1866274)
+++ lib/Apache2/Resource.pm (working copy)
@@ -115,8 +115,7 @@
   "");

 for my $res (keys %$lim) {
-my $val = eval "&BSD::Resource::${res}()";
-my ($soft, $hard) = getrlimit $val;
+my ($soft, $hard) = getrlimit($lim->{$res});
 (my $limit = $res) =~ s/^RLIMIT_//;
 ($soft, $hard) = ("$soft " . BM($soft), "$hard ". BM($hard))
 if $is_mb{$limit};

This avoids making the &BSD::Resource::${res}() calls, which are what
is returning undef for you in some cases. Or does this fail just the
same because some of the values of %$lim are also undef anyway?


Resource-v2.patch
Description: Binary data

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] mod_perl-2.0.11 RC3

[Steve Hay]

On Fri, 27 Sep 2019 at 11:32, Pete Houston  wrote:
>
> On Fri, Sep 27, 2019 at 08:22:11AM +0100, Steve Hay wrote:
> > Can you give any more detail on the apache_resource.t failure? I've
> > not seen that reported anywhere else.
>
> $ t/TEST -verbose t/modules/apache_resource.t
> [warning] setting ulimit to allow core files
> ulimit -c unlimited; /usr/bin/perl /home/pete/src/mod_perl-2.0.11-rc3/t/TEST 
> -verbose 't/modules/apache_resource.t'
> /usr/local/httpd-2.4.41/bin/httpd  -d /home/pete/src/mod_perl-2.0.11-rc3/t -f 
> /home/pete/src/mod_perl-2.0.11-rc3/t/conf/httpd.conf -D APACHE2 -D APACHE2_4 
> -D PERL_USEITHREADS
> using Apache/2.4.41 (event MPM)
>
> waiting 300 seconds for server to start: .[Fri Sep 27 11:21:12.994001 2019] 
> [env:warn] [pid 4182:tid 139866432870272] AH01506: PassEnv variable 
> LD_LIBRARY_PATH was undefined
> [Fri Sep 27 11:21:13.154432 2019] [perl:info] [pid 4182:tid 139866432870272] 
> 6 Apache2:: modules loaded
> [Fri Sep 27 11:21:13.154642 2019] [perl:info] [pid 4182:tid 139866432870272] 
> 0 APR:: modules loaded
> [Fri Sep 27 11:21:13.154837 2019] [perl:info] [pid 4182:tid 139866432870272] 
> base server + 29 vhosts ready to run tests
> ...
> waiting 300 seconds for server to start: ok (waited 3 secs)
> server localhost.localdomain:8529 started
> server localhost.localdomain:8530 listening (filter_out_apache)
> server localhost.localdomain:8531 listening (perlsections)
> server localhost.localdomain:8532 listening (inherit)
> server localhost.localdomain:8533 listening (TestUser::rewrite)
> server localhost.localdomain:8534 listening (TestVhost::log)
> server localhost.localdomain:8535 listening (TestVhost::config)
> server localhost.localdomain:8536 listening (TestModperl::perl_options2)
> server localhost.localdomain:8537 listening (TestModperl::perl_options)
> server localhost.localdomain:8538 listening (TestModperl::merge)
> server localhost.localdomain:8539 listening (TestModperl::setupenv)
> server localhost.localdomain:8540 listening 
> (TestDirective::perlcleanuphandler)
> server localhost.localdomain:8541 listening (TestModules::proxy)
> server localhost.localdomain:8542 listening (TestProtocol::echo_block)
> server localhost.localdomain:8543 listening (TestProtocol::echo_bbs2)
> server localhost.localdomain:8544 listening (TestProtocol::pseudo_http)
> server localhost.localdomain:8545 listening (TestProtocol::echo_timeout)
> server localhost.localdomain:8546 listening (TestProtocol::echo_filter)
> server localhost.localdomain:8547 listening (TestProtocol::echo_nonblock)
> server localhost.localdomain:8548 listening (TestProtocol::echo_bbs)
> server localhost.localdomain:8549 listening (TestPreConnection::note)
> server localhost.localdomain:8550 listening (TestHooks::init)
> server localhost.localdomain:8551 listening (TestHooks::trans)
> server localhost.localdomain:8552 listening (TestHooks::hookrun)
> server localhost.localdomain:8553 listening (TestHooks::stacked_handlers2)
> server localhost.localdomain:8554 listening (TestHooks::startup)
> server localhost.localdomain:8555 listening (TestFilter::both_str_con_add)
> server localhost.localdomain:8556 listening (TestFilter::in_bbs_msg)
> server localhost.localdomain:8557 listening (TestFilter::in_str_msg)
> server localhost.localdomain:8558 listening (TestFilter::in_bbs_inject_header)
> server localhost.localdomain:8559 listening (TestDirective::perlmodule)
> server localhost.localdomain:8560 listening (TestDirective::perlrequire)
> server localhost.localdomain:8561 listening (TestDirective::perlloadmodule4)
> server localhost.localdomain:8562 listening (TestDirective::perlloadmodule5)
> server localhost.localdomain:8563 listening (TestDirective::perlloadmodule3)
> server localhost.localdomain:8564 listening (TestAPI::add_config)
> server localhost.localdomain:8565 listening (TestDirective::perlloadmodule6)
> server localhost.localdomain:8566 listening (TestHooks::push_handlers_anon)
> t/modules/apache_resource.t ..
> 1..1
> # Running under perl version 5.020003 for linux
> # Current time local: Fri Sep 27 11:21:16 2019
> # Current time GMT:   Fri Sep 27 10:21:16 2019
> # Using Test.pm version 1.26
> # Using Apache/Test.pm version 1.42
> not ok 1
> # Failed test 1 in t/modules/apache_resource.t at line 18
> Failed 1/1 subtests
>
> Test Summary Report
> ---
> t/modules/apache_resource.t (Wstat: 0 Tests: 1 Failed: 1)
>   Failed test:  1
> Files=1, Tests=1,  1 wallclock secs ( 0.05 usr  0.02 sys +  0.47 cusr  0.11 
> csys =  0.65 CPU)
> Result: FAIL
> Failed 1/1 test programs. 1/1 subtests failed.
> [warning] server localhost.localdomain:8529 shutdown
> [  error] error running tests (please examine t/logs/error_log)
>

Re: [RELEASE CANDIDATE] mod_perl-2.0.11 RC3

[Steve Hay]

On Thu, 26 Sep 2019 at 23:05, Pete Houston  wrote:
>
> This RC fails for me:
>
> Test Summary Report
> ---
> t/filter/in_bbs_inject_header.t   (Wstat: 0 Tests: 36 Failed: 3)
>   Failed tests:  22, 26, 30
> t/modules/apache_resource.t   (Wstat: 0 Tests: 1 Failed: 1)
>   Failed test:  1
> Files=245, Tests=2607, 138 wallclock secs ( 2.26 usr  0.92 sys + 100.92 cusr 
> 19.69 csys = 123.79 CPU)
> Result: FAIL
> Failed 2/245 test programs. 4/2607 subtests failed.
>
> with apache 2.4.41 (event MPM), perl 5.20.3 on x86_64-linux-thread-multi.
> I did not test any of the previous RCs.
>

I've just updated the README with a note that the recurring
in_bbs_inject_header.t failure is a known problem.

Can you give any more detail on the apache_resource.t failure? I've
not seen that reported anywhere else.

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] mod_perl-2.0.11 RC2

[Steve Hay]

On Thu, 26 Sep 2019 at 16:17, Damyan Ivanov  wrote:
>
> -=| Steve Hay, 26.09.2019 09:02:19 +0100 |=-
> > Before I actually roll out an RC3, please can you confirm that
> > switching to FINFO_SIZE as per this patch works for you (it does for
> > me):
> > …
>
> This works for me:
>
> # testing : $r->finfo
> # expected: 7278
> # received: 7278
> ok 44
>
> (the rest of the tests still pass).
>

Thanks for the confirmation. RC3 is now uploaded.

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[RELEASE CANDIDATE] mod_perl-2.0.11 RC3

[Steve Hay]

Please download, test, and report back on this mod_perl 2.0.11 release
candidate.

https://dist.apache.org/repos/dist/dev/perl/mod_perl-2.0.11-rc3.tar.gz

MD5  = dd6f7918fc492be21868a9774b2e58c5
SHA1 = e4af2eaabb40962cd7e5566e5ee3342090413b30

Changes since RC2 are as follows:

Fix t/api/request_rec.t failures [Steve Hay]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] mod_perl-2.0.11 RC2

[Steve Hay]

On Thu, 26 Sep 2019 at 06:54, Damyan Ivanov  wrote:
>
> -=| Damyan Ivanov, 25.09.2019 23:06:56 +0300 |=-
> > -=| Steve Hay, 25.09.2019 08:34:07 +0100 |=-
> > > What happens if you revert the change in that file? I.e. Change
> > > FINFO_NAME back to FINFO_NORM on line 168, and on line 18 (the
> > > APR::Const -compile line)? Does that fix it for you?
> >
> > With these two lines reverted, all tests pass for me.
>
> Out of interest I also tried using "FINFO_NAME | FINFO_NORM" on line
> 168 and the test failed with the same error. It appears that
> FINFO_NAME triggers it.
>
> Tracing this in APR, the error seems to come from
> file_io/unix/filestat.c¹ where the valid flags on line 73 don't
> include APR_FINFO_NAME. This kind of makes sense, because the stat(2)
> struct lacks information about the file name.
>
>  ¹ https://sources.debian.org/src/apr/1.6.5-1/file_io/unix/filestat.c/
>
> -- Damyan

Good catch! That explains why using FINFO_NAME is failing for you but
working for me (on Windows): apr/file_io/win32/filestat.c *does* have
support for FINFO_NAME in it, filling in finfo->name from filename if
it succeeded in getting filename (using calls other than stat()!).

The point of the request_rec test in question is only to test a single
field anyway (the rest are tested in TestAPR::finfo, as the comment
says) so I will simply change it to test a different field. Evidently
FINFO_NAME was a poor choice!

Before I actually roll out an RC3, please can you confirm that
switching to FINFO_SIZE as per this patch works for you (it does for
me):

Index: t/response/TestAPI/request_rec.pm
===
--- t/response/TestAPI/request_rec.pm (revision 1866274)
+++ t/response/TestAPI/request_rec.pm (working copy)
@@ -15,7 +15,7 @@
 use APR::Pool ();

 use Apache2::Const -compile => qw(OK M_GET M_PUT);
-use APR::Const-compile => qw(FINFO_NAME);
+use APR::Const-compile => qw(FINFO_SIZE);

 #this test module is only for testing fields in the request_rec
 #listed in apache_structures.map
@@ -165,12 +165,13 @@

 # finfo
 {
-my $finfo = APR::Finfo::stat(__FILE__,
APR::Const::FINFO_NAME, $r->pool);
+my $size = (stat __FILE__)[7];
+my $finfo = APR::Finfo::stat(__FILE__,
APR::Const::FINFO_SIZE, $r->pool);
 $r->finfo($finfo);
 # just one field test, all accessors are fully tested in
 # TestAPR::finfo
-ok t_cmp($r->finfo->fname,
- __FILE__,
+ok t_cmp($r->finfo->size,
+ $size,
  '$r->finfo');
 }


request_rec.patch
Description: Binary data

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] mod_perl-2.0.11 RC2

[Steve Hay]

On Wed, 25 Sep 2019 at 03:59, Adam Prime  wrote:
>
> Still seeing the t/api/request_rec.t failures, which do not occur with
> the same machine and same settings with 2.0.10. It's failing after the
> mtime test, before the finfo test.
>
> Test Summary Report
> ---
> t/api/request_rec.t   (Wstat: 0 Tests: 43 Failed: 0)
>Parse errors: Bad plan.  You planned 55 tests but ran 43.
> t/filter/in_bbs_inject_header.t   (Wstat: 0 Tests: 36 Failed: 3)
>Failed tests:  22, 26, 30
> Files=245, Tests=3402, 98 wallclock secs ( 0.71 usr  0.30 sys + 67.05
> cusr 12.05 csys = 80.11 CPU)
> Result: FAIL
> Failed 2/245 test programs. 3/3402 subtests failed.
>
>

(Btw, the filter/in_bbs_inject_header.t failures are an existing
problem - see https://rt.cpan.org/Ticket/Display.html?id=118919)

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] mod_perl-2.0.11 RC2

[Steve Hay]

On Wed, 25 Sep 2019 at 03:59, Adam Prime  wrote:
>
> Still seeing the t/api/request_rec.t failures, which do not occur with
> the same machine and same settings with 2.0.10. It's failing after the
> mtime test, before the finfo test.
>
> Test Summary Report
> ---
> t/api/request_rec.t   (Wstat: 0 Tests: 43 Failed: 0)
>Parse errors: Bad plan.  You planned 55 tests but ran 43.
> t/filter/in_bbs_inject_header.t   (Wstat: 0 Tests: 36 Failed: 3)
>Failed tests:  22, 26, 30
> Files=245, Tests=3402, 98 wallclock secs ( 0.71 usr  0.30 sys + 67.05
> cusr 12.05 csys = 80.11 CPU)
> Result: FAIL
> Failed 2/245 test programs. 3/3402 subtests failed.
>
>
> This is what i see in the error_log:
>
> [Tue Sep 24 22:56:10.012601 2019] [perl:error] [pid 743] [client
> 127.0.0.1:57720] APR::Finfo::stat: (70008) Partial results are valid but
> processing is incomplete at
> /tmp/mod_perl-2.0.11-rc2/t/response/TestAPI/request_rec.pm line 168
>
> Not sure if this is a concern or not, but it seems like it should be?
>

Yes, it's definitely a concern! We have two reports of exactly the
same failure now - this and the one from Damyan Ivanov.

This is exactly the kind of problem that I was getting with libapr's
apr_stat() function (see
https://bz.apache.org/bugzilla/show_bug.cgi?id=51560), which I fixed
for Windows in rev. 1865449. As noted before I do see various test
failures on Windows, but none relating to apr_stat() now, and I don't
get any api/request_rec.t failure at all.

This seems to imply either that other OSes are getting similar
problems that need working around similarly to Windows, or else that
the Windows fixes I've applied have broken other OSes. In this case
I'm wondering if it might the latter:-

The rev. cited above changed exactly the line that the failure is
reported on - line 168 in t/response/TestAPI/request_rec.pm... I
changed FINFO_NORM to FINFO_NAME since it's only fname that we access
so we shouldn't need all the other fields. (Windows has a problem
getting all the fields (specifically, it often fails on the
APR_FINFO_GPROT and APR_FINFO_WPROT bits) so that change dodged the
problem for me.)

What happens if you revert the change in that file? I.e. Change
FINFO_NAME back to FINFO_NORM on line 168, and on line 18 (the
APR::Const -compile line)? Does that fix it for you?

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] mod_perl-2.0.11 RC1

[Steve Hay]

On Tue, 24 Sep 2019 at 15:32, Philippe M. Chiasson  wrote:
>
> On 2019-09-24 4:08 a.m., Steve Hay wrote:
> > Actually, the compat/request.t failure was a trivial fix - done in
> > rev. 1867432. I will make an RC2 with this fix. Thanks for prompting
> > me to look! :-)
>
> Saw similar failures on OS X as well. I'll wait for RC2 to start testing
> again fresh

RC2 now uploaded with this and a couple of other fixes that have come
to my attention.


>
> >
> > On Tue, 24 Sep 2019 at 08:19, Steve Hay  wrote:
> >> I see the compat/request.t failure too, but I also got that in my 2.0.10 
> >> setup.
> >>
> >> (I also get modperl/env.t, modperl/setupenv.t and preconnection.note.t
> >> failures, which also occurred in 2.0.10, plus some lingering
> >> filter/in_bbs_inject_header.t failures that occurred in 2.0.10 which I
> >> thought would be fixed now but aren't. I also have one other report of
> >> the latter. For me it only fails when LWP is present.)
> >>
> >> On Tue, 24 Sep 2019 at 04:04, Adam Prime  wrote:
> >>> I'm seeing more test failures for 2.0.11 RC1 than for 2.0.10 with the
> >>> same setup.
> >>>
> >>> Test Summary Report
> >>> ---
> >>> t/api/request_rec.t   (Wstat: 0 Tests: 43 Failed: 0)
> >>> Parse errors: Bad plan.  You planned 55 tests but ran 43.
> >>> t/compat/request.t(Wstat: 0 Tests: 12 Failed: 0)
> >>> Parse errors: Bad plan.  You planned 22 tests but ran 12.
> >>> t/filter/in_bbs_inject_header.t   (Wstat: 0 Tests: 36 Failed: 3)
> >>> Failed tests:  22, 26, 30
> >>> Files=245, Tests=3392, 102 wallclock secs ( 0.76 usr  0.34 sys + 72.85
> >>> cusr 11.27 csys = 85.22 CPU)
> >>> Result: FAIL
> >>> Failed 3/245 test programs. 3/3392 subtests failed.
> >>>
> >>>
> >>> The compat/request.t failures don't occur on the previous version. Is
> >>> this expected?  I'm using perl 5.28.0, and  httpd 2.4.41 prefork.
> >>>
> >>> Adam
> >>>
> >>> On 9/2/19 8:34 AM, Steve Hay wrote:
> >>>> Please download, test, and report back on this mod_perl 2.0.11 release
> >>>> candidate.
> >>>>
> >>>> https://dist.apache.org/repos/dist/dev/perl/mod_perl-2.0.11-rc1.tar.gz
> >>>>
> >>>> MD5  = 417823274b32e5ca8759cf3760ad1591
> >>>> SHA1 = e47c72337e6766c403d0a76b59d3808625e5162b
> >>>>
> >>>> Major changes in this release are as follows:
> >>>>
> >>>> Fix use-after-free segfault in ap_server_config_defines seen on
> >>>> start-up on OpenBSD. [Found/fixed by Sam Vaughan/Joe Orton]
> >>>>
> >>>> Fix build with Perls earlier than 5.13.6. [Rainer Jung
> >>>> ]
> >>>>
> >>>> Fix filter/in_bbs_inject_header.t test failure with Apache 2.4.25+.
> >>>> [Stefan Fritsch ]
> >>>>
> >>>> Fix apache/read.t test failure with Apache 2.4.25+. [Niko Tyni
> >>>> ]
> >>>>
> > -
> > To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
> > For additional commands, e-mail: dev-h...@perl.apache.org
> >
>
>
> -
> To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
> For additional commands, e-mail: dev-h...@perl.apache.org
>

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[RELEASE CANDIDATE] mod_perl-2.0.11 RC2

[Steve Hay]

Please download, test, and report back on this mod_perl 2.0.11 release
candidate.

https://dist.apache.org/repos/dist/dev/perl/mod_perl-2.0.11-rc2.tar.gz

MD5  = abc2c2168121a09b0fc3b6fc6adc00bd
SHA1 = 36ee7626506a3a461118b3957814dfe9331ef1de

Changes since RC1 are as follows:

Fix [CVE-2011-2767] Arbitrary Perl code execution in the context of the user
account via a user-owned .htaccess. Patch from bugs.debian.org #644169. [Jan
Ingvoldstad ]

Fix potential test suite hangs due to pipelined response deadlocks. Patch
from rt.cpan.org #82409. [Zefram ]

Fix t/compat/request.t failures [Steve Hay]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] mod_perl-2.0.11 RC1

[Steve Hay]

Actually, the compat/request.t failure was a trivial fix - done in
rev. 1867432. I will make an RC2 with this fix. Thanks for prompting
me to look! :-)

On Tue, 24 Sep 2019 at 08:19, Steve Hay  wrote:
>
> I see the compat/request.t failure too, but I also got that in my 2.0.10 
> setup.
>
> (I also get modperl/env.t, modperl/setupenv.t and preconnection.note.t
> failures, which also occurred in 2.0.10, plus some lingering
> filter/in_bbs_inject_header.t failures that occurred in 2.0.10 which I
> thought would be fixed now but aren't. I also have one other report of
> the latter. For me it only fails when LWP is present.)
>
> On Tue, 24 Sep 2019 at 04:04, Adam Prime  wrote:
> >
> > I'm seeing more test failures for 2.0.11 RC1 than for 2.0.10 with the
> > same setup.
> >
> > Test Summary Report
> > ---
> > t/api/request_rec.t   (Wstat: 0 Tests: 43 Failed: 0)
> >Parse errors: Bad plan.  You planned 55 tests but ran 43.
> > t/compat/request.t(Wstat: 0 Tests: 12 Failed: 0)
> >Parse errors: Bad plan.  You planned 22 tests but ran 12.
> > t/filter/in_bbs_inject_header.t   (Wstat: 0 Tests: 36 Failed: 3)
> >Failed tests:  22, 26, 30
> > Files=245, Tests=3392, 102 wallclock secs ( 0.76 usr  0.34 sys + 72.85
> > cusr 11.27 csys = 85.22 CPU)
> > Result: FAIL
> > Failed 3/245 test programs. 3/3392 subtests failed.
> >
> >
> > The compat/request.t failures don't occur on the previous version. Is
> > this expected?  I'm using perl 5.28.0, and  httpd 2.4.41 prefork.
> >
> > Adam
> >
> > On 9/2/19 8:34 AM, Steve Hay wrote:
> > > Please download, test, and report back on this mod_perl 2.0.11 release
> > > candidate.
> > >
> > > https://dist.apache.org/repos/dist/dev/perl/mod_perl-2.0.11-rc1.tar.gz
> > >
> > > MD5  = 417823274b32e5ca8759cf3760ad1591
> > > SHA1 = e47c72337e6766c403d0a76b59d3808625e5162b
> > >
> > > Major changes in this release are as follows:
> > >
> > > Fix use-after-free segfault in ap_server_config_defines seen on
> > > start-up on OpenBSD. [Found/fixed by Sam Vaughan/Joe Orton]
> > >
> > > Fix build with Perls earlier than 5.13.6. [Rainer Jung
> > > ]
> > >
> > > Fix filter/in_bbs_inject_header.t test failure with Apache 2.4.25+.
> > > [Stefan Fritsch ]
> > >
> > > Fix apache/read.t test failure with Apache 2.4.25+. [Niko Tyni
> > > ]
> > >

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] mod_perl-2.0.11 RC1

[Steve Hay]

I see the compat/request.t failure too, but I also got that in my 2.0.10 setup.

(I also get modperl/env.t, modperl/setupenv.t and preconnection.note.t
failures, which also occurred in 2.0.10, plus some lingering
filter/in_bbs_inject_header.t failures that occurred in 2.0.10 which I
thought would be fixed now but aren't. I also have one other report of
the latter. For me it only fails when LWP is present.)

On Tue, 24 Sep 2019 at 04:04, Adam Prime  wrote:
>
> I'm seeing more test failures for 2.0.11 RC1 than for 2.0.10 with the
> same setup.
>
> Test Summary Report
> ---
> t/api/request_rec.t   (Wstat: 0 Tests: 43 Failed: 0)
>Parse errors: Bad plan.  You planned 55 tests but ran 43.
> t/compat/request.t(Wstat: 0 Tests: 12 Failed: 0)
>Parse errors: Bad plan.  You planned 22 tests but ran 12.
> t/filter/in_bbs_inject_header.t   (Wstat: 0 Tests: 36 Failed: 3)
>Failed tests:  22, 26, 30
> Files=245, Tests=3392, 102 wallclock secs ( 0.76 usr  0.34 sys + 72.85
> cusr 11.27 csys = 85.22 CPU)
> Result: FAIL
> Failed 3/245 test programs. 3/3392 subtests failed.
>
>
> The compat/request.t failures don't occur on the previous version. Is
> this expected?  I'm using perl 5.28.0, and  httpd 2.4.41 prefork.
>
> Adam
>
> On 9/2/19 8:34 AM, Steve Hay wrote:
> > Please download, test, and report back on this mod_perl 2.0.11 release
> > candidate.
> >
> > https://dist.apache.org/repos/dist/dev/perl/mod_perl-2.0.11-rc1.tar.gz
> >
> > MD5  = 417823274b32e5ca8759cf3760ad1591
> > SHA1 = e47c72337e6766c403d0a76b59d3808625e5162b
> >
> > Major changes in this release are as follows:
> >
> > Fix use-after-free segfault in ap_server_config_defines seen on
> > start-up on OpenBSD. [Found/fixed by Sam Vaughan/Joe Orton]
> >
> > Fix build with Perls earlier than 5.13.6. [Rainer Jung
> > ]
> >
> > Fix filter/in_bbs_inject_header.t test failure with Apache 2.4.25+.
> > [Stefan Fritsch ]
> >
> > Fix apache/read.t test failure with Apache 2.4.25+. [Niko Tyni
> > ]
> >

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] mod_perl-2.0.11 RC1

[Steve Hay]

On Mon, 2 Sep 2019 at 13:37, Steve Hay  wrote:
>
> On Mon, 2 Sep 2019 at 13:34, Steve Hay  wrote:
> >
> > Please download, test, and report back on this mod_perl 2.0.11 release
> > candidate.
> >
> > https://dist.apache.org/repos/dist/dev/perl/mod_perl-2.0.11-rc1.tar.gz
> >
>
> So far +1 from me in the following builds:
>
> Apache 2.4.33 / Perl 5.28.0 / VC2017 v15.9 (x64) with/without LWP
> Apache 2.4.41 / Perl 5.31.3 / VC2019 v16.1 (x64) with/without LWP

It's sad that there have not been any replies to this in more than a
fortnight... :-/

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] mod_perl-2.0.11 RC1

[Steve Hay]

On Mon, 2 Sep 2019 at 13:34, Steve Hay  wrote:
>
> Please download, test, and report back on this mod_perl 2.0.11 release
> candidate.
>
> https://dist.apache.org/repos/dist/dev/perl/mod_perl-2.0.11-rc1.tar.gz
>

So far +1 from me in the following builds:

Apache 2.4.33 / Perl 5.28.0 / VC2017 v15.9 (x64) with/without LWP
Apache 2.4.41 / Perl 5.31.3 / VC2019 v16.1 (x64) with/without LWP

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[RELEASE CANDIDATE] mod_perl-2.0.11 RC1

[Steve Hay]

Please download, test, and report back on this mod_perl 2.0.11 release
candidate.

https://dist.apache.org/repos/dist/dev/perl/mod_perl-2.0.11-rc1.tar.gz

MD5  = 417823274b32e5ca8759cf3760ad1591
SHA1 = e47c72337e6766c403d0a76b59d3808625e5162b

Major changes in this release are as follows:

Fix use-after-free segfault in ap_server_config_defines seen on
start-up on OpenBSD. [Found/fixed by Sam Vaughan/Joe Orton]

Fix build with Perls earlier than 5.13.6. [Rainer Jung
]

Fix filter/in_bbs_inject_header.t test failure with Apache 2.4.25+.
[Stefan Fritsch ]

Fix apache/read.t test failure with Apache 2.4.25+. [Niko Tyni
]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[ANNOUNCE] Apache-Test-1.42

[Steve Hay]

We are pleased to announce the release of Apache-Test 1.42.

Apache::Test is a test toolkit for testing an Apache server with any
configuration. It works with Apache 1.3 and Apache 2.0/2.2/2.4 and any
of its modules, including mod_perl 1.0 and 2.0. It was originally
developed for testing mod_perl 2.0.

This release is now, or soon will be, available for download from your
favourite CPAN mirror or you can find it at:

https://metacpan.org/release/Apache-Test

Checksums for this release are:

MD5  = 12f29c786eb1be2c56576936feae7d72
SHA1 = 767fc56283daa5a69ffb775667f99891cef2c549

Major changes in this release are as follows:

Fix loading apache_test_config.pm for recent perls in which '.' is no
longer in @INC by default. [Steve Hay]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] Apache-Test-1.42 RC1

[Steve Hay]

On Mon, 19 Aug 2019 at 14:10, Steve Hay  wrote:
>
> Please download, test, and report back on this Apache-Test 1.42
> release candidate.
>
> https://dist.apache.org/repos/dist/dev/perl/Apache-Test-1.42-rc1.tar.gz
>
> MD5  = 2dd753a50d94ee1705f77039d5da2f3a
> SHA1 = c07e19bb63fe5ef845a91fefea58a7c28bfedf34
>
> The only change in this release is:
>
> Fix loading apache_test_config.pm for recent perls in which '.' is
> no longer in @INC by default. [Steve Hay]

+1 from me using VC++ 2019 64-bit, Perl 5.30.0, Apache 2.4.39, with
and without the current Trunk version of mod_perl (soon to be 2.0.11
RC1 once I get this updated Apache-Test in).

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[RELEASE CANDIDATE] Apache-Test-1.42 RC1

[Steve Hay]

Please download, test, and report back on this Apache-Test 1.42
release candidate.

https://dist.apache.org/repos/dist/dev/perl/Apache-Test-1.42-rc1.tar.gz

MD5  = 2dd753a50d94ee1705f77039d5da2f3a
SHA1 = c07e19bb63fe5ef845a91fefea58a7c28bfedf34

The only change in this release is:

Fix loading apache_test_config.pm for recent perls in which '.' is
no longer in @INC by default. [Steve Hay]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[ANNOUNCE] Apache-Test-1.41

[Steve Hay]

We are pleased to announce the release of Apache-Test 1.41.

Apache::Test is a test toolkit for testing an Apache server with any
configuration. It works with Apache 1.3 and Apache 2.0/2.2/2.4 and any
of its modules, including mod_perl 1.0 and 2.0. It was originally
developed for testing mod_perl 2.0.

This release is now, or soon will be, available for download from your
favourite CPAN mirror or you can find it at:

https://metacpan.org/release/Apache-Test

Checksums for this release are:

MD5  = 0ea5f076e8aabd30f27384460b85c457
SHA1 = dc89d4669f356d5a8545ddb545385ec7b0772e93

Major changes in this release are as follows:

Set DefaultStateDir for > 2.5.1 and add -t_state to override. [jorton]

Inherit config via IncludeOptional as well as Include. [jorton]

Increase size of MinSpare, MaxSpare and MaxClients to improve httpd
test framework runs with worker and preform MPMs. [rjung]

Changed the openssl version detection to work with other *SSL libraries. [icing]

Switch test framework from using Net::SSL for raw TLS sockets to
IO::Socket::SSL. [rjung]

Fix mod_ssl tests under OpenSSL 1.1.1 / TLSv1.3. [jorton]

Add cwd to generated lib path in TEST script since Perl >=5.26 don't
do that any more. [jorton]

Override loglevel to trace8 if running in 2.4. [covener]

Allow an empty PREFIX. [sf]

Add need_min_apache_fix(). [covener]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] Apache-Test-1.41 RC1

[Steve Hay]

On Wed, 10 Jul 2019 at 22:07, Philippe M. Chiasson  wrote:
>
> Tested on Fedora 30 with Apache/2.4.39, looking good to me
>
> That's my +1
>

Thanks to all for testing. With the +1 from Adam Prime (sent
privately) that's enough :-)

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] Apache-Test-1.41 RC1

[Steve Hay]

On Wed, 3 Jul 2019 at 08:45, Steve Hay  wrote:
>
> Please download, test, and report back on this Apache-Test 1.41
> release candidate.
>
> https://dist.apache.org/repos/dist/dev/perl/Apache-Test-1.41-rc1.tar.gz
>

+1 from me using the following setups:

VC++ 2013 32-bit, Perl 5.30.0, Apache 2.2.34
VC++ 2019 64-bit, Perl 5.30.0, Apache 2.4.39

Both setups tested in four configurations: With & without mod_perl
2.0.10 / with & without LWP.

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[RELEASE CANDIDATE] Apache-Test-1.41 RC1

[Steve Hay]

Please download, test, and report back on this Apache-Test 1.41
release candidate.

https://dist.apache.org/repos/dist/dev/perl/Apache-Test-1.41-rc1.tar.gz

MD5  = 7933d3a6a762f087bf7883a1ac2086eb
SHA1 = 17aa9a8669023aa2f485aa83f8f389969b8e5f0c

Major changes in this release are as follows:

Set DefaultStateDir for > 2.5.1 and add -t_state to override. [jorton]

Inherit config via IncludeOptional as well as Include. [jorton]

Increase size of MinSpare, MaxSpare and MaxClients to improve httpd
test framework runs with worker and preform MPMs. [rjung]

Changed the openssl version detection to work with other *SSL libraries. [icing]

Switch test framework from using Net::SSL for raw TLS sockets to
IO::Socket::SSL. [rjung]

Fix mod_ssl tests under OpenSSL 1.1.1 / TLSv1.3. [jorton]

Add cwd to generated lib path in TEST script since Perl >=5.26 don't
do that any more. [jorton]

Override loglevel to trace8 if running in 2.4. [covener]

Allow an empty PREFIX. [sf]

Add need_min_apache_fix(). [covener]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: Fix for Segfault with Apache and mod_perl

[Steve Hay]

Yes, I'd like to do a 2.0.11 soon: We have a few bug fixes committed
since 2.0.10 and this crash fix seems like a good time to make a new
release.
I'll try to get the wheels in motion for an RC1 :-)

On Fri, 21 Jun 2019 at 08:56, Sam Vaughan  wrote:
>
> That's great thanks Steve, much appreciated!
>
> Will there be a 2.0.11 release with this fix?
>
> Cheers,
>
> Sam
>
> > On 21 Jun 2019, at 5:38 pm, Steve Hay  wrote:
> >
> > Thanks for the report, Sam (and to Joe for the fix). This is now
> > committed to mod_perl trunk:
> > http://svn.apache.org/viewvc?view=revision&revision=1861755
> >
> > On Thu, 20 Jun 2019 at 09:07, Steve Hay  wrote:
> >>
> >> Looks fine to me too. Will give it a test...
> >>
> >> On Thu, 20 Jun 2019 at 01:28, Philippe Chiasson  
> >> wrote:
> >>>
> >>> Looks good to me
> >>>
> >>> Sent from the depths of my mind on an iPhone
> >>>
> >>>> On Jun 19, 2019, at 17:22, Sam Vaughan  wrote:
> >>>>
> >>>> Hi mod_perl team,
> >>>>
> >>>> I reported a bug yesterday to the Apache team, describing a segfault 
> >>>> that only happens if mod_perl is installed:
> >>>>
> >>>> https://bz.apache.org/bugzilla/show_bug.cgi?id=63516
> >>>>
> >>>> My impression was that it should be fixed in httpd2, but mod_perl is the 
> >>>> only module that adds a configuration flag to the global 
> >>>> ap_server_config_defines array, so it has been suggested that a one-line 
> >>>> change go into the modperl_register_hooks() function instead:
> >>>>
> >>>> Index: src/modules/perl/mod_perl.c
> >>>> --- src/modules/perl/mod_perl.c.orig
> >>>> +++ src/modules/perl/mod_perl.c
> >>>> @@ -861,7 +861,7 @@
> >>>>
> >>>>/* for  and Apache2->define("MODPERL2") */
> >>>>*(char **)apr_array_push(ap_server_config_defines) =
> >>>> -apr_pstrdup(p, "MODPERL2");
> >>>> +apr_pstrdup(ap_server_config_defines->pool, "MODPERL2");
> >>>>
> >>>>ap_hook_pre_config(modperl_hook_pre_config,
> >>>>   NULL, NULL, APR_HOOK_MIDDLE);
> >>>>
> >>>> Essentially, the pool provided as an argument to this function doesn't 
> >>>> have a long enough lifetime to safely be used as backing storage for the 
> >>>> "MODPERL2" string.  This change uses the pool that the 
> >>>> ap_server_config_defines array itself was created in, thereby giving the 
> >>>> "MODPERL2" string the same lifetime as the array it is being added to.
> >>>>
> >>>> As far as I can tell, this use-after-free bug has been sitting there for 
> >>>> quite a few years now but Apache only segfaults if your OS is strict 
> >>>> enough about reclaiming/unmapping freed memory.
> >>>>
> >>>> Currently, this issue causes httpd2 to segfault during startup the 
> >>>> majority of the time on OpenBSD 6.5.
> >>>>
> >>>> I'd really appreciate it if someone could either confirm that the above 
> >>>> patch is OK and apply it to mod_perl, or jump on the above Apache bug 
> >>>> report and add a comment explaining why it isn't.
> >>>>
> >>>> Thanks!
> >>>>
> >>>> Sam
> >>>>
> >>>>
> >>>> -
> >>>> To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
> >>>> For additional commands, e-mail: dev-h...@perl.apache.org
> >>>>
> >>>
> >>>
> >>> -
> >>> To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
> >>> For additional commands, e-mail: dev-h...@perl.apache.org
> >>>
>
>
> -
> To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
> For additional commands, e-mail: dev-h...@perl.apache.org
>

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: Fix for Segfault with Apache and mod_perl

[Steve Hay]

Thanks for the report, Sam (and to Joe for the fix). This is now
committed to mod_perl trunk:
http://svn.apache.org/viewvc?view=revision&revision=1861755

On Thu, 20 Jun 2019 at 09:07, Steve Hay  wrote:
>
> Looks fine to me too. Will give it a test...
>
> On Thu, 20 Jun 2019 at 01:28, Philippe Chiasson  wrote:
> >
> > Looks good to me
> >
> > Sent from the depths of my mind on an iPhone
> >
> > > On Jun 19, 2019, at 17:22, Sam Vaughan  wrote:
> > >
> > > Hi mod_perl team,
> > >
> > > I reported a bug yesterday to the Apache team, describing a segfault that 
> > > only happens if mod_perl is installed:
> > >
> > > https://bz.apache.org/bugzilla/show_bug.cgi?id=63516
> > >
> > > My impression was that it should be fixed in httpd2, but mod_perl is the 
> > > only module that adds a configuration flag to the global 
> > > ap_server_config_defines array, so it has been suggested that a one-line 
> > > change go into the modperl_register_hooks() function instead:
> > >
> > > Index: src/modules/perl/mod_perl.c
> > > --- src/modules/perl/mod_perl.c.orig
> > > +++ src/modules/perl/mod_perl.c
> > > @@ -861,7 +861,7 @@
> > >
> > > /* for  and Apache2->define("MODPERL2") */
> > > *(char **)apr_array_push(ap_server_config_defines) =
> > > -apr_pstrdup(p, "MODPERL2");
> > > +apr_pstrdup(ap_server_config_defines->pool, "MODPERL2");
> > >
> > > ap_hook_pre_config(modperl_hook_pre_config,
> > >NULL, NULL, APR_HOOK_MIDDLE);
> > >
> > > Essentially, the pool provided as an argument to this function doesn't 
> > > have a long enough lifetime to safely be used as backing storage for the 
> > > "MODPERL2" string.  This change uses the pool that the 
> > > ap_server_config_defines array itself was created in, thereby giving the 
> > > "MODPERL2" string the same lifetime as the array it is being added to.
> > >
> > > As far as I can tell, this use-after-free bug has been sitting there for 
> > > quite a few years now but Apache only segfaults if your OS is strict 
> > > enough about reclaiming/unmapping freed memory.
> > >
> > > Currently, this issue causes httpd2 to segfault during startup the 
> > > majority of the time on OpenBSD 6.5.
> > >
> > > I'd really appreciate it if someone could either confirm that the above 
> > > patch is OK and apply it to mod_perl, or jump on the above Apache bug 
> > > report and add a comment explaining why it isn't.
> > >
> > > Thanks!
> > >
> > > Sam
> > >
> > >
> > > -
> > > To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
> > > For additional commands, e-mail: dev-h...@perl.apache.org
> > >
> >
> >
> > -
> > To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
> > For additional commands, e-mail: dev-h...@perl.apache.org
> >

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: Fix for Segfault with Apache and mod_perl

[Steve Hay]

Looks fine to me too. Will give it a test...

On Thu, 20 Jun 2019 at 01:28, Philippe Chiasson  wrote:
>
> Looks good to me
>
> Sent from the depths of my mind on an iPhone
>
> > On Jun 19, 2019, at 17:22, Sam Vaughan  wrote:
> >
> > Hi mod_perl team,
> >
> > I reported a bug yesterday to the Apache team, describing a segfault that 
> > only happens if mod_perl is installed:
> >
> > https://bz.apache.org/bugzilla/show_bug.cgi?id=63516
> >
> > My impression was that it should be fixed in httpd2, but mod_perl is the 
> > only module that adds a configuration flag to the global 
> > ap_server_config_defines array, so it has been suggested that a one-line 
> > change go into the modperl_register_hooks() function instead:
> >
> > Index: src/modules/perl/mod_perl.c
> > --- src/modules/perl/mod_perl.c.orig
> > +++ src/modules/perl/mod_perl.c
> > @@ -861,7 +861,7 @@
> >
> > /* for  and Apache2->define("MODPERL2") */
> > *(char **)apr_array_push(ap_server_config_defines) =
> > -apr_pstrdup(p, "MODPERL2");
> > +apr_pstrdup(ap_server_config_defines->pool, "MODPERL2");
> >
> > ap_hook_pre_config(modperl_hook_pre_config,
> >NULL, NULL, APR_HOOK_MIDDLE);
> >
> > Essentially, the pool provided as an argument to this function doesn't have 
> > a long enough lifetime to safely be used as backing storage for the 
> > "MODPERL2" string.  This change uses the pool that the 
> > ap_server_config_defines array itself was created in, thereby giving the 
> > "MODPERL2" string the same lifetime as the array it is being added to.
> >
> > As far as I can tell, this use-after-free bug has been sitting there for 
> > quite a few years now but Apache only segfaults if your OS is strict enough 
> > about reclaiming/unmapping freed memory.
> >
> > Currently, this issue causes httpd2 to segfault during startup the majority 
> > of the time on OpenBSD 6.5.
> >
> > I'd really appreciate it if someone could either confirm that the above 
> > patch is OK and apply it to mod_perl, or jump on the above Apache bug 
> > report and add a comment explaining why it isn't.
> >
> > Thanks!
> >
> > Sam
> >
> >
> > -
> > To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
> > For additional commands, e-mail: dev-h...@perl.apache.org
> >
>
>
> -
> To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
> For additional commands, e-mail: dev-h...@perl.apache.org
>

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

New Windows mod_perl binaries available

[Steve Hay]

For those who are interested, I have uploaded binary 64-bit builds of
mod_perl-2.0.10 and libapreq2-2.14 components compatible with Apache
Lounge 2.4.27 (Win64, VC15, built 10 July 2017) and each of:

* ActivePerl 5.22.3.2204 (x64)
* Strawberry Perl 5.22.3.1 (64bit)

I will make a 32-bit build compatible with Strawberry Perl's "32bit
without USE_64_BIT_INT" version tomorrow, followed by similar
ActivePerl/Strawberry Perl 5.24.1 versions in the next couple of days.

You can download them from http://home.apache.org/~stevehay/

Regards,
Steve

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: svn commit: r1717474 - in /perl/modperl/trunk: Changes src/modules/perl/mod_perl.c src/modules/perl/modperl_env.c src/modules/perl/modperl_env.h src/modules/perl/modperl_perl.c t/response/TestModp

[Steve Hay]

On 4 January 2017 at 15:33, Rainer Jung  wrote:
> Am 04.01.2017 um 15:51 schrieb Steve Hay:
>>
>> On 4 January 2017 at 14:21, Rainer Jung  wrote:
>>>
>>> Am 04.01.2017 um 14:54 schrieb Steve Hay:
>>>>
>>>>
>>>> On 4 January 2017 at 13:27, Rainer Jung  wrote:
>>>>>
>>>>>
>>>>> Hi there,
>>>>>
>>>>> this commit broke compatibility for mod_perl 2.0.10 with perl before
>>>>> version
>>>>> 5.14. AFAIK API function mg_free_type() used here was only introduced
>>>>> in
>>>>> 5.14.0. Since the above commit was titled "Add support for Perl
>>>>> 5.22.x.",
>>>>> it
>>>>> might be possible to fix this using a version dependent code path. But
>>>>> I
>>>>> don't know enough about these internals to suggest an appropriate
>>>>> patch,
>>>>> sorry.
>>>>>
>>>>
>>>> Thanks for the report.
>>>>
>>>> Are you in a position to try the attached patch, which I think should
>>>> be sufficient to fix it? (mg_free() removes magic of all types, but I
>>>> wouldn't expect any magic other than the type targetted by the
>>>> mg_free_type() call to be present anyway.)
>>>>
>>>> If not then I'll build a 5.14 and try it myself, but I don't have one
>>>> to hand at the moment.
>>>
>>>
>>>
>>> Wow, thanks for the quick patch suggestion. Yes I can test, e.g. using
>>> perl
>>> 5.10.1 on RHEL6. Any specific test scenario, that you would suggest so
>>> that
>>> the code path in question is triggered?
>>>
>>
>> Anything will do - the code is used in start-up and shutdown, so
>> you'll know straight away if it's failing because it probably won't
>> even start up.
>>
>> If you're able to run the test suite as well then that will exercise
>> all the ENV-related code in-between as well, but I think that should
>> all be fine.
>
>
> built and tested on RHEL 6 using platform tools gcc 4.4.5-6 and perl
> v5.10.1. but Apache 2.4.4 (custom build).
>
> Build succeeds without mg_free_type warning (as expected), make test also
> succeeds:
>

Great! Thanks for testing. Now committed in r1777355.

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: svn commit: r1717474 - in /perl/modperl/trunk: Changes src/modules/perl/mod_perl.c src/modules/perl/modperl_env.c src/modules/perl/modperl_env.h src/modules/perl/modperl_perl.c t/response/TestModp

[Steve Hay]

On 4 January 2017 at 14:21, Rainer Jung  wrote:
> Am 04.01.2017 um 14:54 schrieb Steve Hay:
>>
>> On 4 January 2017 at 13:27, Rainer Jung  wrote:
>>>
>>> Hi there,
>>>
>>> this commit broke compatibility for mod_perl 2.0.10 with perl before
>>> version
>>> 5.14. AFAIK API function mg_free_type() used here was only introduced in
>>> 5.14.0. Since the above commit was titled "Add support for Perl 5.22.x.",
>>> it
>>> might be possible to fix this using a version dependent code path. But I
>>> don't know enough about these internals to suggest an appropriate patch,
>>> sorry.
>>>
>>
>> Thanks for the report.
>>
>> Are you in a position to try the attached patch, which I think should
>> be sufficient to fix it? (mg_free() removes magic of all types, but I
>> wouldn't expect any magic other than the type targetted by the
>> mg_free_type() call to be present anyway.)
>>
>> If not then I'll build a 5.14 and try it myself, but I don't have one
>> to hand at the moment.
>
>
> Wow, thanks for the quick patch suggestion. Yes I can test, e.g. using perl
> 5.10.1 on RHEL6. Any specific test scenario, that you would suggest so that
> the code path in question is triggered?
>

Anything will do - the code is used in start-up and shutdown, so
you'll know straight away if it's failing because it probably won't
even start up.

If you're able to run the test suite as well then that will exercise
all the ENV-related code in-between as well, but I think that should
all be fine.

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: svn commit: r1717474 - in /perl/modperl/trunk: Changes src/modules/perl/mod_perl.c src/modules/perl/modperl_env.c src/modules/perl/modperl_env.h src/modules/perl/modperl_perl.c t/response/TestModp

[Steve Hay]

On 4 January 2017 at 13:27, Rainer Jung  wrote:
> Hi there,
>
> this commit broke compatibility for mod_perl 2.0.10 with perl before version
> 5.14. AFAIK API function mg_free_type() used here was only introduced in
> 5.14.0. Since the above commit was titled "Add support for Perl 5.22.x.", it
> might be possible to fix this using a version dependent code path. But I
> don't know enough about these internals to suggest an appropriate patch,
> sorry.
>

Thanks for the report.

Are you in a position to try the attached patch, which I think should
be sufficient to fix it? (mg_free() removes magic of all types, but I
wouldn't expect any magic other than the type targetted by the
mg_free_type() call to be present anyway.)

If not then I'll build a 5.14 and try it myself, but I don't have one
to hand at the moment.
Index: src/modules/perl/modperl_env.c
===
--- src/modules/perl/modperl_env.c	(revision 1777309)
+++ src/modules/perl/modperl_env.c	(working copy)
@@ -670,7 +670,11 @@
mg->mg_ptr, mg->mg_obj, mg->mg_flags);
 
 /* Remove it */
+#if MP_PERL_VERSION_AT_LEAST(5, 13, 6)
 mg_free_type((SV*)ENVHV, PERL_MAGIC_env);
+#else
+mg_free((SV*)ENVHV);
+#endif
 
 /* Add our version instead */
 mg = sv_magicext((SV*)ENVHV, (SV*)NULL, PERL_MAGIC_env, &MP_vtbl_env, (char*)NULL, 0);
@@ -699,7 +703,11 @@
mg->mg_ptr, mg->mg_obj, mg->mg_flags);
 
 /* Remove it */
+#if MP_PERL_VERSION_AT_LEAST(5, 13, 6)
 mg_free_type((SV*)ENVHV, PERL_MAGIC_env);
+#else
+mg_free((SV*)ENVHV);
+#endif
 
 /* Restore perl's original version */
 sv_magicext((SV*)ENVHV, (SV*)NULL, PERL_MAGIC_env, &PL_vtbl_env, (char*)NULL, 0);

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: Absorb win32-apxs into httpd distro?

[Steve Hay]

On 14 December 2016 at 08:13, William A Rowe Jr  wrote:
> Randy wrote http://www.apache.org/dist/perl/win32-bin/ - but I'm wondering
> who else here at httpd is interested in helping maintain and get this code
> into our own distribution? I've shipped this for a decade for my commercial
> customers, for every wrinkle and wart, I think many of our win32 users
> would appreciate this.
>
> I trust the perl pmc and Randy would be good with this adoption?
>

Randy sadly passed away some years ago, but I for one would welcome
the adoption.

The latest version that I'm aware of lives in
https://svn.apache.org/repos/asf/perl/apxs/trunk

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: apreq release

[Steve Hay]

[Resending from an address that's actually subscribed to the lists...]

The current mod_perl release (2.0.10) is taken from trunk. (The
httpd24 branch was only for development work leading towards the
previous release (2.0.9), and is now obsolete.)

On 17 November 2016 at 09:03, Issac Goldstand  wrote:
> Yes, if we go through with this.  The mod_apreq2 stuff is in httpd.
> However I just realized that although it's been in trunk for years now,
> it's never been backported to the 2.4.x branch.  I'm not sure why, and
> my google-fu isn't finding relevant discussion on dev@httpd.
>
> When I initially took a look yesterday, I thought I'd looked at a 2.4
> release.  Can someone from the mod_perl dev community tell me if new
> releases (intended for httpd 2.4) come from trunk or from the httpd24
> branch?  If it's from the branch, then the apreq stuff could be moved to
> trunk (or a submodule, like Apache::Reload and friends if desired).  If
> trunk, then until mod_apreq is backported to httpd 2.4 (or httpd bumps
> versions and rebranches off of trunk)
>
>
>
> On 11/16/2016 3:40 PM, Brian J. France wrote:
>> This is just merging the perl stuff into mod_perl, right?
>>
>> Not merging mod_apreq2 and all the request cache/re-play bucket, POST 
>> reading, file uploading, etc stuff, right?
>>
>> I really don't want to have to include mod_perl so my C modules can read 
>> POST data and handle file uploads.
>>
>> Cheers,
>>
>> Brian
>>
>>
>>> On Nov 16, 2016, at 4:42 AM, Issac Goldstand  wrote:
>>>
>>> Given that the C was (finally) merged into httpd years ago, and given
>>> that there are no proposed code changes, I'd say that's not such a bad
>>> idea...
>>>
>>> I've become a bit rusty in Perl (and even with apreq) over the years,
>>> but IIRC, all of the Perl glue is in
>>> http://svn.apache.org/repos/asf/httpd/apreq/trunk/glue/perl/
>>>
>>> I'll take a crack at seeing if I can fold it into mod_perl despite the rust.
>>>
>>>
>>> On 11/15/2016 2:45 PM, Andres Thomas Stivalet wrote:
 Good news!! No idea why apreq hasn't just been merged into mod_perl
 after all these years.

 A+++

 On Nov 15, 2016 3:27 AM, "Issac Goldstand" >>> > wrote:

Hi all,

Someone (finally) noticed that apreq's test suite isn't compatible with
Apache 2.4 and requested a change.  Given that we haven't released an
updated apreq in nearly 6 years, I'm inclined to make/test the changes
to the test suite and immediately go to a release cycle.

Does anyone want time to add anything else to libapreq-2.14 before I
start tarring and voting (in the next few days, I hope)?


-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org

For additional commands, e-mail: dev-h...@perl.apache.org


>>>
>>
>

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: apreq release

[Steve Hay]

On 15 November 2016 at 09:26, Issac Goldstand  wrote:
> Hi all,
>
> Someone (finally) noticed that apreq's test suite isn't compatible with
> Apache 2.4 and requested a change.  Given that we haven't released an
> updated apreq in nearly 6 years, I'm inclined to make/test the changes
> to the test suite and immediately go to a release cycle.
>
> Does anyone want time to add anything else to libapreq-2.14 before I
> start tarring and voting (in the next few days, I hope)?
>

All looks good here on Win32 - the current SVN version passes all
tests using VC++ 2010, perl-5.25.3, mod_perl-2.0.10 and either
httpd-2.2.31 or httpd-2.4.23. (Whatever the httpd-2.4 incompatibility
was seems not to affect anything on Win32!)

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[ANNOUNCE] mod_perl-2.0.10

[Steve Hay]

We are pleased to announce the release of mod_perl 2.0.10.

mod_perl is an Apache HTTP Server module for embedding a Perl
interpreter in your web server, giving you super-fast dynamic content
by avoiding the overhead of starting an external interpreter.

This release is now, or soon will be, available for download from a
mirror site near you via:

http://perl.apache.org/download/index.html

or in the meantime directly from:

http://apache.org/dist/perl/
http://search.cpan.org/dist/mod_perl/
https://metacpan.org/release/SHAY/mod_perl-2.0.10

Checksums for this release are:

MD5  = cef55e715b5770a63b3becbe9d271121
SHA1 = 61b5b0fe4449440258ad45dee6efa0e2264a9701

Major changes in this release are as follows:

Add support for Perl 5.22.x. [Niko Tyni , Steve Hay]

Fix non-threaded Perl 5.22.x build and tests. [Klaus S. Madsen
]

Automatically select the appropriate c89 option when modperl is being
built with either gcc 5 or clang. [Klaus S. Madsen ]

Declare MP_vtbl_env and MP_vtbl_envelem as 'extern' to fix linker
errors on OSX/Darwin. [Michael Schout ]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[RELEASE CANDIDATE] mod_perl-2.0.10 RC2

[Steve Hay]

Please download, test, and report back on this mod_perl 2.0.10 release
candidate.

https://dist.apache.org/repos/dist/dev/perl/mod_perl-2.0.10-rc2.tar.gz

MD5  = 087c5ef93c0dd4bd696827a9552b92d5
SHA1 = e774414cf07d80dc4d5bf81ba5abaf512a0758e6

Changes since RC1 are as follows:

Declare MP_vtbl_env and MP_vtbl_envelem as 'extern' to fix linker
errors on OSX/Darwin. [Michael Schout ]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] mod_perl-2.0.10 RC1

[Steve Hay]

On 30 September 2016 at 14:13, Steve Hay  wrote:
> Please download, test, and report back on this mod_perl 2.0.10 release
> candidate.
>
> https://dist.apache.org/repos/dist/dev/perl/mod_perl-2.0.10-rc1.tar.gz
>

+1 (modulo the expected failures noted in the RC announcement) on
Windows 8.1 / VC++ 2010 with all eight combinations of:

Perl 5.20.3, 5.22.3-RC2, 5.24.1-RC2, 5.25.3
Apache httpd 2.2.31, 2.4.23

except that t/filter/in_bbs_inject_header.t is also failing tests 22,
26 and 30 with httpd 2.4, although I don't think that's new from
looking through the archives, e.g. see:

https://marc.info/?l=apache-modperl&m=143696414409949&w=2 (2015-07-15)
https://marc.info/?l=apache-modperl-dev&m=146450325022969&w=2 (2016-05-29)
https://marc.info/?l=apache-modperl&m=146922430215771&w=2 (2016-07-22)

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[RELEASE CANDIDATE] mod_perl-2.0.10 RC1

[Steve Hay]

Please download, test, and report back on this mod_perl 2.0.10 release
candidate.

https://dist.apache.org/repos/dist/dev/perl/mod_perl-2.0.10-rc1.tar.gz

MD5  = f29feb099e0cff545da347b1bcead671
SHA1 = 07c8c656eecbc436ba285aa5a9a3c69947613a16

NOTE: Unfortunately, there is a new test failure on Windows when using
Perls built
with the PERL_IMPLICIT_SYS build option enabled (which it is by
default). This is in addition to the existing four known test failures
on Windows (t/protocol/echo_block.t, t/protocol/echo_nonblock.t,
t/protocol/echo_timeout.t and t/protocol/pseudo_http.t) which were
present in the previous release (2.0.9).

Major changes in this release are as follows:

Automatically select the appropriate c89 option when modperl is being
built with either gcc 5 or clang. [Klaus S. Madsen ]

Fix non-threaded Perl 5.22.x build and tests. [Klaus S. Madsen
]

Add support for Perl 5.22.x. [Niko Tyni , Steve Hay]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] Apache-Test-1.40 RC1

[Steve Hay]

On 28 August 2016 at 04:55, Jie Gao  wrote:
> I have just tried to build mod_perl from mod_perl-SVN as at Sun Aug 28 
> 03:25:32 UTC 2016, "make test" hang at:
>
> t/protocol/eliza.t .. 1/7
>

That may simply be because you haven't got Chatbot::Eliza installed?

Few people do have, which is why t/protocol/eliza.t is omitted from
the mod_perl distribution (by ModPerl::Manifest).

Sorry - this (and a few other cases like it) is a problem with testing
from SVN but I hope to get a mod_perl-2.0.10-RC1 out imminently, and
then testing will be easier :-)

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[ANNOUNCE] Apache-Test 1.40 released

[Steve Hay]

We are pleased to announce the release of Apache-Test 1.40.

This release is now, or soon will be, available for download from your
favourite CPAN mirror or you can find it at:

https://metacpan.org/release/Apache-Test

Checksums for this release are:

MD5  = 78e059bccfa0c2f0297ff2fca591eca9
SHA1 = 578210e8da1b01e624db8577c4d8ddfc3f221ae2

Major changes in this release are as follows:

Specify licence (Apache 2.0) in META.yml. [Steve Hay, CPAN RT#111359]

Fix broken POD. [Steve Hay]

Switch argument order in "openssl gendsa". [rjung]

Add (limited) checks for *_SAN_*_n and *_DN_Email variables. [kbrand]

Update key sizes and message digest to what is common in 2015. [kbrand]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Apache-Test-1.40-RC1 out, mod_perl-2.0.10-RC1 coming!

[Steve Hay]

Hi,

Hopefully you've seen by now that I've just released an RC1 for
Apache-Test-1.40.

The reason for this is that I'd like to make a new mod_perl release
(2.0.10) soon, and it's worth getting the latest Apache-Test changes
into it.

I still have new test failures in mod_perl in a default configuration
build on Windows. I can get back to the usual (as of mod_perl-2.0.9 /
httpd-2.4) set of failures by disabling the PERL_IMPLICIT_SYS build
option in perl.

That's on by default, but is Windows-specific, so I'm hoping that
things will be OK on other OSes. If that is indeed the case then I do
not want to hold up releasing 2.0.10 over this issue because it
contains support for perl >= 5.22, which is particularly important
right now because CVE-2016-1238 means that all perls < the imminent
5.22.3 and 5.24.1 are insecure. (Also, perls < 5.22 are officially out
of support as far as perl5-porters is concerned -- see
perlpolicy.pod.)

So in short, expect to see a mod_perl-2.0.10-RC1 soon after
Apache-Test-1.40 is done!

If anyone has any spare tuits for some pre-RC1 testing of what is
currently in SVN then that could only be a good thing.

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE] Apache-Test-1.40 RC1

[Steve Hay]

On 24 August 2016 at 18:02, Steve Hay  wrote:
> Please download, test, and report back on this Apache-Test 1.40
> release candidate.
>
> https://dist.apache.org/repos/dist/dev/perl/Apache-Test-1.40-rc1.tar.gz
>

+1 on Win8/VC10 with default configuration builds of:

Perl 5.20.3 + Apache 2.2.31
Perl 5.20.3 + Apache 2.4.23

Perl 5.22.3-RC2 + Apache 2.2.31
Perl 5.22.3-RC2 + Apache 2.4.23

Perl 5.24.1-RC2 + Apache 2.2.31
Perl 5.24.1-RC2 + Apache 2.4.23

Perl 5.25.3 + Apache 2.2.31
Perl 5.25.3 + Apache 2.4.23

all with and without mod_perl-SVN and LWP-6.15.

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[RELEASE CANDIDATE] Apache-Test-1.40 RC1

[Steve Hay]

Please download, test, and report back on this Apache-Test 1.40
release candidate.

https://dist.apache.org/repos/dist/dev/perl/Apache-Test-1.40-rc1.tar.gz

MD5  = ecabfbf3a782601aa1511ae3c89601b0
SHA1 = a046f0ccd3a696ecbcaa207dbe92a83c92e5f4c4

=item 1.40-rc1

Specify licence (Apache 2.0) in META.yml. [Steve Hay, CPAN RT#111359]

Fix broken POD. [Steve Hay]

Switch argument order in "openssl gendsa". [rjung]

Add (limited) checks for *_SAN_*_n and *_DN_Email variables. [kbrand]

Update key sizes and message digest to what is common in 2015. [kbrand]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: Automatic compiler flags for gcc 5 and clang

[Steve Hay]

On 3 March 2016 at 19:58, Klaus S. Madsen  wrote:
> Hi,
>
> Trying to get modperl compiling on a recent Ubuntu, I encountered the
> "inline semantics changed" issue, which is also mentioned here:
>
> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=777935#14
>
> To make compilation easier on gcc5/clang hosts, I've created the attached
> patch, which automatically selects the appropriate c89 option, when modperl
> is being built with either gcc 5 or clang.
>
> I've tested the patch on Ubuntu 15.10 (with gcc 5.2.1), Fedora 23 (with gcc
> 5.3.1) and FreeBSD 10.2 (with clang).
>

Thanks, committed in rThanks, committed in r1733566.

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: Compile errors with SVN revision 1733039 for unthreaded perl

[Steve Hay]

On 2 March 2016 at 15:35, Klaus S. Madsen  wrote:
> Sorry,
>
> A small correction. The test works with a "home compiled" perl 5.16.3
> without ithreads support. With 5.22.1 without ithreads support I run into
> the same issue that r1702395 should have solved with sprintf warnings.
>
> While I'm no expert in this, I believe that the comment in r1702395 is
> wrong. It's not because of the thread context that the number of arguments
> is wrong, it's because of the way WrapXS.pm handles functions with variable
> parameter length (i.e. ... functions).
>
> In the .xs code items is the number of elements in the MARK array. However
> when WrapXS.pm generates code for a ... function, it will increment MARK, so
> that it points to the second argument to the function. It will not, however
> decrement items, which means that items and MARK now are out of sync. I
> suspect that MARK is incremented to skip the object that the function is
> called on.
>
> I've attached a simple patch that updates the comment and removes the ifdef
> around the decrementation of items, which solves the test-failures for
> non-threaded perl 5.22.
>

Thanks, committed in r1733564.

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: Compile errors with SVN revision 1733039 for unthreaded perl

[Steve Hay]

On 2 March 2016 at 10:15, Klaus S. Madsen  wrote:
> Hi,
>
> I'm looking into upgrading our perl version to 5.22.1. Since mod_perl 2.0.9
> doen't work with 5.22, I thought that it would be a good idea to test the
> SVN version of mod_perl.
>
> We compile our own perl, and we compile it without ithread support, which
> the new env-support doesn't appear to work with. I get the following compile
> errors:
>
> modperl_env.c: In function 'modperl_env_init':
> modperl_env.c:657:10: error: 'my_perl' undeclared (first use in this
> function)
> modperl_env.c:657:10: note: each undeclared identifier is reported only once
> for each function it appears in
> modperl_env.c: In function 'modperl_env_unload':
> modperl_env.c:688:10: error: 'my_perl' undeclared (first use in this
> function)
>
> As far as I can tell, this is because pTHX expands to define a my_perl
> parameter for modperl_env_init and modperl_env_unload if perl is compiled
> with ithreads support. However without ithreads support pTHX expands to
> nothing, causing the compile error.
>
> To fix this, I've tried to simply delete the if (!my_perl) tests in
> modperl_env_init and modperl_env_unload. I'm fairly sure that this is safe
> to do for modperl_env_unload, since it is only called from
> modperl_perl_destruct, which appears to have a perl interpreter instance.
> However I'm not so sure for modperl_env_init.
>
> Removing those two if-statements reveal that we should call
> modperl_env_unload with aTHX as the parameter in modperl_perl_descruct, so
> that the number of arguments is correct both with and without ithread
> support enabled.
>
> The attached patch makes both changes. Apply with -p0.
>
> I've tested this against our "home-compiled" perl 5.22 where it now compiles
> and passes the tests. I've also tried against the Ubuntu 12.04 supplied perl
> with ithreads support, which also compiles and passes all the tests.
>

Thanks, committed in r1733563.

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[ANNOUNCE] mod_perl 2.0.9

[Steve Hay]

We are delighted to announce the long-awaited release of mod_perl
2.0.9, which you will soon be able to download from a mirror site near
you via:

http://perl.apache.org/download/index.html

or in the meantime directly from:

http://apache.org/dist/perl/
http://search.cpan.org/dist/mod_perl/
https://metacpan.org/release/mod_perl/

MD5: fdab9a145bf35f54fce997c96c76f8e2 *mod_perl-2.0.9.tar.gz
SHA1: 5de3018214da21de186d758a429d5c78f827caa5 *mod_perl-2.0.9.tar.gz

This is the first official release of mod_perl with support for httpd
2.4.x. Your humble release engineer would like to give a big thank you
to all those who helped to make it happen.

Please note that perl 5.22.0 is not supported by this release. We hope
to have this fixed soon in mod_perl 2.0.10.

See below for the full list of changes in this release.

Enjoy!

Steve Hay


Add note to README about MP_INLINE problem when building with GCC 5.
[Niko Tyni ]

Fix t/api/aplog.t for apr-1.5.2. [Steve Hay]

Note that Perl 5.22.x is currently not supported. This is logged as
CPAN RT#101962 and will hopefully be addressed in 2.0.10. [Steve Hay]

Fix unthreaded build, which was broken in 2.0.9-rc2. [Steve Hay]

Remove PerlInterpScope. This has not been working properly with
threaded MPMs with httpd-2.4.x and the use-case of this directive was
questionable. [Jan Kaluza]

Allow running the test suite with httpd-2.4.x when mod_access_compat
is not loaded. [Steve Hay]

Add support for Apache httpd-2.4.x. [Torsten Foertsch, Jan Kaluza,
Steve Hay, Gozer]

Don't call modperl_threaded_mpm() et al. from XS code. Fixes Debian
Bug #765174. [Niko Tyni ]

Make sure modperl_interp_select uses r->server rather than the passed
s parameter to find the interpreter pool to pull an interpreter from.
This fixes an issue with vhosts with a separate interpreter pool and
runtime dir-config merges that used to pull the interpreter from the
wrong pool. [Torsten Foertsch]

PerlInterpScope is now more advisory. Using $(c|r)->pnotes will bind
the current interpreter to that object for it's lifetime.
$(c|r)->pnotes_kill() can be used to prematurely drop pnotes and
remove this binding. [Torsten Foertsch]

Now correctly invokes PerlCleanupHandlers, even if they are the only
handler type configured for that request [Torsten Foertsch]

For threaded MPMs, change interpreter managment to a new,
reference-counted allocation model. [Torsten Foertsch]

Expose modperl_interp_pool_t via ModPerl::InterpPool, modperl_tipool_t
via ModPerl::TiPool and modperl_tipool_config_t via
ModPerl::TiPoolConfig [Torsten Foertsch]

Expose modperl_interp_t via ModPerl::Interpreter [Torsten Foertsch]

Fix t/compat/apache_file.t on Windows. Apache::File->tmpfile() wants
TMPDIR or TEMP from the environment, or else defaults to /tmp. The
latter is no good on Windows, so make sure the environment variables
are passed through. (TEMP should be set to something suitable on
Windows.) [Steve Hay]

Fix t/api/err_headers_out.t with HTTP::Headers > 6.00. [Rolando
]

Fix the build with VC++ and dmake (rather than nmake) on Windows. The
Makefile generated by Apache2::Build uses shell commands for the
manifest file, but neglected to tell dmake to use the shell. [Steve
Hay]

Don't write an 'rpm' target into the Makefile on Windows. It isn't
relevant on Windows, and the (hard-coded, not MakeMaker-generated)
recipe group has syntax which dmake doesn't understand. [Steve Hay]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: mod_perl2, MP_INLINE and gcc 5

[Steve Hay]

On 17 June 2015 at 19:57, Niko Tyni  wrote:
> On Wed, Jun 17, 2015 at 08:27:28PM +0300, Niko Tyni wrote:
>
>> mod_perl2 (up to and including 2.0.9-RC3) fails to build with GCC 5,
>> because the default inline semantics have changed from GNU89 to C99 and
>> require changes in mod_perl source.
>
> Ah, reading my own references properly, it looks like it might be a
> bit more complicated than that, and probably only happens when apr and
> mod_perl are compiled with different inline semantics. Is that correct?
>
> Sorry for the confusion. I still think a note somewhere could help...

I was just about to roll the release this evening when I saw this. A
minor README update surely doesn't warrant a new RC so I'm prepared to
slip the note in now. Is the wording still accurate, given that the
situation is more complicated than you first thought?

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE]: mod_perl-2.0.9 RC3

[Steve Hay]

On 10 Jun 2015 21:01, "David E. Wheeler"  wrote:
>
> On Jun 10, 2015, at 10:13 AM, Steve Hay 
wrote:
>
> > Note that Perl 5.22.x is currently not supported. This is logged as
> > CPAN RT#101962 and will hopefully be addressed in 2.0.10. [Steve Hay]
>
> Oh, bummer. I was just looking at this, as it’s core-dumping. Is it
difficult to fix? Is 2.0.10 likely to come sooner than 2.0.9 did?
>

I first came across it a while ago and hoped to have 2.0.9 out before
5.22.0, so I put it to one side at the time.

Unfortunately 5.22.0 has beaten us to a release, but I didn't want to hold
up 2.0.9 any longer to try to fix it.

Rest assured, though, that 2.0.10 won't be anything like as long coming as
2.0.9 has been. I hope the fix won't be too difficult, in which case we
should have 2.0.10 fairly soon after 2.0.9. It might also give us a chance
to address any more problems found with 2.0.9, especially with its httpd
2.4.x support, after it's escaped into the wild!

Re: [RELEASE CANDIDATE]: mod_perl-2.0.9 RC2

[Steve Hay]

On 10 June 2015 at 09:50, Jan Kaluža  wrote:
> On 06/09/2015 10:23 AM, Steve Hay wrote:
>>
>> On 8 June 2015 at 08:14, Jan Kaluža  wrote:
>>>
>>> On 05/31/2015 03:12 AM, Kevin A. McGrail wrote:
>>>>
>>>>
>>>> On 5/30/2015 2:20 PM, Steve Hay wrote:
>>>>>
>>>>>
>>>>> Please download, test, and report back on this release candidate of
>>>>> the long-awaited mod_perl 2.0.9.
>>>>>
>>>>> http://people.apache.org/~stevehay/mod_perl-2.0.9-rc2.tar.gz
>>>>>
>>>>> MD5 = 6dbf61e0724a245f1c015687fb3e4213
>>>>> SHA1 = faf63c98b338ef0373ccbbe775c52506270ff3b9
>>>>>
>>>>> Changes since RC1:
>>>>>
>>>>> Remove PerlInterpScope. This has not been working properly with
>>>>> threaded MPMs with httpd-2.4.x and the use-case of this directive was
>>>>> questionable. [Jan Kaluza]
>>>>>
>>>>> Allow running the test suite with httpd-2.4.x when mod_access_compat
>>>>> is not loaded. [Steve Hay]
>>>>
>>>>
>>>> The test suite finished with one failure against 2.4.12 on my test rig:
>>>>
>>>> Test Summary Report
>>>> ---
>>>> t/api/aplog.t (Wstat: 0 Tests: 36 Failed: 1)
>>>> Failed test:  24
>>>> Files=245, Tests=4646, 304 wallclock secs ( 3.34 usr  0.49 sys + 227.74
>>>> cusr 50.53 csys = 282.10 CPU)
>>>> Result: FAIL
>>>> Failed 1/245 test programs. 1/4646 subtests failed.
>>>> [warning] server localhost.localdomain:8529 shutdown
>>>> [warning] port 8529 still in use...
>>>> ...done
>>>> [  error] error running tests (please examine t/logs/error_log)
>>>> ++
>>>> | Please file a bug report: http://perl.apache.org/bugs/ |
>>>> ++
>>>> Makefile:1112: recipe for target 'run_tests' failed
>>>> make: *** [run_tests] Error 1
>>>>
>>>>
>>>>
>>>> Repeating with TEST_VERBOSE=1 added...  Here's the relevant output:
>>>>
>>>> ok 23
>>>> # testing : $s->log_serror(LOG_MARK, LOG_DEBUG, APR::Const::EGENERAL...)
>>>> # expected: qr/(?^:Internal error: log_serror test 2)/
>>>> # received: '
>>>> # *** The following warn entry is expected and harmless ***
>>>> # [Sat May 30 21:04:59.085379 2015] [perl:debug] [pid 9642:tid 3076]
>>>> aplog.pm(90): (20014)Internal error (specific information not
>>>> available): log_serror test 2
>>>> # '
>>>> not ok 24
>>>> # testing : $r->log_rerror(LOG_MARK, LOG_CRIT, APR::Const::ENOTIME...)
>>>> # expected: qr/(?^:\[\w*:crit\] \[pid[^]]+\] .*?: \[[^]]+\] log_rerror
>>>> test)/
>>>> # received: '# Failed test 24 in
>>>> /usr/src/mod_perl-2.0.9-rc2/t/response/TestAPI/aplog.pm at line 92
>>>
>>>
>>>
>>> Something is bad here I think, because test on line 92 is not the one
>>> which
>>> is failing according to the TEST_VERBOSE output. Could you check what's
>>> in
>>> aplog.pm at line 92?
>>>
>>> Jan Kaluza
>>>
>>>> # *** The following error entry is expected and harmless ***
>>>> # [Sat May 30 21:04:59.086325 2015] [perl:crit] [pid 9642:tid 3076]
>>>> (20007)No time was provided and one was required.: [client
>>>> 127.0.0.1:56778] log_rerror test
>>>> # '
>>>> ok 25
>>
>>
>> Ah... I think the "testing/expected/received" text is appearing
>> *before* the "ok/not ok" text.
>>
>> Test 25 has succeeded because the quoted text from "# Failed test 24"
>> through "log_rerror test #" does indeed match the pattern
>> qr/\[\w*:crit\] \[pid[^]]+\] .*?: \[[^]]+\] log_rerror test/
>>
>> But it is test 24 on lin 92 that has failed because the received text
>>
>> # *** The following warn entry is expected and harmless ***
>> # [Sat May 30 21:04:59.085379 2015] [perl:debug] [pid 9642:tid 3076]
>> aplog.pm(90): (20014)Internal error (specific information not
>> available): log_serror test 2
>> #
>>
>> doesn't match the pattern qr/$egeneral: log_serror test 2/, where
>>
>>  my $egeneral = have_min_apache_version('2.1.0')
>> ? "Internal error"
>> : "Error string not specified yet";
>>
>> It fails to match because of the " (specific information not available)"
>> part.
>>
>> Does the following/attached patch fix it?
>
>
> Great! I've been able to reproduce it. That error string changed in
> apr-1.5.2 and I've been testing with apr-1.5.1 earlier. I've verified that
> the patch you've attached fixes the issue for me.
>
> I've tested with the patch with both apr-1.5.2 and apr-1.5.1 and it worked
> in both cases for me, so +1 from my point of view.
>

Thanks for testing. I've now committed this patch, and just uploaded
RC3. Hopefully there will be nothing unexpected this time!

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[RELEASE CANDIDATE]: mod_perl-2.0.9 RC3

[Steve Hay]

Please download, test, and report back on this release candidate of
the long-awaited mod_perl 2.0.9.

http://people.apache.org/~stevehay/mod_perl-2.0.9-rc3.tar.gz

MD5 = 61d07fe00919d9da2b49dbf7b821b1a7
SHA1 = 09e1d5f19312742db9da38c8e7f8955a77d29dfd

Changes since RC2:

Fix t/api/aplog.t for apr-1.5.2. [Steve Hay]

Note that Perl 5.22.x is currently not supported. This is logged as
CPAN RT#101962 and will hopefully be addressed in 2.0.10. [Steve Hay]

Fix unthreaded build, which was broken in 2.0.9-rc2. [Steve Hay]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE]: mod_perl-2.0.9 RC2

[Steve Hay]

On 9 June 2015 at 23:42, Gregg Smith  wrote:
> I need to remember reply-all in this list.
>
>
> On 6/9/2015 12:56 AM, Steve Hay wrote:
>>
>> On 4 June 2015 at 05:47, Gregg Smith  wrote:
>>>
>>> On 6/3/2015 1:03 AM, Steve Hay wrote:
>>>>
>>>>
>>>> On 2 June 2015 at 23:30, Gregg Smith   wrote:
>>>>>
>>>>>
>>>>> On 6/2/2015 10:35 AM, Steve Hay wrote:
>>>>>>
>>>>>>
>>>>>> On 2 June 2015 at 17:52, Gregg Smithwrote:
>>>>>>>
>>>>>>>
>>>>>>> On 6/2/2015 12:49 AM, Steve Hay wrote:
>>>>>>>>
>>>>>>>>
>>>>>>>> On 1 June 2015 at 18:59, Gregg Smith wrote:
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> On 5/30/2015 11:20 AM, Steve Hay wrote:
>>>>
>>>>
>>>> Is there anything more in t/logs/error_log?
>>>
>>>
>>>
>>> Sorry for the delay, had two very busy days.
>>>
>>> The 403 in both_str_req_proxy.t is no real surprise to me and
>>> mod_access_compat being the culprit is what I had guessed.
>>> mod_access_compat
>>> should not be relied upon on 2.4. It certainly does not work as I had
>>> expected it to when I switched over from 2.2 and I see support questions
>>> because of it in forums all the time.
>>>
>>> The proxy.t error might just be Vista when I think about it since you do
>>> not
>>> get the same on Win7. Vista has a very limited number of connections that
>>> can be open. I should upgrade to 8.1 but it's so much work to get all my
>>> other needed software installed.
>>>
>>>
>>>
>>> t\filter\both_str_req_proxy.t ...
>>> 1..1
>>> # Running under perl version 5.016003 for MSWin32
>>> # Win32::BuildNumber 1604
>>> # Current time local: Wed Jun  3 20:48:44 2015
>>> # Current time GMT:   Thu Jun  4 03:48:44 2015
>>> # Using Test.pm version 1.26
>>> # Using Apache/Test.pm version 1.39
>>> # testing : lc input and reverse output filters
>>> # Failed test 1 in t\filter\both_str_req_proxy.t at line 18
>>> # expected: 'abcdefghijklmnopqrstuvwxyz0123456789'
>>> # received: '
>>> #
>>> #403 Forbidden
>>> #
>>> #Forbidden
>>> #You don\'t have permission to access
>>> /TestFilter__both_str_req_proxy/foo
>>> # on this server.
>>> #
>>> #
>>> #Apache/2.4.12 (Win32) world domination series/2.0
>>> OpenSSL/1.0.1m
>>> mod_perl/2.0.9-rc2 Perl/
>>> v5.16.3 Server at gls3 Port 8529
>>> #
>>> # '
>>> not ok 1
>>> Failed 1/1 subtests
>>>
>>> [Wed Jun 03 20:48:43.885200 2015] [charset_lite:debug] [pid 3124:tid
>>> 2344]
>>> mod_charset_lite.c(216): [client 10.0.0.33:59366] AH01448: incomplete
>>> configuration: src unspecified, dst unspecified
>>> [Wed Jun 03 20:48:43.885200 2015] [deflate:debug] [pid 3124:tid 2344]
>>> mod_deflate.c(853): [client 10.0.0.33:59366] AH01384: Zlib: Compressed 21
>>> to
>>> 26 : URL /TestFilter__both_str_req_mix
>>> [Wed Jun 03 20:48:50.842800 2015] [access_compat:error] [pid 3124:tid
>>> 2344]
>>> [client 10.0.0.33:59367] AH01797: client denied by server configuration:
>>> proxy:http://GLS3:8529/TestFilter__both_str_req_proxy_content/foo
>>> [Wed Jun 03 20:48:57.909600 2015] [authz_core:debug] [pid 3124:tid 2344]
>>> mod_authz_core.c(834): [client 10.0.0.33:59368] AH01628: authorization
>>> result: granted (no directives)
>>> [Wed Jun 03 20:48:57.909600 2015] [charset_lite:debug] [pid 3124:tid
>>> 2344]
>>> mod_charset_lite.c(216): [client 10.0.0.33:59368] AH01448: incomplete
>>> configuration: src unspecified, dst unspecified
>>>
>>>
>>>
>>> # connecting to http://GLS3:8538/TestModules__proxy
>>> 1..1
>>> # Running under perl version 5.016003 for MSWin32
>>> # Win32::BuildNumber 1604
>>> # Current time local: Wed Jun  3 21:04:05 2015
>>> # Current time GMT:   Thu Jun  4 04:04:05 2015
>>> # Using Test.pm version 1.26
>>> # Using Apache/Test.pm version 1.39
>>> request has failed (the response code was: 503)
>>> see t/logs/error_log for more details
>>> Dubious, test returned 255 (wstat 65280, 0xff00)
>>> Failed 

Re: [RELEASE CANDIDATE]: mod_perl-2.0.9 RC2

[Steve Hay]

On 8 June 2015 at 08:14, Jan Kaluža  wrote:
> On 05/31/2015 03:12 AM, Kevin A. McGrail wrote:
>>
>> On 5/30/2015 2:20 PM, Steve Hay wrote:
>>>
>>> Please download, test, and report back on this release candidate of
>>> the long-awaited mod_perl 2.0.9.
>>>
>>> http://people.apache.org/~stevehay/mod_perl-2.0.9-rc2.tar.gz
>>>
>>> MD5 = 6dbf61e0724a245f1c015687fb3e4213
>>> SHA1 = faf63c98b338ef0373ccbbe775c52506270ff3b9
>>>
>>> Changes since RC1:
>>>
>>> Remove PerlInterpScope. This has not been working properly with
>>> threaded MPMs with httpd-2.4.x and the use-case of this directive was
>>> questionable. [Jan Kaluza]
>>>
>>> Allow running the test suite with httpd-2.4.x when mod_access_compat
>>> is not loaded. [Steve Hay]
>>
>> The test suite finished with one failure against 2.4.12 on my test rig:
>>
>> Test Summary Report
>> ---
>> t/api/aplog.t (Wstat: 0 Tests: 36 Failed: 1)
>>Failed test:  24
>> Files=245, Tests=4646, 304 wallclock secs ( 3.34 usr  0.49 sys + 227.74
>> cusr 50.53 csys = 282.10 CPU)
>> Result: FAIL
>> Failed 1/245 test programs. 1/4646 subtests failed.
>> [warning] server localhost.localdomain:8529 shutdown
>> [warning] port 8529 still in use...
>> ...done
>> [  error] error running tests (please examine t/logs/error_log)
>> ++
>> | Please file a bug report: http://perl.apache.org/bugs/ |
>> ++
>> Makefile:1112: recipe for target 'run_tests' failed
>> make: *** [run_tests] Error 1
>>
>>
>>
>> Repeating with TEST_VERBOSE=1 added...  Here's the relevant output:
>>
>> ok 23
>> # testing : $s->log_serror(LOG_MARK, LOG_DEBUG, APR::Const::EGENERAL...)
>> # expected: qr/(?^:Internal error: log_serror test 2)/
>> # received: '
>> # *** The following warn entry is expected and harmless ***
>> # [Sat May 30 21:04:59.085379 2015] [perl:debug] [pid 9642:tid 3076]
>> aplog.pm(90): (20014)Internal error (specific information not
>> available): log_serror test 2
>> # '
>> not ok 24
>> # testing : $r->log_rerror(LOG_MARK, LOG_CRIT, APR::Const::ENOTIME...)
>> # expected: qr/(?^:\[\w*:crit\] \[pid[^]]+\] .*?: \[[^]]+\] log_rerror
>> test)/
>> # received: '# Failed test 24 in
>> /usr/src/mod_perl-2.0.9-rc2/t/response/TestAPI/aplog.pm at line 92
>
>
> Something is bad here I think, because test on line 92 is not the one which
> is failing according to the TEST_VERBOSE output. Could you check what's in
> aplog.pm at line 92?
>
> Jan Kaluza
>
>> # *** The following error entry is expected and harmless ***
>> # [Sat May 30 21:04:59.086325 2015] [perl:crit] [pid 9642:tid 3076]
>> (20007)No time was provided and one was required.: [client
>> 127.0.0.1:56778] log_rerror test
>> # '
>> ok 25

Ah... I think the "testing/expected/received" text is appearing
*before* the "ok/not ok" text.

Test 25 has succeeded because the quoted text from "# Failed test 24"
through "log_rerror test #" does indeed match the pattern
qr/\[\w*:crit\] \[pid[^]]+\] .*?: \[[^]]+\] log_rerror test/

But it is test 24 on lin 92 that has failed because the received text

# *** The following warn entry is expected and harmless ***
# [Sat May 30 21:04:59.085379 2015] [perl:debug] [pid 9642:tid 3076]
aplog.pm(90): (20014)Internal error (specific information not
available): log_serror test 2
#

doesn't match the pattern qr/$egeneral: log_serror test 2/, where

my $egeneral = have_min_apache_version('2.1.0')
   ? "Internal error"
   : "Error string not specified yet";

It fails to match because of the " (specific information not available)" part.

Does the following/attached patch fix it?

Index: t/response/TestAPI/aplog.pm
===
--- t/response/TestAPI/aplog.pm (revision 1684346)
+++ t/response/TestAPI/aplog.pm (working copy)
@@ -83,7 +83,7 @@

 # the APR_EGENERAL error string changed for APR 1.0
 my $egeneral = have_min_apache_version('2.1.0')
-   ? "Internal error"
+   ? qr/Internal error(?: \(specific information not available\))?/
: "Error string not specified yet";

 t_server_log_warn_is_expected();
Index: t/response/TestAPI/aplog.pm
===
--- t/response/TestAPI/aplog.pm	(revision 1684346)
+++ t/response/TestAPI/aplog.pm	(working copy)
@@ -83,7 +83,7 @@
 
 # the APR_EGENERAL error string changed for APR 1.0
 my $egeneral = have_min_apache_version('2.1.0')
-   ? "Internal error"
+   ? qr/Internal error(?: \(specific information not available\))?/
: "Error string not specified yet";
 
 t_server_log_warn_is_expected();

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE]: mod_perl-2.0.9 RC2

[Steve Hay]

On 4 June 2015 at 05:47, Gregg Smith  wrote:
> On 6/3/2015 1:03 AM, Steve Hay wrote:
>>
>> On 2 June 2015 at 23:30, Gregg Smith  wrote:
>>>
>>> On 6/2/2015 10:35 AM, Steve Hay wrote:
>>>>
>>>> On 2 June 2015 at 17:52, Gregg Smith   wrote:
>>>>>
>>>>> On 6/2/2015 12:49 AM, Steve Hay wrote:
>>>>>>
>>>>>> On 1 June 2015 at 18:59, Gregg Smithwrote:
>>>>>>>
>>>>>>> On 5/30/2015 11:20 AM, Steve Hay wrote:
>>
>> Is there anything more in t/logs/error_log?
>
>
> Sorry for the delay, had two very busy days.
>
> The 403 in both_str_req_proxy.t is no real surprise to me and
> mod_access_compat being the culprit is what I had guessed. mod_access_compat
> should not be relied upon on 2.4. It certainly does not work as I had
> expected it to when I switched over from 2.2 and I see support questions
> because of it in forums all the time.
>
> The proxy.t error might just be Vista when I think about it since you do not
> get the same on Win7. Vista has a very limited number of connections that
> can be open. I should upgrade to 8.1 but it's so much work to get all my
> other needed software installed.
>
>
>
> t\filter\both_str_req_proxy.t ...
> 1..1
> # Running under perl version 5.016003 for MSWin32
> # Win32::BuildNumber 1604
> # Current time local: Wed Jun  3 20:48:44 2015
> # Current time GMT:   Thu Jun  4 03:48:44 2015
> # Using Test.pm version 1.26
> # Using Apache/Test.pm version 1.39
> # testing : lc input and reverse output filters
> # Failed test 1 in t\filter\both_str_req_proxy.t at line 18
> # expected: 'abcdefghijklmnopqrstuvwxyz0123456789'
> # received: '
> # 
> # 403 Forbidden
> # 
> # Forbidden
> # You don\'t have permission to access
> /TestFilter__both_str_req_proxy/foo
> # on this server.
> # 
> # 
> # Apache/2.4.12 (Win32) world domination series/2.0 OpenSSL/1.0.1m
> mod_perl/2.0.9-rc2 Perl/
> v5.16.3 Server at gls3 Port 8529
> # 
> # '
> not ok 1
> Failed 1/1 subtests
>
> [Wed Jun 03 20:48:43.885200 2015] [charset_lite:debug] [pid 3124:tid 2344]
> mod_charset_lite.c(216): [client 10.0.0.33:59366] AH01448: incomplete
> configuration: src unspecified, dst unspecified
> [Wed Jun 03 20:48:43.885200 2015] [deflate:debug] [pid 3124:tid 2344]
> mod_deflate.c(853): [client 10.0.0.33:59366] AH01384: Zlib: Compressed 21 to
> 26 : URL /TestFilter__both_str_req_mix
> [Wed Jun 03 20:48:50.842800 2015] [access_compat:error] [pid 3124:tid 2344]
> [client 10.0.0.33:59367] AH01797: client denied by server configuration:
> proxy:http://GLS3:8529/TestFilter__both_str_req_proxy_content/foo
> [Wed Jun 03 20:48:57.909600 2015] [authz_core:debug] [pid 3124:tid 2344]
> mod_authz_core.c(834): [client 10.0.0.33:59368] AH01628: authorization
> result: granted (no directives)
> [Wed Jun 03 20:48:57.909600 2015] [charset_lite:debug] [pid 3124:tid 2344]
> mod_charset_lite.c(216): [client 10.0.0.33:59368] AH01448: incomplete
> configuration: src unspecified, dst unspecified
>
>
>
> # connecting to http://GLS3:8538/TestModules__proxy
> 1..1
> # Running under perl version 5.016003 for MSWin32
> # Win32::BuildNumber 1604
> # Current time local: Wed Jun  3 21:04:05 2015
> # Current time GMT:   Thu Jun  4 04:04:05 2015
> # Using Test.pm version 1.26
> # Using Apache/Test.pm version 1.39
> request has failed (the response code was: 503)
> see t/logs/error_log for more details
> Dubious, test returned 255 (wstat 65280, 0xff00)
> Failed 1/1 subtests
>
> [Wed Jun 03 21:04:13.855200 2015] [proxy:debug] [pid 3124:tid 2340]
> mod_proxy.c(1161): [client 10.0.0.33:59664] AH01143: Running scheme http
> handler (attempt 0)
> [Wed Jun 03 21:04:13.855200 2015] [proxy_fcgi:debug] [pid 3124:tid 2340]
> mod_proxy_fcgi.c(859): [client 10.0.0.33:59664] AH01076: url:
> http://gls3:8538/TestModules__proxy_real proxyname: (null) proxyport: 0
> [Wed Jun 03 21:04:13.855200 2015] [proxy_fcgi:debug] [pid 3124:tid 2340]
> mod_proxy_fcgi.c(864): [client 10.0.0.33:59664] AH01077: declining URL
> http://gls3:8538/TestModules__proxy_real
> [Wed Jun 03 21:04:13.855200 2015] [proxy_scgi:debug] [pid 3124:tid 2340]
> mod_proxy_scgi.c(516): [client 10.0.0.33:59664] AH00865: declining URL
> http://gls3:8538/TestModules__proxy_real
> [Wed Jun 03 21:04:13.855200 2015] [proxy:debug] [pid 3124:tid 2340]
> proxy_util.c(2138): AH00942: HTTP: has acquired connection for (*)
> [Wed Jun 03 21:04:13.855200 2015] [proxy:debug] [pid 3124:tid 2340]
> proxy_util.c(2192): [client 10.0.0.33:59664] AH00944: connecting
> http://gls3:8538/TestModules__proxy_real to gls3:8538
> [Wed Jun 03 21:04:

Re: [RELEASE CANDIDATE]: mod_perl-2.0.9 RC2

[Steve Hay]

On 5 June 2015 at 13:01, Jan Kaluža  wrote:
> On 06/04/2015 06:47 AM, Gregg Smith wrote:
>>
>> On 6/3/2015 1:03 AM, Steve Hay wrote:
>>>
>>> On 2 June 2015 at 23:30, Gregg Smith  wrote:
>>>>
>>>> On 6/2/2015 10:35 AM, Steve Hay wrote:
>>>>>
>>>>> On 2 June 2015 at 17:52, Gregg Smith   wrote:
>>>>>>
>>>>>> On 6/2/2015 12:49 AM, Steve Hay wrote:
>>>>>>>
>>>>>>> On 1 June 2015 at 18:59, Gregg Smithwrote:
>>>>>>>>
>>>>>>>> On 5/30/2015 11:20 AM, Steve Hay wrote:
>>>
>>> Is there anything more in t/logs/error_log?
>>
>>
>> Sorry for the delay, had two very busy days.
>>
>> The 403 in both_str_req_proxy.t is no real surprise to me and
>> mod_access_compat being the culprit is what I had guessed.
>> mod_access_compat should not be relied upon on 2.4. It certainly does
>> not work as I had expected it to when I switched over from 2.2 and I see
>> support questions because of it in forums all the time.
>
>
> Could you please try to fix it, since you are able to reproduce this issue?
> Or is there anyone else with Windows who can reproduce it and fix it?
>
> It would be really great to spend some time fixing these bugs which cannot
> be reproduced globally, so we can finally release 2.4.x support :).

I intend to look at it, but if anyone else wants to get there first
then please feel free :-)


>
>> The proxy.t error might just be Vista when I think about it since you do
>> not get the same on Win7. Vista has a very limited number of connections
>> that can be open. I should upgrade to 8.1 but it's so much work to get
>> all my other needed software installed.
>>
>>
>>
>> t\filter\both_str_req_proxy.t ...
>> 1..1
>> # Running under perl version 5.016003 for MSWin32
>> # Win32::BuildNumber 1604
>> # Current time local: Wed Jun  3 20:48:44 2015
>> # Current time GMT:   Thu Jun  4 03:48:44 2015
>> # Using Test.pm version 1.26
>> # Using Apache/Test.pm version 1.39
>> # testing : lc input and reverse output filters
>> # Failed test 1 in t\filter\both_str_req_proxy.t at line 18
>> # expected: 'abcdefghijklmnopqrstuvwxyz0123456789'
>> # received: '
>> # 
>> # 403 Forbidden
>> # 
>> # Forbidden
>> # You don\'t have permission to access
>> /TestFilter__both_str_req_proxy/foo
>> # on this server.
>> # 
>> # 
>> # Apache/2.4.12 (Win32) world domination series/2.0
>> OpenSSL/1.0.1m mod_perl/2.0.9-rc2 Perl/
>> v5.16.3 Server at gls3 Port 8529
>> # 
>> # '
>> not ok 1
>> Failed 1/1 subtests
>>
>> [Wed Jun 03 20:48:43.885200 2015] [charset_lite:debug] [pid 3124:tid
>> 2344] mod_charset_lite.c(216): [client 10.0.0.33:59366] AH01448:
>> incomplete configuration: src unspecified, dst unspecified
>> [Wed Jun 03 20:48:43.885200 2015] [deflate:debug] [pid 3124:tid 2344]
>> mod_deflate.c(853): [client 10.0.0.33:59366] AH01384: Zlib: Compressed
>> 21 to 26 : URL /TestFilter__both_str_req_mix
>> [Wed Jun 03 20:48:50.842800 2015] [access_compat:error] [pid 3124:tid
>> 2344] [client 10.0.0.33:59367] AH01797: client denied by server
>> configuration:
>> proxy:http://GLS3:8529/TestFilter__both_str_req_proxy_content/foo
>> [Wed Jun 03 20:48:57.909600 2015] [authz_core:debug] [pid 3124:tid 2344]
>> mod_authz_core.c(834): [client 10.0.0.33:59368] AH01628: authorization
>> result: granted (no directives)
>> [Wed Jun 03 20:48:57.909600 2015] [charset_lite:debug] [pid 3124:tid
>> 2344] mod_charset_lite.c(216): [client 10.0.0.33:59368] AH01448:
>> incomplete configuration: src unspecified, dst unspecified
>>
>>
>>
>> # connecting to http://GLS3:8538/TestModules__proxy
>> 1..1
>> # Running under perl version 5.016003 for MSWin32
>> # Win32::BuildNumber 1604
>> # Current time local: Wed Jun  3 21:04:05 2015
>> # Current time GMT:   Thu Jun  4 04:04:05 2015
>> # Using Test.pm version 1.26
>> # Using Apache/Test.pm version 1.39
>> request has failed (the response code was: 503)
>> see t/logs/error_log for more details
>> Dubious, test returned 255 (wstat 65280, 0xff00)
>> Failed 1/1 subtests
>>
>> [Wed Jun 03 21:04:13.855200 2015] [proxy:debug] [pid 3124:tid 2340]
>> mod_proxy.c(1161): [client 10.0.0.33:59664] AH01143: Running scheme http
>> handler (attempt 0)
>> [Wed Jun 03 21:04:13.855200 2015] [proxy_fcgi:debug] [pid 3124:tid 2340]
>> mod_proxy_fcgi.c(859): [client 10.0

Re: [RELEASE CANDIDATE]: mod_perl-2.0.9 RC2

[Steve Hay]

On 2 June 2015 at 23:30, Gregg Smith  wrote:
> On 6/2/2015 10:35 AM, Steve Hay wrote:
>>
>> On 2 June 2015 at 17:52, Gregg Smith  wrote:
>>>
>>> On 6/2/2015 12:49 AM, Steve Hay wrote:
>>>>
>>>>
>>>> On 1 June 2015 at 18:59, Gregg Smith   wrote:
>>>>>
>>>>>
>>>>> On 5/30/2015 11:20 AM, Steve Hay wrote:
>>>>>>
>>>>>>
>>>>>>
>>>>>> Please download, test, and report back on this release candidate of
>>>>>> the long-awaited mod_perl 2.0.9.
>>>>>>
>>>>>> http://people.apache.org/~stevehay/mod_perl-2.0.9-rc2.tar.gz
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> Server version: Apache/2.4.12 (Win32)
>>>>> Server built:   Mar 19 2015 18:23:40
>>>>> Server loaded:  APR 1.5.1, APR-UTIL 1.5.4
>>>>> Architecture:   32-bit
>>>>> Server MPM: WinNT
>>>>> All module loaded
>>>>>
>>>>> Test Summary Report
>>>>> ---
>>>>> t\filter\both_str_req_proxy.t (Wstat: 0 Tests: 1 Failed: 1)
>>>>> Failed test:  1
>>>>> t\modperl\setupenv.t  (Wstat: 0 Tests: 63 Failed: 6)
>>>>> Failed tests:  8, 22, 29, 36, 50, 57
>>>>> t\modules\proxy.t (Wstat: 65280 Tests: 0 Failed: 0)
>>>>> Non-zero exit status: 255
>>>>> Parse errors: Bad plan.  You planned 1 tests but ran 0.
>>>>> t\preconnection\note.t(Wstat: 0 Tests: 1 Failed: 1)
>>>>> Failed test:  1
>>>>
>>>>
>>>>
>>>> I get setupenv.t and note.t both failing on one test machine (Windows
>>>> 8.1), but not on my other (Windows 7). I haven't looked into those
>>>> yet.
>>>>
>>>> Can you send verbose output of the both_str_req_proxy.t and proxy.t
>>>> failures? I haven't seen those before.
>>>>
>>>>
>>>>> t\protocol\echo_block.t   (Wstat: 0 Tests: 3 Failed: 2)
>>>>> Failed tests:  2-3
>>>>> t\protocol\echo_nonblock.t(Wstat: 0 Tests: 3 Failed: 1)
>>>>> Failed test:  2
>>>>> t\protocol\echo_timeout.t (Wstat: 0 Tests: 5 Failed: 4)
>>>>> Failed tests:  2-5
>>>>> t\protocol\pseudo_http.t  (Wstat: 0 Tests: 13 Failed: 9)
>>>>> Failed tests:  3-8, 11-13
>>>>
>>>>
>>>>
>>>> These four failures are documented in README as known problems on
>>>> Windows with httpd-2.4. I haven't figured out why they fail yet. They
>>>> work fine on httpd-2.2.
>>>>
>>>> I wondered if they had something to do with APR functions and posted
>>>> here about it but got no reply:
>>>>
>>>>
>>>>
>>>> http://mail-archives.apache.org/mod_mbox/apr-dev/201405.mbox/%3CCADED%3DK5ZNs9vm6k_Uy9S%3DpkjjD%3DjTSJR_cE7HJ7M3-LwD-nONw%40mail.gmail.com%3E
>>>
>>>
>>>
>>>
>>> t\filter\both_str_req_proxy.t ...
>>> 1..1
>>> # Running under perl version 5.016003 for MSWin32
>>> # Win32::BuildNumber 1604
>>> # Current time local: Tue Jun  2 09:16:48 2015
>>> # Current time GMT:   Tue Jun  2 16:16:48 2015
>>> # Using Test.pm version 1.26
>>> # Using Apache/Test.pm version 1.39
>>> # testing : lc input and reverse output filters
>>> # Failed test 1 in t\filter\both_str_req_proxy.t at line 18
>>> # expected: 'abcdefghijklmnopqrstuvwxyz0123456789'
>>> # received: '
>>> #
>>> #403 Forbidden
>>> #
>>> #Forbidden
>>> #You don\'t have permission to access
>>> /TestFilter__both_str_req_proxy/foo
>>> # on this server.
>>> #
>>> #
>>> #Apache/2.4.12 (Win32) world domination series/2.0
>>> OpenSSL/1.0.1m
>>> mod_perl/2.0.9-rc2 Perl/
>>> v5.16.3 Server at gls3 Port 8529
>>> #
>>> # '
>>> not ok 1
>>> Failed 1/1 subtests
>>> t\modules\proxy.t ...
>>> # connecting to http://GLS3:8538/TestModules__proxy
>>> 1..1
>>> # Running under perl version 5.016003 for MSWin32
>>> # Win32::BuildNumber 1604
>>> # Current time local: Tue Jun  2 09:30:55 2015
>>> # Current time GMT:   Tue Jun  2 16:30:55 2015
>>> # Using Test.pm version 1.26
>>> # Using Apache/Test.pm version 1.39
>>> request has failed (the response code was: 503)
>>> see t/logs/error_log for more details
>>> Dubious, test returned 255 (wstat 65280, 0xff00)
>>> Failed 1/1 subtests
>>>
>>
>> Thanks. Unfortunately, not much to go on there. I wonder if this is
>> related to your Apache not having all modules loaded? If so, could you
>> send me your httpd.conf so that I can see what's missing and try to
>> reproduce the failures myself.
>
>
> No, fresh build w/ all modules loaded. It seems to me however that APR 1.5.1
> may be different to you. IIRC it released after 2.4.12.
>
> httpd.conf is from the 2.4.12 built against.
> t.httpd.conf is the one that was generated by the test.
>
>
>

No, I've got APR 1.5.1 too (and APR-UTIL 1.5.4) - from the
httpd-2.4.12-deps.tar.gz file.

Is there anything more in t/logs/error_log?

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: Fw: Re: [RELEASE CANDIDATE]: mod_perl-2.0.9 RC2

[Steve Hay]

On 2 June 2015 at 17:18, Dominic Hargreaves  wrote:
> On Tue, Jun 02, 2015 at 03:22:21PM +0200, Vincent Veyron wrote:
>>
>> On Tue, 2 Jun 2015 08:41:49 +0100
>> Steve Hay  wrote:
>> >
>> > Just follow the instructions in INSTALL in the tarball, i.e. something 
>> > like:
>> >
>> >   % perl Makefile.PL MP_APXS=/usr/local/apache2/bin/apxs
>> >   % make && make test
>>
>>
>> hh. hitting a wall, I'm getting errors (listed below). I tried 
>> several things, but can't get 'make test' to work
>>
>> My system runs Debian Stable ( 8/Jessie ), with this command to install 
>> Apache httpd/mod_perl :
>>
>> apt-get install apache2-mpm-prefork libapache2-request-perl 
>> libapache2-modperl2 libapache2-mod-apreq2 apache2.2-common apache2-dev
>>
>> So  presumably, it's running RC1 now (dpkg -l shows : 2.0.9~1624218-2)
>
> This is actually an svn snapshot from a while ago, so not RC1.
>
> If your main concern is whether RC2's test suite works within Debian,
> then the answer is yes:
>
> https://buildd.debian.org/status/fetch.php?pkg=libapache2-mod-perl2&arch=amd64&ver=2.0.9~rc2-1&stamp=1433186697
>
> (at least except with perl 5.22, as I posted about separately).
>

What's the problem with 5.22? I don't see a separate post, except from
David Wheeler, reporting a problem with 5.22. I also found a problem
late in the 5.21 development cycle that I knew would cause a problem
for 5.22: https://rt.cpan.org/Ticket/Display.html?id=101962

Is the problem anything to do with that?

I'm inclinced to just add a note to README about 5.22 woes for now so
that we can get a 2.0.9 out there that works with the latest httpd,
and then work on supporting this latest perl for 2.0.10.

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE]: mod_perl-2.0.9 RC2

[Steve Hay]

On 1 June 2015 at 18:59, Gregg Smith  wrote:
> On 5/30/2015 11:20 AM, Steve Hay wrote:
>>
>> Please download, test, and report back on this release candidate of
>> the long-awaited mod_perl 2.0.9.
>>
>> http://people.apache.org/~stevehay/mod_perl-2.0.9-rc2.tar.gz
>
>
> Server version: Apache/2.4.12 (Win32)
> Server built:   Mar 19 2015 18:23:40
> Server loaded:  APR 1.5.1, APR-UTIL 1.5.4
> Architecture:   32-bit
> Server MPM: WinNT
> All module loaded
>
> Test Summary Report
> ---
> t\filter\both_str_req_proxy.t (Wstat: 0 Tests: 1 Failed: 1)
>   Failed test:  1
> t\modperl\setupenv.t  (Wstat: 0 Tests: 63 Failed: 6)
>   Failed tests:  8, 22, 29, 36, 50, 57
> t\modules\proxy.t (Wstat: 65280 Tests: 0 Failed: 0)
>   Non-zero exit status: 255
>   Parse errors: Bad plan.  You planned 1 tests but ran 0.
> t\preconnection\note.t(Wstat: 0 Tests: 1 Failed: 1)
>   Failed test:  1

I get setupenv.t and note.t both failing on one test machine (Windows
8.1), but not on my other (Windows 7). I haven't looked into those
yet.

Can you send verbose output of the both_str_req_proxy.t and proxy.t
failures? I haven't seen those before.


> t\protocol\echo_block.t   (Wstat: 0 Tests: 3 Failed: 2)
>   Failed tests:  2-3
> t\protocol\echo_nonblock.t(Wstat: 0 Tests: 3 Failed: 1)
>   Failed test:  2
> t\protocol\echo_timeout.t (Wstat: 0 Tests: 5 Failed: 4)
>   Failed tests:  2-5
> t\protocol\pseudo_http.t  (Wstat: 0 Tests: 13 Failed: 9)
>   Failed tests:  3-8, 11-13

These four failures are documented in README as known problems on
Windows with httpd-2.4. I haven't figured out why they fail yet. They
work fine on httpd-2.2.

I wondered if they had something to do with APR functions and posted
here about it but got no reply:

http://mail-archives.apache.org/mod_mbox/apr-dev/201405.mbox/%3CCADED%3DK5ZNs9vm6k_Uy9S%3DpkjjD%3DjTSJR_cE7HJ7M3-LwD-nONw%40mail.gmail.com%3E

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE]: mod_perl-2.0.9 RC2

[Steve Hay]

On 31 May 2015 at 11:07, Vincent Veyron  wrote:
> On Sat, 30 May 2015 19:20:34 +0100
> Steve Hay  wrote:
>
>> Please download, test, and report back on this release candidate of
>> the long-awaited mod_perl 2.0.9.
>>
>
> Hi Steve,
>
> I've been seeing the requests for tests, and meaning to run them for a long 
> while now, but for the life of me, I can't figure out how to install the 
> thing, and it's not for lack of trying.
>
> Is there a place that explains how to install and run the test suite? this 
> page for instance, does not (unless something escapes me totally):
>
> https://perl.apache.org/docs/general/testing/testing.html
>

Just follow the instructions in INSTALL in the tarball, i.e. something like:

  % perl Makefile.PL MP_APXS=/usr/local/apache2/bin/apxs
  % make && make test
  % make install

(You can skip the last step. It's the results of "make test" that
we're interested in here.)

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: [RELEASE CANDIDATE]: mod_perl-2.0.9 RC2

[Steve Hay]

On 31 May 2015 at 02:12, Kevin A. McGrail  wrote:
> ok 23
> # testing : $s->log_serror(LOG_MARK, LOG_DEBUG, APR::Const::EGENERAL...)
> # expected: qr/(?^:Internal error: log_serror test 2)/
> # received: '
> # *** The following warn entry is expected and harmless ***
> # [Sat May 30 21:04:59.085379 2015] [perl:debug] [pid 9642:tid 3076]
> aplog.pm(90): (20014)Internal error (specific information not available):
> log_serror test 2
> # '
> not ok 24
> # testing : $r->log_rerror(LOG_MARK, LOG_CRIT, APR::Const::ENOTIME...)
> # expected: qr/(?^:\[\w*:crit\] \[pid[^]]+\] .*?: \[[^]]+\] log_rerror
> test)/
> # received: '# Failed test 24 in
> /usr/src/mod_perl-2.0.9-rc2/t/response/TestAPI/aplog.pm at line 92
> #
> # *** The following error entry is expected and harmless ***
> # [Sat May 30 21:04:59.086325 2015] [perl:crit] [pid 9642:tid 3076]
> (20007)No time was provided and one was required.: [client 127.0.0.1:56778]
> log_rerror test
> # '

Weird. I don't see this on my system, although I see that many other
people are reporting the same problem.

I really don't know why the test fails given that what it says it
received DOES match what it says it's expecting!?

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

[RELEASE CANDIDATE]: mod_perl-2.0.9 RC2

[Steve Hay]

Please download, test, and report back on this release candidate of
the long-awaited mod_perl 2.0.9.

http://people.apache.org/~stevehay/mod_perl-2.0.9-rc2.tar.gz

MD5 = 6dbf61e0724a245f1c015687fb3e4213
SHA1 = faf63c98b338ef0373ccbbe775c52506270ff3b9

Changes since RC1:

Remove PerlInterpScope. This has not been working properly with
threaded MPMs with httpd-2.4.x and the use-case of this directive was
questionable. [Jan Kaluza]

Allow running the test suite with httpd-2.4.x when mod_access_compat
is not loaded. [Steve Hay]

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: PerlInterpScope removed, the compatibility with threaded MPMs fixed

[Steve Hay]

On 29 May 2015 at 06:45, Jan Kaluža  wrote:
> Hi,
>
> as discussed in the thread about commit r1676417, I've removed
> PerlInterpScope directive. The reasoning is described in the
> http://svn.apache.org/r1682369 commit message, but I will paste it here too:
>
> ~~~
> Remove PerlInterpScope directive as discussed on mailing list.
> PerlInterpScope did not work correctly with httpd-2.4.x threaded MPMs.
> For example when using 'PerlInterpScope connection' and multiple
> requests are handled using the single connection, every request tries
> to fetch new interpreter. This leads to deadlock soon, because we can always
> have more requests than interpreters.
>
> There are more situations like the one described above and the fix is quite
> hard with the way how the usage of PerlInterpeters is designed currently.
> ~~~
>
> This and other changes I've done in trunk today should fix the
> modperl-2.0.9-RC1 freeze seen with threaded MPMs on httpd-2.4.x. Could you
> please test it?
>
> If things will look OK, maybe we could release RC2 to let more people to
> test it.
>

Fantastic! Thanks for this. I will test it this evening/tomorrow and
roll out an RC2 after that if all is looking well.

-
To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
For additional commands, e-mail: dev-h...@perl.apache.org

Re: svn commit: r1676417 - /perl/modperl/trunk/src/modules/perl/modperl_interp.c

[Steve Hay]

On 25 May 2015 10:11, "Jan Kaluža"  wrote:
>
> On 05/24/2015 04:50 PM, Steve Hay wrote:
>>
>> On 20 May 2015 at 14:51, Jan Kaluža  wrote:
>>>
>>> On 05/20/2015 11:49 AM, Jan Kaluža wrote:
>>>>
>>>>
>>>> On 05/15/2015 12:26 PM, Jan Kaluža wrote:
>>>>>
>>>>>
>>>>> On 05/15/2015 11:57 AM, Jan Kaluža wrote:
>>>>>>
>>>>>>
>>>>>> On 05/15/2015 10:01 AM, Steve Hay wrote:
>>>>>>>
>>>>>>>
>>>>>>> On 15 May 2015 at 08:56, Steve Hay 
wrote:
>>>>>>>>
>>>>>>>>
>>>>>>>> On 15 May 2015 at 07:14, Jan Kaluža  wrote:
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> On 05/14/2015 07:42 PM, Steve Hay wrote:
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> On 14 May 2015 at 12:48, Jan Kaluža  wrote:
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> On 05/14/2015 11:24 AM, Niko Tyni wrote:
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>> On Sun, May 10, 2015 at 01:47:19PM +0100, Steve Hay wrote:
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>> On 28 April 2015 at 07:51,   wrote:
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>
>>>>>>>>>>>>>> Author: jkaluza
>>>>>>>>>>>>>> Date: Tue Apr 28 06:51:12 2015
>>>>>>>>>>>>>> New Revision: 1676417
>>>>>>>>>>>>>>
>>>>>>>>>>>>>> URL: http://svn.apache.org/r1676417
>>>>>>>>>>>>>> Log:
>>>>>>>>>>>>>> Initialize interp->refcnt to 1 in modperl_interp_select.
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>> I cannot understand why, but since this patch was applied I
find
>>>>>>>>>>>>> that
>>>>>>>>>>>>> t\modules\proxy.t fails every time when I run the full "nmake
>>>>>>>>>>>>> test",
>>>>>>>>>>>>> but it always succeeds when I run it in isolation so I'm at a
>>>>>>>>>>>>> loss to
>>>>>>>>>>>>> find out what is going wrong. All other tests (apart from
those
>>>>>>>>>>>>> known
>>>>>>>>>>>>> Win32-specific failures documented in README) still pass.
>>>>>>>>>>>>> Reverting
>>>>>>>>>>>>> the patch "fixes" the proxy.t problem, but probably isn't the
>>>>>>>>>>>>> right
>>>>>>>>>>>>> solution.
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> It's caused by Perl_croak/modperl_croak.
>>>>>>>>>>>
>>>>>>>>>>> Lets take modperl_run_filter as an example. When following
>>>>>>>>>>> code-path is
>>>>>>>>>>> executed ...
>>>>>>>>>>>
>>>>>>>>>>>modperl_croak(aTHX_ MODPERL_FILTER_ERROR,
>>>>>>>>>>>  

Re: [RELEASE CANDIDATE]: mod_perl-2.0.9 RC1

[Steve Hay]

On 27 May 2015 21:50, "Gregg Smith"  wrote:
>
> Hi,
>
> Apache 2.4.12 Perl 5.16.2 VC++ 2008
>
> Build went fine but 'nmake test' gives me;
>
> C:\usr\bin\perl.exe -Iblib\arch -Iblib\lib  t/TEST -clean
>   C:\usr\bin\perl.exe -Iblib\arch -Iblib\lib  t/TEST -bugreport
-verbose=0
> C:\Apache24\bin\httpd.exe  -d C:/build4/mod_perl-2.0.9-rc1/t -f
C:/build4/mod_perl-2.0.9-rc1/t/conf/
> httpd.conf -D APACHE2 -D APACHE2_4 -D PERL_USEITHREADS
> using Apache/2.4.12 (winnt MPM)
>
> waiting 300 seconds for server to start: ...[Wed May 27 13:25:18.706600
2015] [env:warn] [pid 3524:t
> id 292] AH01506: PassEnv variable LD_LIBRARY_PATH was undefined
> AH00526: Syntax error on line 2006 of
C:/build4/mod_perl-2.0.9-rc1/t/conf/httpd.conf:
> Invalid command 'Order', perhaps misspelled or defined by a module not
included in the server config
> uration
>
> 'Order' ... that would be alright if mod_access_compat.so would also get
loaded, but obviously it doesn't.
>
> I also see this is the generated httpd.conf
> 
>  2.4.1>
>
> That would be fine if it would load mod_version, but doesn't.
>
> I'm sure this is somehow my fault.  Steve, how did you test this on
Apache 2.4 Win32?

Is it a new, clean build of Apache? I think the generated httpd.conf is
influenced by what's in the httpd.conf of the Apache you build against...

I used a new Apache and all was well with 2.4. With 2.2 I had to edit the
httpd.conf of my Apache installation (even though it was a fresh build
again) in order to enable all the LoadModule lines, otherwise something was
missing since not everything is enabled by default with 2.2 even though all
modules are built ...

Coming to think of it, it might have been mod_version that was the problem,
but a new build of 2.4 does have it enabled already (at least, when
building with all modules enabled).

Steve

>
> Regards,
>
> Gregg
>
>
>
>
>
> -
> To unsubscribe, e-mail: dev-unsubscr...@perl.apache.org
> For additional commands, e-mail: dev-h...@perl.apache.org
>

Re: svn commit: r1676417 - /perl/modperl/trunk/src/modules/perl/modperl_interp.c

[Steve Hay]

On 20 May 2015 at 14:51, Jan Kaluža  wrote:
> On 05/20/2015 11:49 AM, Jan Kaluža wrote:
>>
>> On 05/15/2015 12:26 PM, Jan Kaluža wrote:
>>>
>>> On 05/15/2015 11:57 AM, Jan Kaluža wrote:
>>>>
>>>> On 05/15/2015 10:01 AM, Steve Hay wrote:
>>>>>
>>>>> On 15 May 2015 at 08:56, Steve Hay  wrote:
>>>>>>
>>>>>> On 15 May 2015 at 07:14, Jan Kaluža  wrote:
>>>>>>>
>>>>>>> On 05/14/2015 07:42 PM, Steve Hay wrote:
>>>>>>>>
>>>>>>>>
>>>>>>>> On 14 May 2015 at 12:48, Jan Kaluža  wrote:
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> On 05/14/2015 11:24 AM, Niko Tyni wrote:
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> On Sun, May 10, 2015 at 01:47:19PM +0100, Steve Hay wrote:
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> On 28 April 2015 at 07:51,   wrote:
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>> Author: jkaluza
>>>>>>>>>>>> Date: Tue Apr 28 06:51:12 2015
>>>>>>>>>>>> New Revision: 1676417
>>>>>>>>>>>>
>>>>>>>>>>>> URL: http://svn.apache.org/r1676417
>>>>>>>>>>>> Log:
>>>>>>>>>>>> Initialize interp->refcnt to 1 in modperl_interp_select.
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>> I cannot understand why, but since this patch was applied I find
>>>>>>>>>>> that
>>>>>>>>>>> t\modules\proxy.t fails every time when I run the full "nmake
>>>>>>>>>>> test",
>>>>>>>>>>> but it always succeeds when I run it in isolation so I'm at a
>>>>>>>>>>> loss to
>>>>>>>>>>> find out what is going wrong. All other tests (apart from those
>>>>>>>>>>> known
>>>>>>>>>>> Win32-specific failures documented in README) still pass.
>>>>>>>>>>> Reverting
>>>>>>>>>>> the patch "fixes" the proxy.t problem, but probably isn't the
>>>>>>>>>>> right
>>>>>>>>>>> solution.
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> It's caused by Perl_croak/modperl_croak.
>>>>>>>>>
>>>>>>>>> Lets take modperl_run_filter as an example. When following
>>>>>>>>> code-path is
>>>>>>>>> executed ...
>>>>>>>>>
>>>>>>>>>   modperl_croak(aTHX_ MODPERL_FILTER_ERROR,
>>>>>>>>> "a filter calling $f->read "
>>>>>>>>> "must return OK and not DECLINED");
>>>>>>>>>
>>>>>>>>> ... the MP_INTERP_PUTBACK is not reached for some reason (I
>>>>>>>>> presume it's
>>>>>>>>> because of Perl_croak, but I don't understand why it stops the
>>>>>>>>> execution
>>>>>>>>> of
>>>>>>>>> the rest of modperl_run_filter method).
>>>>>>>>>
>>>>>>>>> Because of that, the interp->refcnt is not decreased, and the
>>>>>>>>> interp is
>>>>>>>>> not
>>>>>>>>> freed.
>>>>>>>>>
>>>>>>>>> I has been able to "fix" it by attached patch, but I would like to
>>>>>>>>> discuss
>>>>>>>>> more generic way how to fix that problem.