from:"perl"

When code is updated during developing period, httpd progress must be
reload or restart so that to see the change. If is it possible to see
the change without reload httpd?

--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: reload or restart httpd

Yes, I am using Apache2, the doc is very useful.
in httpd.conf file, I use PerlInitHandler example.pm's handler() as
starting point, when I update the code in handler(), I didn't see the
updates, so I asked previous question.
After I read the head of the doc, I make sense that I need to modify
the global %INC after server is started.
I am going to do like this
sub handler {
my $r = shift;
my $s = Apache2::ServerUtil->server;
my $WRoot = $s->dir_config('WRoot');
use lib "$FindBin::Bin/$WRoot";
}

Am I correct?

On 2/6/06, Sean Davis <[EMAIL PROTECTED]> wrote:
>
> On 2/6/06 7:10 AM, "Ken Perl" <[EMAIL PROTECTED]> wrote:
>
> > When code is updated during developing period, httpd progress must be
> > reload or restart so that to see the change. If is it possible to see
> > the change without reload httpd?
>
>
> You didn't say which version of mod_perl you are using, but try looking
> here:
>
> <http://search.cpan.org/~pgollucci/mod_perl-2.0.2/docs/api/Apache2/Reload.po
> d>
>
> Sean
>
>
>

--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

interactive debuging modperl2 using Apache::DB

Hi,
I configured Apache::DB in my modperl2 and got a notice message in
apache2 startup error.log,
[notice] Apache::DB initialized in child 1123

I think I have setuped it successfully.
And I also see the debugger welcome message,
Loading DB routines from perl5db.pl version 1.28
Editor support available.

Enter h or `h h' for help, or `man perldebug' for more help.

But I don't know to interactive with modperl to debug my program, I
can only see  DB<1>  prompt symbol in the log file.

--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

timestamp of Apache::DB

when Apache::DB is initialized it will write a line of log into
error.log, but without a timestamp,
[notice] Apache::DB initialized in child 1957
May I add one?

--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

[MP2] set breakpoint

hi,
I'd like to set a breakpoint at the first line in the module's handler,
sub handler {
my $r = shift;   <-- here
 ...  ...
   $r->set_handlers(PerlResponseHandler => \&contentHandler);
}

So, I tried PerFixupHandler, PerlHandler  and PerlInitHander
respectively in below Location directive, but all didn't stop at the
point I wanted, instead, It always stopped at  the line of
'&contentHandler',

PerlInitHandler +Apache::DB
    

what I did wrong?

--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: [MP2] set breakpoint

No IDE for me too, I am using vi on a Debian terminal too.
warn is easy and good, I also want a more common and more powerful
interactive debugger tool, now I am trying Apache::DB and have this
trouble.

On 2/7/06, John ORourke <[EMAIL PROTECTED]> wrote:
> What sort of IDE, debugger, and platform are you using?
>
> Personally for debugging handlers I find it easier to just stick a line
> like: warn "got here, r=$r\n"
>
> then just tail -f the error_log.  Proper old school.  Then again I
> develop with vi on an 80x24 terminal, am I missing out on something?
>
> John
>
> Ken Perl wrote:
>
> >hi,
> >I'd like to set a breakpoint at the first line in the module's handler,
> >sub handler {
> >my $r = shift;   <-- here
> > ...  ...
> >   $r->set_handlers(PerlResponseHandler => \&contentHandler);
> >}
> >
> >So, I tried PerFixupHandler, PerlHandler  and PerlInitHander
> >respectively in below Location directive, but all didn't stop at the
> >point I wanted, instead, It always stopped at  the line of
> >'&contentHandler',
> >
> >    PerlInitHandler +Apache::DB
> >
> >
> >what I did wrong?
> >
> >--
> >perl -e 'print unpack(u,"62V5N\"FME;G\!E >")'
> >
> >
> >
>
>


--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: timestamp of Apache::DB

Cool, I guess you mean add these lines into the conf file,


use CGI::Carp;


Thanks, It works.

On 2/7/06, John Doe <[EMAIL PROTECTED]> wrote:
> Ken Perl am Dienstag, 7. Februar 2006 07.44:
> > when Apache::DB is initialized it will write a line of log into
> > error.log, but without a timestamp,
> > [notice] Apache::DB initialized in child 1957
> > May I add one?
> >
> > --
> > perl -e 'print unpack(u,"62V5N\"FME;G\!E > ")'
>
> yes, by putting the line
>
> use CGI::Carp
>
> into the httpd.conf file :-)
>
> hth,
> joe
>


--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: reload or restart httpd

tried adding the two lines to the conf file, didn't work.

On 2/6/06, Sean Davis <[EMAIL PROTECTED]> wrote:
>
>
>
> On 2/6/06 9:06 AM, "Ken Perl" <[EMAIL PROTECTED]> wrote:
>
> > Yes, I am using Apache2, the doc is very useful.
> > in httpd.conf file, I use PerlInitHandler example.pm's handler() as
> > starting point, when I update the code in handler(), I didn't see the
> > updates, so I asked previous question.
> > After I read the head of the doc, I make sense that I need to modify
> > the global %INC after server is started.
> > I am going to do like this
> > sub handler {
> > my $r = shift;
> > my $s = Apache2::ServerUtil->server;
> > my $WRoot = $s->dir_config('WRoot');
> > use lib "$FindBin::Bin/$WRoot";
> > }
>
> Unless I am misunderstanding what you are trying to do, I think you just
> need to add these two lines to your httpd.conf to have any changed modules
> reloaded:
>
>  PerlModule Apache2::Reload
>  PerlInitHandler Apache2::Reload
>
> I don't think you have to do anything special with %INC.
>
> Sean
>
>
>


--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: reload or restart httpd

I also turned the debug on, but didn't see any debug message in the
error.log, anyone knows why?

 PerlSetVar ReloadDebug On


On 2/7/06, Ken Perl <[EMAIL PROTECTED]> wrote:
> tried adding the two lines to the conf file, didn't work.
>
> On 2/6/06, Sean Davis <[EMAIL PROTECTED]> wrote:
> >
> >
> >
> > On 2/6/06 9:06 AM, "Ken Perl" <[EMAIL PROTECTED]> wrote:
> >
> > > Yes, I am using Apache2, the doc is very useful.
> > > in httpd.conf file, I use PerlInitHandler example.pm's handler() as
> > > starting point, when I update the code in handler(), I didn't see the
> > > updates, so I asked previous question.
> > > After I read the head of the doc, I make sense that I need to modify
> > > the global %INC after server is started.
> > > I am going to do like this
> > > sub handler {
> > > my $r = shift;
> > > my $s = Apache2::ServerUtil->server;
> > > my $WRoot = $s->dir_config('WRoot');
> > > use lib "$FindBin::Bin/$WRoot";
> > > }
> >
> > Unless I am misunderstanding what you are trying to do, I think you just
> > need to add these two lines to your httpd.conf to have any changed modules
> > reloaded:
> >
> >  PerlModule Apache2::Reload
> >  PerlInitHandler Apache2::Reload
> >
> > I don't think you have to do anything special with %INC.
> >
> > Sean
> >
> >
> >
>
>
> --
> perl -e 'print unpack(u,"62V5N\"FME;G\!E ")'
>


--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: [MP2] set breakpoint

You catched my problem, I didn't load debugger before this handler, thanks.
After I changed the order in the httpd config file, my problem is
solved,  It stopped at the point I wanted,

 my $r = shift;

Thanks again.

On 2/8/06, Perrin Harkins <[EMAIL PROTECTED]> wrote:
> On Tue, 2006-02-07 at 16:10 +0800, Ken Perl wrote:
> > hi,
> > I'd like to set a breakpoint at the first line in the module's handler,
> > sub handler {
> > my $r = shift;   <-- here
> >  ...  ...
> >$r->set_handlers(PerlResponseHandler => \&contentHandler);
> > }
> >
> > So, I tried PerFixupHandler, PerlHandler  and PerlInitHander
> > respectively in below Location directive, but all didn't stop at the
> > point I wanted, instead, It always stopped at  the line of
> > '&contentHandler',
> > 
> > PerlInitHandler +Apache::DB
> > 
> >
> > what I did wrong?
>
> Did you load this handler before you loaded the debugger?  You should
> call Apache::DB->init() explicitly before loading any of the code you
> want to debug, or it won't get debugging symbols.  See the documentation
> for Apache::DB.
>
> - Perrin
>
>


--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

module bug?

I see a info in the error.log when I access the uri / and works fine,
[Wed Feb 08 13:22:35 2006] [info] [client 10.0.0.1] Module bug? 
Request filename is missing for URI /

Is the info reported by modperl2? anything wrong?

--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: module bug?

Sounds reasonable.
Based on your anaIysis,  I added

   warn "reach last line of WebGUI.pm";

into the handler before the last line.

sub handler{
... ...
else {
$r->set_handlers(PerlResponseHandler => \&contentHandler);
$r->set_handlers(PerlTransHandler => sub { return Apache2::Const
::OK});
}
warn "reach last line of WebGUI.pm";
return Apache2::Const::DECLINED;
}

then I got these two lines of log message like this,
[Wed Feb  8 20:45:57 2006] -e: got last line of WebGUI.pm at
/data/WebGUI/lib/WebGUI.pm line 65.
[Wed Feb 08 20:45:57 2006] [info] [client 10.0.0.1] Module bug? 
Request filename is missing for URI /

You may notice that the "Module bug?" message written out after the
debug message I added, so I think, the info is not caused by
PerlTransHander. True?


On 2/8/06, Torsten Foertsch <[EMAIL PROTECTED]> wrote:
> On Wednesday 08 February 2006 07:13, Ken Perl wrote:
> > I see a info in the error.log when I access the uri / and works fine,
> > [Wed Feb 08 13:22:35 2006] [info] [client 10.0.0.1] Module bug?
> > Request filename is missing for URI /
> >
> > Is the info reported by modperl2? anything wrong?
>
> You have written a translation handler that doesn't set $r->filename and
> returns OK, right?
>
> The info you are seeing is generated by the core-maptostorage handler. You can
> avoid it either
>
> a) by setting $r->filename prior to maptostorage or
>
> b) by returning DECLINED from your translation handler and thus letting the
> core translation handler figure the filename out or
>
> c) by inserting a PerlMapToStorage handler that returns OK and thus skipping
> the core-maptostorage handler.
>
> Torsten
>
>
>


--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: timestamp of Apache::DB

That helps already, but need to add the statement into the 
directory and works fine!

On 2/8/06, John Doe <[EMAIL PROTECTED]> wrote:
> Perrin Harkins am Dienstag, 7. Februar 2006 18.45:
> > On Tue, 2006-02-07 at 08:56 +0100, John Doe wrote:
> > > Ken Perl am Dienstag, 7. Februar 2006 07.44:
> > > > when Apache::DB is initialized it will write a line of log into
> > > > error.log, but without a timestamp,
> > > > [notice] Apache::DB initialized in child 1957
> > > > May I add one?
> > >
> > > yes, by putting the line
> > >
> > > use CGI::Carp
> > >
> > > into the httpd.conf file :-)
> >
> > Huh?  How is that going to help?
>
> It doesn't help in this early startup phase; I misread the question. :-(
>


--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: module bug?

And I found just now, I got the info message before the root / page is returned.

On 2/8/06, Ken Perl <[EMAIL PROTECTED]> wrote:
> Sounds reasonable.
> Based on your anaIysis,  I added
>
>   warn "reach last line of WebGUI.pm";
>
> into the handler before the last line.
>
> sub handler{
>... ...
>else {
>$r->set_handlers(PerlResponseHandler => \&contentHandler);
>$r->set_handlers(PerlTransHandler => sub { return 
> Apache2::Const
> ::OK});
>}
>warn "reach last line of WebGUI.pm";
>return Apache2::Const::DECLINED;
> }
>
> then I got these two lines of log message like this,
> [Wed Feb  8 20:45:57 2006] -e: got last line of WebGUI.pm at
> /data/WebGUI/lib/WebGUI.pm line 65.
> [Wed Feb 08 20:45:57 2006] [info] [client 10.0.0.1] Module bug?
> Request filename is missing for URI /
>
> You may notice that the "Module bug?" message written out after the
> debug message I added, so I think, the info is not caused by
> PerlTransHander. True?
>
>
> On 2/8/06, Torsten Foertsch <[EMAIL PROTECTED]> wrote:
> > On Wednesday 08 February 2006 07:13, Ken Perl wrote:
> > > I see a info in the error.log when I access the uri / and works fine,
> > > [Wed Feb 08 13:22:35 2006] [info] [client 10.0.0.1] Module bug?
> > > Request filename is missing for URI /
> > >
> > > Is the info reported by modperl2? anything wrong?
> >
> > You have written a translation handler that doesn't set $r->filename and
> > returns OK, right?
> >
> > The info you are seeing is generated by the core-maptostorage handler. You 
> > can
> > avoid it either
> >
> > a) by setting $r->filename prior to maptostorage or
> >
> > b) by returning DECLINED from your translation handler and thus letting the
> > core translation handler figure the filename out or
> >
> > c) by inserting a PerlMapToStorage handler that returns OK and thus skipping
> > the core-maptostorage handler.
> >
> > Torsten
> >
> >
> >
>
>
> --
> perl -e 'print unpack(u,"62V5N\"FME;G\!E ")'
>


--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

the modperl book link is broken

Hi,
The page 
http://perl.apache.org/docs/offsite/books.html#The_mod_perl_Developer_s_Cookbook
introduces two books, their links http://modperlbook.org/ and
http://www.modperlcookbook.org/ are both broken now, Does anyone have
these two ebooks?

--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: the modperl book link is broken

I still can't connect the site, : -(

I prefer e-book and appreciate that if anyone who have those two books
could send me the e-book copy.

On 2/8/06, Perrin Harkins <[EMAIL PROTECTED]> wrote:
> Ken Perl wrote:
> > The page 
> > http://perl.apache.org/docs/offsite/books.html#The_mod_perl_Developer_s_Cookbook
> > introduces two books, their links http://modperlbook.org/ and
> > http://www.modperlcookbook.org/ are both broken now
>
> Those links both work fine for me.  Maybe there's a network problem on
> your end.  If you network is too unreliable to read them on-line, you
> can purchase both of these books in print.
>
> - Perrin
>


--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Apache2::SizeLimit with non-threaded MPMs

I know Apache2:SizeLimit works ony with non-threaded MPMs, but I don't
know how to fix the issue, any comments?

[Thu Feb 09 13:46:21 2006] [error] [client 221.219.202.148] failed to
resolve handler `Apache2::SizeLimit': Apache2::SizeLimit at the moment
works only with non-threaded MPMs at /usr/share/perl/5.8/CGI/Carp.pm
line 314.\nBEGIN failed--compilation aborted at
/usr/lib/perl5/Apache2/SizeLimit.pm line 97.\nCompilation failed in
require at (eval 913) line 3.\n

--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

debugger command always return IO::Handle::DESTROY

2006-02-09 Thread Ken Perl

I am using Apache::DB as debugger to debug my modperl2 program. No
breakpoints are defined, I input many 'r' and the return is always
like below, and the home page can't be displayed in the debug mode
while turning off debug mode it home page can be rendered as normal.

  DB<1> r
void context return from IO::Handle::DESTROY
IO::Handle::DESTROY(/usr/lib/perl/5.8/IO/Handle.pm:75):
75: sub DESTROY {}
  DB<1> r
void context return from IO::Handle::DESTROY
Apache2::SizeLimit::handler(/usr/lib/perl5/Apache2/SizeLimit.pm:248):
248:my $r = shift;

Anybody knows about the problem in it?

--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

detecting server start, stop, graceful in startup.pl

2006-02-11 Thread Ken Perl

>From the Apache2::ServerUtil API doc, to do something only when server
restarts (httpd -k start or httpd -k graceful), check whether
restart_count() is bigger than 1:

  my $cnt = Apache2::ServerUtil::restart_count();
  do_something() if $cnt > 1;

However, the API doesn't work as above. I tried to add some debug code
to my startup.pl script and tried all the command options of
apache2ctl, start, stop, graceful, the $cnt is always equal 1.
I also tried to write the value $cnt to a /tmp/out file just like in
the API doc and got same result, that is
 % apache2ctl -k start
  cnt: 1
  cnt: 2

  % apache2ctl -k graceful
  cnt: 1
  cnt: 3

  % apache2ctl -k graceful
  cnt: 1
  cnt: 4

  % apache2ctl -k stop
  cnt: 1

But, I can't detecting start, stop, graceful in startup.pl coding just
like in the API doc, Could you give me some comments? thanks.

--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: debugger command always return IO::Handle::DESTROY

And I also can't quit Apache::DB by entering many 'q' commands,
  DB<5> q
IO::Handle::DESTROY(/usr/lib/perl/5.8/IO/Handle.pm:75):
75: sub DESTROY {}
  DB<5> q
IO::Handle::DESTROY(/usr/lib/perl/5.8/IO/Handle.pm:75):
75: sub DESTROY {}
  DB<5> q
IO::Handle::DESTROY(/usr/lib/perl/5.8/IO/Handle.pm:75):
75: sub DESTROY {}
  DB<5> q

the last 'q' command hang there, any comments?


On 2/9/06, Ken Perl <[EMAIL PROTECTED]> wrote:
> I am using Apache::DB as debugger to debug my modperl2 program. No
> breakpoints are defined, I input many 'r' and the return is always
> like below, and the home page can't be displayed in the debug mode
> while turning off debug mode it home page can be rendered as normal.
>
>  DB<1> r
> void context return from IO::Handle::DESTROY
> IO::Handle::DESTROY(/usr/lib/perl/5.8/IO/Handle.pm:75):
> 75: sub DESTROY {}
>  DB<1> r
> void context return from IO::Handle::DESTROY
> Apache2::SizeLimit::handler(/usr/lib/perl5/Apache2/SizeLimit.pm:248):
> 248:    my $r = shift;
>
> Anybody knows about the problem in it?
>
> --
> perl -e 'print unpack(u,"62V5N\"FME;G\!E ")'
>


--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: detecting server start, stop, graceful in startup.pl

No, I want to do something in the startup.pl script.
When apache start, I also want to say my modperl program is starting
and started, when stop, I want to say my modperl program is stopping
and stopped. When graceful, I want to say reloading and reloaded.
But I didn't find a API could help, any ideas?

On 2/12/06, Frank Wiles <[EMAIL PROTECTED]> wrote:
> On Sun, 12 Feb 2006 15:02:38 +0800
> Ken Perl <[EMAIL PROTECTED]> wrote:
>
> > >From the Apache2::ServerUtil API doc, to do something only when
> > >server
> > restarts (httpd -k start or httpd -k graceful), check whether
> > restart_count() is bigger than 1:
> >
> >   my $cnt = Apache2::ServerUtil::restart_count();
> >   do_something() if $cnt > 1;
> >
> > However, the API doesn't work as above.
> >
> > [...snip...]
> >
> > But, I can't detecting start, stop, graceful in startup.pl coding just
> > like in the API doc, Could you give me some comments? thanks.
>
>  I think you are misreading what the docs are saying.
>  Apache2::ServerUtil::restart_count() is just that a count of the
>  number of times the server was restarted.  So a start will show a
>  count of two and gracefuls would be greater than two.
>
>  The server restarts itself at least once during start up and you
>  typically do not want to do anything during that initial restart.
>
>  -
>   Frank Wiles <[EMAIL PROTECTED]>
>   http://www.wiles.org
>  -
>
>


--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: detecting server start, stop, graceful in startup.pl

I add your code to my startup.pl, but it doesn't work.

# apache2ctl -k start
Program is starting
Stopping...

#apache2ctl -k graceful


# apache2ctl -k stop


On 2/13/06, Frank Wiles <[EMAIL PROTECTED]> wrote:
> On Mon, 13 Feb 2006 09:17:57 +0800
> Ken Perl <[EMAIL PROTECTED]> wrote:
>
> > No, I want to do something in the startup.pl script.
> > When apache start, I also want to say my modperl program is starting
> > and started, when stop, I want to say my modperl program is stopping
> > and stopped. When graceful, I want to say reloading and reloaded.
> > But I didn't find a API could help, any ideas?
>
>  Hi Ken,
>
>  I haven't tested this, but it should work. In your startup.pl do
>  something like this:
>
>  my $started = 0;
>  my $cnt = Apache2::ServerUtil::restart_count();
>
>  if( !$started and $cnt == 1 ) {
>   $started = 1;
>   print "Program is starting\n";
>  }
>
>  if( $cnt > 2 ) {
>   print "Restarting";
>  }
>
>  if( $started and $cnt == 1 ) {
>   print "Stopping...";
>  }
>
>  -
>   Frank Wiles <[EMAIL PROTECTED]>
>   http://www.wiles.org
>  -
>
>


--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: debugger command always return IO::Handle::DESTROY

I tried to stop apache by issuing apache2ctl -k stop in another
terminal, the command exit without any output, but the debugger is
still hung there. I have to send signal KILL to the process.

On 2/12/06, Frank Wiles <[EMAIL PROTECTED]> wrote:
> On Sun, 12 Feb 2006 18:55:13 +0800
> Ken Perl <[EMAIL PROTECTED]> wrote:
>
> > And I also can't quit Apache::DB by entering many 'q' commands,
> >   DB<5> q
> > IO::Handle::DESTROY(/usr/lib/perl/5.8/IO/Handle.pm:75):
> > 75: sub DESTROY {}
> >   DB<5> q
> > IO::Handle::DESTROY(/usr/lib/perl/5.8/IO/Handle.pm:75):
> > 75: sub DESTROY {}
> >   DB<5> q
> > IO::Handle::DESTROY(/usr/lib/perl/5.8/IO/Handle.pm:75):
> > 75: sub DESTROY {}
> >   DB<5> q
> >
> > the last 'q' command hang there, any comments?
>
>  I believe you have to stop Apache to exit the debugger when using it.
>  If that doesn't fix it let me know.
>
>  -
>   Frank Wiles <[EMAIL PROTECTED]>
>   http://www.wiles.org
>  -
>
>


--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: debugger command always return IO::Handle::DESTROY

Normaly, ctrl +c works for a simple modperl program, I am trying to
debug a more complex modperl program.

I'd like to provide all info you need, maybe I need some guide on what's needed.

What I want to do is stepping into the modperl2 WebGUI
program(https://svn.webgui.org/svnroot/WebGUI/ and fix some bugs.

When I request the home page, I get the prompt of the debugger, then I
input command 'r' to let debugger return from the contentHandler(),
and input many other 'r' to return too, I just want to see if I can
see the home page under debugging mode, but I can't, only lots of
IO::Handle::DESTROY, and finally, hang there,
I can't input any more debugger command, kill the apache2 process is
the only thing left could be done.
And on the browser side, I see from the status bar it is still waiting
to dowload the .gif images, however, if trun the debugging mode off,
home page can be displayed smoothly.

In my httpd.conf, I enabled Apache2::Reload and Apache2::Status.

Anything I missed or anything you want to know, please let me know.

On 2/13/06, Perrin Harkins <[EMAIL PROTECTED]> wrote:
> Ken Perl wrote:
> > I tried to stop apache by issuing apache2ctl -k stop in another
> > terminal, the command exit without any output, but the debugger is
> > still hung there. I have to send signal KILL to the process.
>
> That will work, but the usual way to stop is to hit Ctrl-C in the
> debugger terminal.  The apache2ctl script probably fails because you are
> running with a single process so the PID file isn't there or isn't accurate.
>
> I haven't seen your IO::Handle issue before, but I suspect the debugger
> is just informing you about some code that ran after your return.  The
> real issue is why your home page doesn't display as you expect.  To help
> you with that, you need to give us more details about what happens and
> what was supposed to happen.
>
> - Perrin
>

--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: debugger command always return IO::Handle::DESTROY

2006-02-13 Thread Ken Perl

Yes, my browser gets most of the home page back if I input lots of
'r', except some images are still waiting for downloading.
If starting server with -X only the page could be retrieved complete
successfully.


On 2/14/06, Perrin Harkins <[EMAIL PROTECTED]> wrote:
> On Mon, 2006-02-13 at 14:27 +0800, Ken Perl wrote:
> > Normaly, ctrl +c works for a simple modperl program, I am trying to
> > debug a more complex modperl program.
>
> I don't see how that would a make difference.  You're just trying to
> send a signal to the httpd process by hitting Ctrl-C.
>
> > I'd like to provide all info you need, maybe I need some guide on what's 
> > needed.
>
> Your previous message said that it didn't work, but didn't say what
> would have happened if it did work, or what it actually did.  The
> information you gave here is better.
>
> > When I request the home page, I get the prompt of the debugger
>
> That's good.
>
> > then I
> > input command 'r' to let debugger return from the contentHandler(),
> > and input many other 'r' to return too, I just want to see if I can
> > see the home page under debugging mode, but I can't
>
> Your browser never gets anything back?  Try using telnet or lwp-get to
> test it, so you can see if anything gets sent back.  You don't need to
> test the images.
>
> > only lots of
> > IO::Handle::DESTROY, and finally, hang there,
> > I can't input any more debugger command, kill the apache2 process is
> > the only thing left could be done.
> > And on the browser side, I see from the status bar it is still waiting
> > to dowload the .gif images, however, if trun the debugging mode off,
> > home page can be displayed smoothly.
>
> What happens if you run the server with -X but don't add the debugger?
> Can you get the page then?
>
> - Perrin
>
>


--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: detecting server start, stop, graceful in startup.pl

2006-02-17 Thread Ken Perl

Better than before, but only 'start' works.

debian:~/WebGUI# apache2ctl -k start
[Fri Feb 17 22:34:21 2006] preload.perl: Program is starting
Loading WebGUI.conf
debian:~/WebGUI# apache2ctl -k graceful
[Fri Feb 17 22:34:55 2006] preload.perl: Program is starting
Loading WebGUI.conf
[Fri Feb 17 22:34:57 2006] -e: I am shutting down at
/root/WebGUI/sbin/preload.perl line 33.
debian:~/WebGUI# apache2ctl -k stop
[Fri Feb 17 22:35:28 2006] preload.perl: Program is starting
Loading WebGUI.conf
[Fri Feb 17 22:35:29 2006] -e: I am shutting down at
/root/WebGUI/sbin/preload.perl line 33.

On 2/17/06, Philippe M. Chiasson <[EMAIL PROTECTED]> wrote:
> Ken Perl wrote:
> > I add your code to my startup.pl, but it doesn't work.
>
> Change 'print' to warn and you should see the expected behaviour.
>
> The only exception is stop, that doesn't go thru a restart cycle, so the
> correct way to detect a shutdown is to register a server_shutdown callback
> like this :
>
> Apache2::ServerUtil::server_shutdown_cleanup_register(sub { warn "I am 
> shutting down" });
>
> > # apache2ctl -k start
> > Program is starting
> > Stopping...
> >
> > #apache2ctl -k graceful
> > 
> >
> > # apache2ctl -k stop
> > 
> >
> > On 2/13/06, Frank Wiles <[EMAIL PROTECTED]> wrote:
> >
> >>On Mon, 13 Feb 2006 09:17:57 +0800
> >>Ken Perl <[EMAIL PROTECTED]> wrote:
> >>
> >>
> >>>No, I want to do something in the startup.pl script.
> >>>When apache start, I also want to say my modperl program is starting
> >>>and started, when stop, I want to say my modperl program is stopping
> >>>and stopped. When graceful, I want to say reloading and reloaded.
> >>>But I didn't find a API could help, any ideas?
> >>
> >> Hi Ken,
> >>
> >> I haven't tested this, but it should work. In your startup.pl do
> >> something like this:
> >>
> >> my $started = 0;
> >> my $cnt = Apache2::ServerUtil::restart_count();
> >>
> >> if( !$started and $cnt == 1 ) {
> >>  $started = 1;
> >>  print "Program is starting\n";
> >> }
> >>
> >> if( $cnt > 2 ) {
> >>  print "Restarting";
> >> }
> >>
> >> if( $started and $cnt == 1 ) {
> >>  print "Stopping...";
> >> }
> >>
>
> 
> Philippe M. Chiasson m/gozer\@(apache|cpan|ectoplasm)\.org/ GPG KeyID : 
> 88C3A5A5
> http://gozer.ectoplasm.org/ F9BF E0C2 480E 7680 1AE5 3631 CB32 A107 
> 88C3A5A5
>
>
>


--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: detecting server start, stop, graceful in startup.pl

2006-02-19 Thread Ken Perl

Though it will write to the log and it is wrong as the console.

On 2/18/06, Philippe M. Chiasson <[EMAIL PROTECTED]> wrote:
> Ken Perl wrote:
> > Better than before, but only 'start' works.
> >
> > debian:~/WebGUI# apache2ctl -k start
> > [Fri Feb 17 22:34:21 2006] preload.perl: Program is starting
> > Loading WebGUI.conf
> > debian:~/WebGUI# apache2ctl -k graceful
> > [Fri Feb 17 22:34:55 2006] preload.perl: Program is starting
> > Loading WebGUI.conf
> > [Fri Feb 17 22:34:57 2006] -e: I am shutting down at
> > /root/WebGUI/sbin/preload.perl line 33.
> > debian:~/WebGUI# apache2ctl -k stop
> > [Fri Feb 17 22:35:28 2006] preload.perl: Program is starting
> > Loading WebGUI.conf
> > [Fri Feb 17 22:35:29 2006] -e: I am shutting down at
> > /root/WebGUI/sbin/preload.perl line 33.
>
> the restarts should show up too in your error_log, not on the console.
>
> --
> Philippe M. Chiasson m/gozer\@(apache|cpan|ectoplasm)\.org/ GPG KeyID : 
> 88C3A5A5
> http://gozer.ectoplasm.org/ F9BF E0C2 480E 7680 1AE5 3631 CB32 A107 
> 88C3A5A5
>
>
>
>


--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

due core cpu

2006-02-22 Thread Ken Perl

Will a modperl2 program run faster on a cue core cpu machine than a
single cpu machine if we assume their speeds are same? In other words,
should I buy a new due core cpu machine?

--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

clusters

2006-05-09 Thread Ken Perl


Is it possible to make a modperl application to run in cluster? if
yes, how to do that? any doc?

--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: clusters

2006-05-09 Thread Ken Perl

Do you mean that the pound work great with apache 2.0 and apache 1.3 both?

On 5/9/06, Harmen <[EMAIL PROTECTED]> wrote:

On Tue, May 09, 2006 at 04:47:14PM +0800, Ken Perl wrote:
> Is it possible to make a modperl application to run in cluster? if
> yes, how to do that? any doc?

Take a few mod_perl servers and put one pound (http://www.apsis.ch/poung/) 
reverse
proxy in front of them.
Works great.

--
The Moon is Waxing Gibbous (85% of Full)

--
perl -e 'print unpack(u,"62V5N\"FME;G\!E

how to use Apache::DBI

2007-07-30 Thread Ken Perl

I've configured  the Apache::DBI in httpd.conf like this,

PerlModule Apache::DBI

I didn't have  Apache::DBI->connect_on_init($data_source, $username,
$auth, \%attr) in startup.pl since we don't use startup.pl. and the
doc says to config it in httpd.conf is OK.

my question is I can not find any persistent connections for my CGI
script. any usage error here?

-- 
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: how to use Apache::DBI

2007-07-30 Thread Ken Perl

I didn't run it under ModPerl::Registry, is there any risk to use the
module? maybe I have to run lots of testing to the existing scripts.

Could I just run use the previous connections via DBI->connect_cache
instead of the Apache::DBI in one single cgi script(but the script
will call other modules which may make new db connections with same db
password and attributes)? In my testing, I saw some different database
handlers returned instead of the same hash reference address.

On 7/30/07, Clinton Gormley <[EMAIL PROTECTED]> wrote:
> On Mon, 2007-07-30 at 18:21 +0800, Ken Perl wrote:
> > I've configured  the Apache::DBI in httpd.conf like this,
> >
> > PerlModule Apache::DBI
> >
> > I didn't have  Apache::DBI->connect_on_init($data_source, $username,
> > $auth, \%attr) in startup.pl since we don't use startup.pl. and the
> > doc says to config it in httpd.conf is OK.
> >
> > my question is I can not find any persistent connections for my CGI
> > script. any usage error here?
>
> Persistent connections will not work for straight CGI scripts, because
> CGI is not persistent.
>
> For each CGI request, a new perl interpreter is loaded, your script is
> compiled, run, and the Perl process exits.
>
> mod_perl IS persistent, as are your CGI scripts run via
> ModPerl::Registry etc.
>
> You don't need connect_on_init to work, but you do need mod_perl
>
> Clint
> >
>
>

-- 
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: how to use Apache::DBI

2007-07-30 Thread Ken Perl

OK, got it. Is it possible to use the same db connections in one
request? if yes, could you please show me how to implement this if a
cgi script calls many times other perl modules which requests db
connections to work.

On 7/31/07, Perrin Harkins <[EMAIL PROTECTED]> wrote:
> On 7/30/07, Ken Perl <[EMAIL PROTECTED]> wrote:
> > I didn't run it under ModPerl::Registry
>
> Did you run it under some other mod_perl module, like
> ModPerl::PerlRun?  If it's a CGI script, and you aren't running it
> through something like this, then you aren't running it through
> mod_perl and you won't be able to use persistent connections.
>
> > Could I just run use the previous connections via DBI->connect_cache
> > instead of the Apache::DBI in one single cgi script
>
> Not if you aren't running it through mod_perl.  CGI scripts exit after
> each request, so they can't do persistent connections.
>
> - Perrin
>


-- 
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: how to use Apache::DBI

2007-08-01 Thread Ken Perl

ok, I'll try connect_cached too.

On 8/1/07, Perrin Harkins <[EMAIL PROTECTED]> wrote:
> On 7/31/07, Clinton Gormley <[EMAIL PROTECTED]> wrote:
> > What you could do instead is to use DBI's connect_cached method, which
> > provides similar functionality.  I actually use this instead of
> > Apache::DBI, even when running under mod_perl.
>
> I also use connect_cached in many cases.  Just be aware of what you're
> missing out on, because Apache::DBI does some extra things to protect
> you.
>
> Here are the main differences:
> - Apache::DBI doesn't require code changes.  connect_cached() does.
> - Apache::DBI will avoid caching a connection that you open during
> server startup, so that you don't accidentally try to use the same
> connection in multiple processes (due to forking).
> - Apache::DBI issues an automatic rollback at the end of every
> request, so that if your code dies while it has a transaction going,
> that will be freed before the next request.  (This is only if you open
> your handles with AutoCommit off though.)
>
> Hmm, maybe this should be in the Apache::DBI documentation.
>
> - Perrin
>


-- 
perl -e 'print unpack(u,"62V5N\"FME;G\!E

Re: how to use Apache::DBI

2007-08-01 Thread Ken Perl

it is really good gotchas and should be included to the
ModPerl::Registry document. thanks a lot.

On 7/31/07, Clinton Gormley <[EMAIL PROTECTED]> wrote:
> On Tue, 2007-07-31 at 10:09 +0800, Ken Perl wrote:
> > I didn't run it under ModPerl::Registry, is there any risk to use the
> > module? maybe I have to run lots of testing to the existing scripts.
> >
>
>
> There is a risk to using this module, but also a significant benefit:
> speed.
>
> The risks come from:
>
>  - the fact that your script has been written to run one and then exit,
>so you may have a number of variables that aren't initialised
>properly - this can produce unexpected errors
>
>  - ModPerl::Registry wraps your script in another sub, which may produce
>unexpected closures (you would see warnings about variables not
>remaining shared in your log)
>
> So if you move to using Registry, then you will need to check your
> scripts.
>
> However, using Registry means that your script is loaded and compiled
> once, avoiding this performance hit for subsequent requests - hence it
> runs faster.
>
> Have a look at this post that I wrote about gotchas under Registry, it
> may help you migrate:
> http://www.gossamer-threads.com/lists/modperl/modperl/94203#94203
>
> Clint
>
>
>


-- 
perl -e 'print unpack(u,"62V5N\"FME;G\!E

mod_perl and CGI::Application.. Strange Behaviour

2007-10-26 Thread Perl Modulator


Hi everyone,

Recently I have ported my Apache server to mod_perl and the server is
working fine. I can say this by checking the error log and it says:

[Fri Oct 26 10:52:16 2007] [notice] SIGHUP received.  Attempting to restart
[Fri Oct 26 10:52:17 2007] [warn] module mod_php4.c is already added,
skipping
[Fri Oct 26 10:52:17 2007] [notice] Apache/1.3.33 (Darwin) PHP/4.3.9
mod_perl/1.26 configured -- resuming normal operations
[Fri Oct 26 10:52:17 2007] [notice] Accept mutex: flock (Default: flock)

And I am running this Apache on MAC. The entries for mod_perl in httpd.conf
is as follows:

Alias /perl /docroot/cgi-perl
PerlModule Apache::Registry

   SetHandler perl-script
   PerlHandlerApache::Registry
   PerlSendHeader On
   Options+ExecCGI


Problem

I am trying to create mod_perl based appln using CGI::Application and I am
just printing a message. Here is the Perl Module Test.pm, which will be
loaded.


package Test ;
use strict;
#our @ISA = qw[ CGI::Application ];
use base 'CGI::Application';

sub setup {

   my $self = shift;
   $self->run_modes( 'mode1' => 'mode1_sub' );
   $self->start_mode( 'mode1' );

}

sub mode1_sub {

   my $self = shift;
   my $req = $self->query();

   $req->content_type( 'text/html' );
   $req->send_http_header;

   return "using CGI::Application and mod_perl";
}

1;

Here is my CGI Script test.cgi:

use strict;
use lib '/docroot/cgi-perl/test';
use Test;

my $webapp = Test->new();
$webapp->run();

Both the CGI script and the module are in same directory. I am getting
following error message in my error log and I am just could not understand,
where I am doing wrong. 

[Fri Oct 26 11:05:39 2007] [error] Can't locate object method "new" via
package "Test" at /docroot/cgi-perl/test/test.cgi line 5.\n

Please guide me in this regard and it will be great help for my application.

Thanks.
-- 
View this message in context: 
http://www.nabble.com/mod_perl-and-CGI%3A%3AApplication..-Strange-Behaviour-tf4697759.html#a13428708
Sent from the mod_perl - General mailing list archive at Nabble.com.

[Newbie] MP2.0: PerlSendHeader versus PerlOptions (and Apache::Registry grrr...)

2008-04-24 Thread Perl Junkie

Okay, it's been a while, despite the name, that I've done any serious 
playing around with MP.  Since 1.0.  I've just installed and gotten 
running MP 2.0 on Apache 2.2.8...  I'm doing the "ground up" and trying 
to get registry scripts working first and I've run into the following, 
right-off-the-bat head scratchers:


(1) The documentation says that:


 |ParseHeaders|
 
<http://perl.apache.org/docs/2.0/user/config/config.html#toc_C_ParseHeaders_>

   Scan output for HTTP headers, same functionality as mod_perl 1.0's
   |PerlSendHeader|, but more robust. This option is usually needs to
   be enabled for registry scripts which send the HTTP header with:

 print "Content-type: text/html\n\n";


Doesn't work for me...  When I run registry scripts using in httpd.conf:

   Alias /mp "/var/www/mp"
   PerlModule ModPerl::Registry
   
  SetHandler perl-script
  Options +ExecCGI
  PerlHandler ModPerl::Registry
  *PerlOptions ParseHeaders*
  Order allow,deny
  Allow from all
   

The content-type is never sent and the response delivers text/plain -- I 
see the HTML in plain text.  When I go back to the MP 1.0 way in 
httpd.conf, it works:


   Alias /mp "/var/www/mp"
   PerlModule ModPerl::Registry
   
  SetHandler perl-script
  Options +ExecCGI
  PerlHandler ModPerl::Registry
  *PerlSendHeader On*
  Order allow,deny
  Allow from all
   

What's the story here?

ALSO...

(2) It took me a while to figure out that Apache::Registry has become 
ModPerl::Registry.  I saw nothing and still see nothing in the MP 2.0 
docs about this, what I would consider MAJOR change, and as I said, it 
took me a while to realize what happened here...  What gives?  What was 
wrong with Apache::Registry???


I had Stas' book, which I know was written during MP 1.0, so between 
these two things,  it leaves me wondering how much more of the book is 
now practically worthless??  I'm fast beginning to feel MP 2.0 has now 
become the domain of "old MP 1.0 hands" and if you didn't follow MP from 
1.0 days -- which I did not in a consistent way, then you are out of 
luck...  Where can I get the latest MP 2.0 beef?  I think the MP 2.0 
documentation at perl.apache.org leaves a lot to be desired...


Anyways, if someone can explain the ParseHeaders versus PerlSendHeader 
and the ModPerl::Registry versus Apache::Registry changes, that would 
help at least get registry scripts running in the optimal way.  If the 
MP 2.0 docs say that ParseHeaders is better, than I'd like to be able to 
use it, but right now, it doesn't seem to work as the docs say. :-(


-PJ

Re: [Newbie] MP2.0: PerlSendHeader versus PerlOptions (and Apache::Registry grrr...)

2008-04-24 Thread Perl Junkie

Perrin Harkins wrote:

On Fri, Apr 25, 2008 at 12:18 AM, Perl Junkie <[EMAIL PROTECTED]> wrote:

PerlOptions ParseHeaders

I think that should be +ParseHeaders. See
http://perl.apache.org/docs/2.0/user/porting/compat.html#C_PerlSendHeader_

I did finally see that in another section. I *have* to bother thousands
of people with a question before I find it in the docs, even if I've
already looked. I have no idea why it works out that way. :-( :-)

(2) It took me a while to figure out that Apache::Registry has become
ModPerl::Registry. I saw nothing and still see nothing in the MP 2.0 docs
about this, what I would consider MAJOR change, and as I said, it took me a
while to realize what happened here... What gives? What was wrong with
Apache::Registry???

This is answered in the migration guide for converting from 1.x to 2.x:
http://perl.apache.org/docs/2.0/user/porting/compat.html#C_Apache__Registry___C_Apache__PerlRun__and_Friends

Nice.

I had Stas' book, which I know was written during MP 1.0, so between these
two things, it leaves me wondering how much more of the book is now
practically worthless??

The information about system architecture, performance tuning,
debugging techniques, and general perl issues is all applicable, but
of course the API for mod_perl 2 is different and there's no way
around that. There is a new book by Stas and Jim Brandt about
mod_perl 2:
http://modperl2book.org/

I'll look into it. Thanks. Sounds invaluable.

I think the MP 2.0 documentation at
perl.apache.org leaves a lot to be desired...

Sorry to hear you don't like the documentation. If you have some
ideas for how to improve it, send them over.

H, seems like maybe my ability to spot the right information IN the
docs is what is leaving a lot to be desired right now. :-) Maybe I need
to just settle down and read some more first before assuming it's "not
there." Thanks!

The only suggestion so far would be to make that one section I quoted on
the ParseHeaders option a little more clearer than it is albeit it's
documented elsewhere, as you pointed out above.

What nice people here at MP... :-) Alright, I'll leave y'all alone and
do some more reading first. The migration guide seems like a good place
to start to convert all my past (way, way back in the past) remembrances
into the new MP2 reality.

-PJ

Re: decline and fall of modperl?

2009-03-23 Thread Perl Junkie


Byrne Reese wrote:
The problem is that there are no very many big sites that use perl 
either.
I knew that Amazon used Perl, than tried to use Java, than... I don't 
know what they use now.
Google uses Python, Yahoo uses PHP, Microsoft probably uses DotNet 
and Sun probably uses Java.


I will add:

* LiveJournal
* TypePad
* Vox
* Popular MT sites like:
  - Huffington Post
  - Gothamist
  - Talking Points Memo
  - many, many, many more of course


As far as I know, these sites also use Perl or mod_perl:

o Internet Movie Database (Amazon)
o TicketMaster (last I checked, but that was a couple of years ago and 
they were recently sold or bought or something)

o CitySearch.com
o ArtToday.com
o PBS Online

I also know that a very large international cosmetics company uses 
mod_perl quite extensively.  I've done some work for them and I don't 
know if or how they would care for their name being mentioned here, so I 
won't, but...  If you want "large" -- they define "large."


-
I wouldn't "count" a company that uses a Perl "product" or CMS or some 
sort (such as MT) as a "perl site" in the strict sense of the word.  
Some of the techies at places like Huffington Post may be aware of the 
underlying technology, but I doubt very much that "Perl" itself drove 
the decision as much as the great work that Ben and Mena Trott did at 
SixApart.com in creating such a product.


-
And I have to chime in and say I've seen plenty of ugly, ridiculous code 
written in every language imaginable (everyone's precious "Java" and JEE 
very much included.)  It's not a product of the language as much as the 
language bearer.  Sorry.  That's always been the case.  Margin for error 
on a tight mountain road doesn't automatically a bad driver make -- only 
the one who will not heed the type of road and car he's in in the 
situation.  Anyone can zoom through an intersection full of traffic on a 
wide-open, flat country road as well.  The road being the intent and 
direction and the car being Perl (and pretty cotton-pickin' good one, I 
might add), I don't see the correlation here.  Again, this is clever 
smoke and mirrors by consultants with a different agenda and product to 
sell.  The craft of programming and development are only slightly 
impacted by these kinds of things.


-pj

Re: decline and fall of modperl?

2009-03-23 Thread Perl Junkie


Alexandr Ciornii wrote:

Hi.

It possible to encrypt perl sources with same safety as with PHP -
with possibility of source decryption. But Perl developers are in
general more advanced than PHP developers so they know how to decrypt
it, in contrast to PHP developers that do not know that encrypted PHP
sources can easily be decrypted. There are also some bytecode
compilers for Perl (alpha stage).

2009/3/23 Octavian Râsnita :

  

PHP, C#, Java are much more prefered, because the programs created with them
can hide the source code much better, while this is not possible with Perl.
This is a big reason why the software companies that create custom programs
for their clients prefer to use those languages and not perl. Not because
perl is bad.



  
Not to keep the PHP thing going, but I would tend to de-classify about 
30% of the people who call themselves "PHP developers."  There is a VERY 
high propensity among many in this community to go out and "borrow code" 
and "retrofit it."  I'm not saying Perl developers don't do this, but I 
don't believe there are very many heavily invested in the Perl community 
who have this bent -- many are devoted to the craft of true, honest to 
goodness OO development.


I also left out a very important and I would say very well known Perl 
*powerhouse* in corporate America:


WhitePages.com

Some of the best of the best work there.  You can't even get a sniff 
from them (eg. interview) without being in the upper crust of Perl 
development.  No Perl "wanna-bes" allowed there.  All very solid OO 
development only.


-pj

Re: test

2018-05-22 Thread Practical Perl

welcome.

On Tue, May 22, 2018 at 5:34 AM, Jie Gao  wrote:

> test
>
>

CPAN will shutdown

2018-05-22 Thread Practical Perl

 *search.cpan.org <http://search.cpan.org> is shutting down*
For details read Perl NOC
<https://log.perl.org/2018/05/goodbye-search-dot-cpan-dot-org.html>. After
June 25th this page will redirect to MetaCPAN.org <http://search.mcpan.org/>


So does as well as Perl?

Output filters, data encoding


Hi.
I'm writing a new PerlOutputFilter, stream version.
I have written several working ones before, so I know the general scheme.
But in this latest filter, I have a problem with the data encoding, which I did not 
encounter previously.
I did not find an answer in the on-line mod_perl documentation, which seems silent about 
any encoding of the data that one might read in a filter.

(I also tried the WWW, without any more success).

To read the response data coming from upstream, I use the standard $f->read(), 
as in

while (my $sz = $f->read(my $buffer, BUFF_LEN)) {
..

and then I need to pass this data to another module for processing 
(Template::Toolkit).
To make a long story short, Template::Toolkit misinterprets the data I'm sending to it, 
because this data /is/ actually UTF-8, but apparently not marked so internally by the 
$f->read(). So TT2 re-encodes it, leading to double UTF-8 encoding.


My question is : can I - and how -, set the filehandle that corresponds to the $f->read(), 
to a UTF-8 layer ?

I have tried

line 155: binmode($f,'encoding:(UTF-8)');

and that triggers an error :
 Not a GLOB reference at (my filter) line 155.\n
)

Or do I need to read the data 'as is', and separately do an

 $decoded_buffer = decode('UTF-8', $buffer);

?

Note 1 : I can of course do the above decode(), but it seems more elegant to just insert 
the I/O layer.


Note 2 : I know that the pre-filter data is UTF-8 (or at least I choose to believe that it 
is), because I check $f->r->content_type(), which returns "text/html;charset=UTF-8".
Also, the filter otherwise works as expected, and I get the expected output in the 
browser, except that e.g. "München" becomes "MÃ¼nchen".


Thanks in advance for any tips
André

Re: Output filters, data encoding


On 13.11.2019 19:17, p...@cpan.org wrote:

On Wednesday 13 November 2019 19:12:10 André Warnier (tomcat/perl) wrote:

My question is : can I - and how -, set the filehandle that corresponds to
the $f->read(), to a UTF-8 layer ?
I have tried

line 155: binmode($f,'encoding:(UTF-8)');


Hi André! When specifying PerlIO layer for file handle, you need to
write colon character before layer name. So correct binmode call is:

   binmode($f, ':encoding(UTF-8)');


and that triggers an error :
  Not a GLOB reference at (my filter) line 155.\n
)


Thanks. Ooops, that was a typo (also in my filter, not only in the list 
message).
But correcting it, does not change the GLOB error message.

Re: Output filters, data encoding


On 13.11.2019 19:53, p...@cpan.org wrote:

On Wednesday 13 November 2019 19:52:25 André Warnier (tomcat/perl) wrote:

On 13.11.2019 19:17, p...@cpan.org wrote:

On Wednesday 13 November 2019 19:12:10 André Warnier (tomcat/perl) wrote:

My question is : can I - and how -, set the filehandle that corresponds to
the $f->read(), to a UTF-8 layer ?
I have tried

line 155: binmode($f,'encoding:(UTF-8)');


Hi André! When specifying PerlIO layer for file handle, you need to
write colon character before layer name. So correct binmode call is:

binmode($f, ':encoding(UTF-8)');


and that triggers an error :
   Not a GLOB reference at (my filter) line 155.\n
)


Thanks. Ooops, that was a typo (also in my filter, not only in the list 
message).
But correcting it, does not change the GLOB error message.


Ok. What is the $f? It is object or what kind of scalar?


It is the Apache2::Filter object.
See : http://perl.apache.org/docs/2.0/api/Apache2/Filter.html
Configured in httpd as :   PerlOutputFilterHandler MyFilter
See also :  http://perl.apache.org/docs/2.0/user/handlers/filters.html

My (hopeful) thinking was that considering the
$f->read()
the Apache2::Filter object may also be a FileHandle, hence the attempt at
binmode($f,..)
But that seems to be incorrect.
(And I don't see any (documented) method of Apache2::Filter that would return the 
underlying FileHandle either)

Re: Output filters, data encoding


On 13.11.2019 19:37, Damyan Ivanov wrote:

-=| André Warnier (tomcat/perl), 13.11.2019 19:12:10 +0100 |=-

while (my $sz = $f->read(my $buffer, BUFF_LEN)) {
..

and then I need to pass this data to another module for processing 
(Template::Toolkit).
To make a long story short, Template::Toolkit misinterprets the data I'm
sending to it, because this data /is/ actually UTF-8, but apparently not
marked so internally by the $f->read(). So TT2 re-encodes it, leading to
double UTF-8 encoding.

My question is : can I - and how -, set the filehandle that corresponds to
the $f->read(), to a UTF-8 layer ?
I have tried

line 155: binmode($f,'encoding:(UTF-8)');

and that triggers an error :
  Not a GLOB reference at (my filter) line 155.\n
)

Or do I need to read the data 'as is', and separately do an

  $decoded_buffer = decode('UTF-8', $buffer);


There's a middle ground - partial decoding. See Encode(1)/FB_QUIET:

If CHECK is set to "Encode::FB_QUIET", encoding and decoding
immediately return the portion of the data that has been processed so
far when an error occurs. The data argument is overwritten with
everything after that point; that is, the unprocessed portion of the
data.  This is handy when you have to call "decode" repeatedly in the
case where your source data may contain partial multi-byte character
sequences, (that is, you are reading with a fixed-width buffer). Here's
some sample code to do exactly that:

my($buffer, $string) = ("", "");
while (read($fh, $buffer, 256, length($buffer))) {
$string .= decode($encoding, $buffer, Encode::FB_QUIET);
# $buffer now contains the unprocessed partial character
}

Looks exactly like your case.


Thanks for the response and the tip.

My idea of adding a UTF-8 layer to the filehandle through which Apache2::Filter reads the 
incoming data was probably wrong anyway : it cannot do that, because it gets this data 
originally in chunks, as "bucket brigades" from Apache httpd.  And there is no guarantee 
that such a bucket brigade would always end in "complete" UTF-8 character sequences.

At the very least, this would probably complicate the code underlying 
$f->read() quite a bit.
It is clearer to handle that in the filter itself.

The Encode::FB_QUIET flag above, with the incremental buffer read, is really 
smart.
Unfortunately, the Apache2::Filter read() method does not allow as many arguments, and all 
one has is something like this :


my $accumulated_content = "";
while (my $sz = $f->read(my $buffer, BUFF_LEN)) {
$accumulated_content .= $buffer;
}

Luckily, in this case, I have to accumulate the complete response content anyway, before I 
can decide to call Template::Toolkit on it or not. So I can do a single decode() on 
$accumulated_content. Not the most efficient memory-wise, but good enough in this case.

Re: Output filters, data encoding

2019-11-14 Thread tomcat/perl


On 14.11.2019 01:09, Hua, Yong wrote:

Hi

on 2019/11/14 2:12, André Warnier (tomcat/perl) wrote:

I'm writing a new PerlOutputFilter, stream version.


Can you give a more general introduction for what is "stream version"?

Thank you.

You shoud read the pages which I referred to previously, they explain this better than I 
could do :

1) http://perl.apache.org/docs/2.0/user/handlers/filters.html
2) http://perl.apache.org/docs/2.0/api/Apache2/Filter.html

See in particular here :
http://perl.apache.org/docs/2.0/user/handlers/filters.html#Two_Methods_for_Manipulating_Data

Re: decrypt the posted content

2019-11-20 Thread tomcat/perl


On 20.11.2019 10:26, Tillman Peng wrote:

hello

My client post the data body which is encrypted with public-key.
the private key is deployed in web server,powered by mp2.
How can I correctly decrypt the data with private key from within modperl 
handler?



Hi.
Do you have a separate command-line program on the server which can decrypt 
that content ?
If yes : if you do not find an appropriate perl module to do this decryption, your 
mod_perl handler can always execute that external program using the system() function.

(See : https://perldoc.perl.org/5.30.0/functions/system.html)

General idea :
- get the encrypted content from the request
- write this encrypted content to a file in some appropriate work directory on 
the server
- compose the external command that reads the encrypted data, and writes the decrypted 
content to a file

- execute that command with system()
- check for errors
- read the decrypted results file
- clean up

If you end up using this method, and you are doing this from within an Apache/mod_perl 
handler, you have to be extra careful about many aspects, such as :
- catching any errors which may happen in the external program, and interpret them 
correctly in the calling module.

- logging the errors properly, so that if "it doesn't work", you can find out 
why
- taking into account that your webserver may receive several simultaneous requests for 
such content, and thus that there may be several instances of that external command 
running at the same time (think about the temporary files that you may need, and make sure 
that each instance uses its own unique files)

- cleaning up after succesfully running the command
- maybe selectively "not cleaning up" if there were any problems, so that you can inspect 
what happened
- check permissions (the external program will run under the same user-id as the 
webserver, so whatever it writes, must be in a directory writeable by the webserver)
- verify that the external command cannot be running for too long, causing the client to 
time-out waiting for a response, and closing the connection to the webserver
- make extra sure that the client cannot, through some malicious use of the parameters 
that it sends to the server (e.g. filenames), result in damage on your server

(e.g. system("program > /etc/passwd"))
- etc.

If you prefer to use a perl module to do the decryption, you will have to look at what is 
available on CPAN.  Most modules that relate to encryption/decryption are in the "Crypt" 
namespace, such as : https://metacpan.org/search?q=crypt%3A%3A

installing mod_perl on Windows 10

2020-01-08 Thread tomcat/perl


Hi.
After (very) long and faithful service, I had to replace my old Windows XP development 
laptop by a new one, which of course came with Windows 10.

And I am having problems installing Apache 2.4 + Strawberry perl + mod_perl.

Apache 2.4 is from ApacheLounge 
(https://www.apachelounge.com/download/VS16/binaries/httpd-2.4.41-win64-VS16.zip) and 
installed under C:\apache24

Strawberry perl is (see below)
and both seem to work separately.

Next I try to install mod_perl2 from CPAN and I get the following, see below.
Any clue as to what I'm doing wrong ?

(Note : prior to this, I tried to install mod_perl manually from the downloaded file 
"mod_perl-2.0.10-strawberryperl-5.26.1.1-64bit.zip", but this seems to result in modules 
having wrong line endings and crashing the compiler).


C:\Users\aw>cpan mod_perl2
Loading internal null logger. Install Log::Log4perl for logging messages
CPAN: CPAN::SQLite loaded ok (v0.211)
CPAN: LWP::UserAgent loaded ok (v6.33)
CPAN: Time::HiRes loaded ok (v1.9758)
Fetching with LWP:
http://cpan.strawberryperl.com/authors/01mailrc.txt.gz
CPAN: YAML::XS loaded ok (v0.69)
Fetching with LWP:
http://cpan.strawberryperl.com/modules/02packages.details.txt.gz
Fetching with LWP:
http://cpan.strawberryperl.com/modules/03modlist.data.gz
Creating database file ...
Done!
Running install for module 'mod_perl2'
Fetching with LWP:
http://cpan.strawberryperl.com/authors/id/S/SH/SHAY/mod_perl-2.0.11.tar.gz
CPAN: Digest::SHA loaded ok (v6.01)
Fetching with LWP:
http://cpan.strawberryperl.com/authors/id/S/SH/SHAY/CHECKSUMS
CPAN: Compress::Zlib loaded ok (v2.081)
Checksum for 
C:\STRAWB~1\cpan\sources\authors\id\S\SH\SHAY\mod_perl-2.0.11.tar.gz ok
CPAN: Archive::Tar loaded ok (v2.26)
CPAN: CPAN::Meta::Requirements loaded ok (v2.140)
CPAN: Parse::CPAN::Meta loaded ok (v2.150010)
CPAN: CPAN::Meta loaded ok (v2.150010)
Configuring S/SH/SHAY/mod_perl-2.0.11.tar.gz with Makefile.PL
no conflicting prior mod_perl version found - good.

Next we need to know where the 'apxs' script is located. This script
provides a lot of information about the Apache installation, and makes
it easier to find things on your system. Normally it's located in the
same directory as the 'httpd' executable.

If you don't yet have Apache installed you can build Apache against
the Apache source code, but you won't be able to run the test suite (a
very important step). Therefore you may want to install Apache before
proceeding.


Please provide a full path to 'apxs' executable
(press Enter if you don't have it installed):


Configure mod_perl with C:\Apache24? [y] y
Configuring Apache/2.4.41 mod_perl/2.0.11 Perl/v5.26.2
Use of uninitialized value $httpd_version in string lt at 
lib/ModPerl/MapUtil.pm line 51.
Checking if your kit is complete...
Looks good
Warning (mostly harmless): No library found for 
C:/STRAWB~1/cpan/build/mod_perl-2.0.11-0/src/modules/perl/mod_perl.lib
ERROR from evaluation of 
C:\STRAWB~1\cpan\build\mod_perl-2.0.11-0\Apache-Reload/Makefile.PL: Use of uninitialized 
value $v in substitution (s///) at Apache-Test/lib/Apache/TestRun.pm line 1100.

Warning: No success on command[C:\Strawberry\perl\bin\perl.exe Makefile.PL]
  SHAY/mod_perl-2.0.11.tar.gz
  C:\Strawberry\perl\bin\perl.exe Makefile.PL -- NOT OK
Stopping: 'install' failed for 'mod_perl2'.

C:\Users\aw>

Perl :
C:\Users\aw>perl -V
Summary of my perl5 (revision 5 version 26 subversion 2) configuration:

  Platform:
osname=MSWin32
osvers=10.0.16299.371
archname=MSWin32-x64-multi-thread
uname='Win32 strawberry-perl 5.26.2.1 #1 Sun Apr 15 13:52:49 2018 x64'
config_args='undef'
hint=recommended
useposix=true
d_sigaction=undef
useithreads=define
usemultiplicity=define
use64bitint=define
use64bitall=undef
uselongdouble=undef
usemymalloc=n
default_inc_excludes_dot=define
bincompat5005=undef
  Compiler:
cc='gcc'
ccflags =' -s -O2 -DWIN32 -DWIN64 -DCONSERVATIVE -D__USE_MINGW_ANSI_STDIO 
-DPERL_TEXTMODE_SCRIPTS -DPERL_IMPLICIT_CONTEXT -DPERL_IMPLICIT_SYS -DUSE_PERLIO -fwrapv 
-fno-strict-aliasing -mms-bitfields'

optimize='-s -O2'
cppflags='-DWIN32'
ccversion=''
gccversion='7.1.0'
gccosandvers=''
intsize=4
longsize=4
ptrsize=8
doublesize=8
byteorder=12345678
doublekind=3
d_longlong=define
longlongsize=8
d_longdbl=define
longdblsize=16
longdblkind=3
ivtype='long long'
ivsize=8
nvtype='double'
nvsize=8
Off_t='long long'
lseeksize=8
alignbytes=8
prototype=define
  Linker and Libraries:
ld='g++'
ldflags ='-s -L"C:\STRAWB~1\perl\lib\CORE" -L"C:\STRAWB~1\c\lib"'
libpth=C:\STRAWB~1\c\lib C:\STRAWB~1\c\x86_64-w64-mingw32\lib 
C:

Re: installing mod_perl on Windows 10

2020-01-08 Thread tomcat/perl


On 08.01.2020 17:19, Edward J. Sabol wrote:

On Jan 8, 2020, at 10:17 AM, André Warnier (tomcat/perl)  
wrote:

Please provide a full path to 'apxs' executable
(press Enter if you don't have it installed):


Configure mod_perl with C:\Apache24? [y] y


It seems to me you didn’t provide the *full* path to the apxs executable. I 
would think that the full path should end with “\bin\apxs”.

Hope this helps,
Ed



In fact it never really asks. The only thing I entered, was the "cpan 
mod_perl2" command.
All the rest is just output, without any prompt.
For example here :

>> Configure mod_perl with C:\Apache24? [y] y

the "y" answer was not entered by me, it did it itself.

Re: installing mod_perl on Windows 10

2020-01-14 Thread tomcat/perl


Hi.
One more time.
I am trying to install Apache 2.4 (64-bit, ApacheLounge) + perl (Strawberry, 5.26.2, 
64-bit) + mod_perl 2 (all versions shown below in cpanm output), on a (German) Windows 10 
laptop. (*)


The work directory below is leftover from a previous command "cpanm mod_perl2".
Apache 2.4 is installed and starts up (as a service), error log :
[Wed Jan 08 14:38:40.071509 2020] [mpm_winnt:notice] [pid 15800:tid 820] AH00455: 
Apache/2.4.41 (Win64) configured -- resuming normal operations

..
(note : the c:\apache24\bin directory contains no apxs script, and according to the source 
files of mod_perl, apxs does not seem to work under MSWin anyway)


perl works

but the mod_perl install breaks, build log below.
Any clues anyone ?

log (in Windows 10 command window, run as administrator) :

C:\Users\aw\.cpanm\work\1578998136.19100\mod_perl-2.0.11>
C:\Users\aw\.cpanm\work\1578998136.19100\mod_perl-2.0.11>cpanm --interactive 
--prompt .
--> Working on .
Configuring mod_perl-2.0.11 ... no conflicting prior mod_perl version found - 
good.

Next we need to know where the 'apxs' script is located. This script
provides a lot of information about the Apache installation, and makes
it easier to find things on your system. Normally it's located in the
same directory as the 'httpd' executable.

If you don't yet have Apache installed you can build Apache against
the Apache source code, but you won't be able to run the test suite (a
very important step). Therefore you may want to install Apache before
proceeding.


Please provide a full path to 'apxs' executable
(press Enter if you don't have it installed):


Configure mod_perl with C:\Apache24? [y] y
Configuring Apache/2.4.41 mod_perl/2.0.11 Perl/v5.26.2
Checking if your kit is complete...
Looks good
Warning (mostly harmless): No library found for 
C:/Users/aw/.cpanm/work/1578998136.19100/mod_perl-2.0.11/src/modules/perl/mod_perl.lib
ERROR from evaluation of 
C:\Users\aw\.cpanm\work\1578998136.19100\mod_perl-2.0.11\Apache-Reload/Makefile.PL: Use of 
uninitialized value $v in substitution (s///) at Apache-Test/lib/Apache/TestRun.pm line 1100.

Configuring  failed.
You can s)kip, r)etry, e)xamine build log, or l)ook ? [s] l
...


(*) which may interfere with the process in unfathomable ways.
E.g. there is no directory "c:\users" visible in Windows Explorer, it is instead called 
"C:\Benutzer". But that does not seem to immediately stop the command-line process from 
changing to the work directory  C:/Users/aw/.cpanm/work/, go figure..

mod_perl, compilation issues

2020-01-14 Thread tomcat/perl


Hi.

OS : Windows 10
Apache httpd 2.4.41
perl (Strawberry) : perl 5, version 26, subversion 2 (v5.26.2) built for 
MSWin32-x64-multi-thread

mod_perl : from Steve Hay's mod_perl-2.0.10-strawberryperl-5.26.1.1-64bit.zip,
as per Apache log :
Apache/2.4.41 (Win64) mod_perl/2.0.10 Perl/v5.26.2 configured

Problem :

C:\>perl -cw D:\develop\06_SVN\AP2lib\trunk\modlib\AUTH\UMA2.pm
Can't load 'C:/Strawberry/perl/site/lib/auto/ModPerl/Const/Const.dll' for module 
ModPerl::Const: load_file:Das angegebene Modul wurde nicht gefunden at 
C:/Strawberry/perl/lib/DynaLoader.pm line 193.

  at C:/Strawberry/perl/site/lib/Apache2/Const.pm line 19.
Compilation failed in require at C:/Strawberry/perl/site/lib/Apache2/Const.pm 
line 19.
BEGIN failed--compilation aborted at 
C:/Strawberry/perl/site/lib/Apache2/Const.pm line 19.
Compilation failed in require at 
D:\develop\06_SVN\AP2lib\trunk\modlib\AUTH\UMA2.pm line 28.
BEGIN failed--compilation aborted at D:\develop\06_SVN\AP2lib\trunk\modlib\AUTH\UMA2.pm 
line 28.


Line 28 of UMA2.pm is :

use Apache2::Const -compile => qw(:common :http :methods :override :proxy :log REDIRECT 
AUTHZ_GRANTED AUTHZ_DENIED AUTHZ_DENIED_NO_USER AUTHZ_GENERAL_ERROR);


(wrapping due to email)

C:/Strawberry/perl/site/lib/Apache2/Const.pm line 19 is :

use ModPerl::Const ();

File C:\strawberry\perl\site\lib\ModPerl\Const.pm exists, but has no 
"load_file" in it.

The file C:/Strawberry/perl/site/lib/auto/ModPerl/Const/Const.dll exists.

For non-german-speakers, the above message "load_file:Das angegebene Modul wurde nicht 
gefunden" translates as "load_file:the specified module was not found".


The above module "UMA2.pm" compiles without error on a Linux server with similar httpd, 
perl and mod_perl versions.


Any clue or additional questions welcome.
Thanks

mod_perl, compilation issues, cont

2020-01-16 Thread tomcat/perl


Can anyone tell me what the error below really means ?
(and if possible, give me a clue as to what may be happening ?)

To get possibly more details, I re-ran the compile command after setting PERL_DL_DEBUG in 
the environment (just something I found in Dynaloader.pm). It does give some more detail, 
but I am still stumped..


Full config described below.
D:\develop\06_SVN\AP2lib\trunk\modlib\AUTH\UMA2.pm is my module, which compiles just fine 
on Linux servers with similar software installed (Apache, mod_perl, perl).



C:\>perl -cw D:\develop\06_SVN\AP2lib\trunk\modlib\AUTH\UMA2.pm
DynaLoader.pm loaded (C:/Strawberry/perl/site/lib C:/Strawberry/perl/vendor/lib 
C:/Strawberry/perl/lib, C:\STRAWB~1\c\lib C:\STRAWB~1\c\x86_64-w64-mingw32\lib 
C:\STRAWB~1\c\lib\gcc\x86_64-w64-mingw32\7.1.0)

DynaLoader::bootstrap for ModPerl::Const (auto/ModPerl/Const/Const.xs.dll)
dl_findfile(-LC:/Strawberry/perl/site/lib/auto/ModPerl/Const -LC:/Strawberry/perl/site/lib 
-LC:/Strawberry/perl/vendor/lib -LC:/Strawberry/perl/lib Const)

 checking in C:/Strawberry/perl/site/lib/auto/ModPerl/Const for Const.xs.dll
 checking in C:/Strawberry/perl/site/lib/auto/ModPerl/Const for Const.dll
dl_findfile found: C:/Strawberry/perl/site/lib/auto/ModPerl/Const/Const.dll
Can't load 'C:/Strawberry/perl/site/lib/auto/ModPerl/Const/Const.dll' for module 
ModPerl::Const: load_file:Das angegebene Modul wurde nicht gefunden at 
C:/Strawberry/perl/lib/DynaLoader.pm line 193.

  at C:/Strawberry/perl/site/lib/Apache2/Const.pm line 19.
Compilation failed in require at C:/Strawberry/perl/site/lib/Apache2/Const.pm 
line 19.
BEGIN failed--compilation aborted at 
C:/Strawberry/perl/site/lib/Apache2/Const.pm line 19.
Compilation failed in require at 
D:\develop\06_SVN\AP2lib\trunk\modlib\AUTH\UMA2.pm line 28.
BEGIN failed--compilation aborted at D:\develop\06_SVN\AP2lib\trunk\modlib\AUTH\UMA2.pm 
line 28.


-

OS : Windows 10
Apache httpd 2.4.41
perl (Strawberry) : perl 5, version 26, subversion 2 (v5.26.2) built for 
MSWin32-x64-multi-thread

mod_perl : from Steve Hay's mod_perl-2.0.10-strawberryperl-5.26.1.1-64bit.zip,
as per Apache log :
Apache/2.4.41 (Win64) mod_perl/2.0.10 Perl/v5.26.2 configured

Problem :

C:\>perl -cw D:\develop\06_SVN\AP2lib\trunk\modlib\AUTH\UMA2.pm
Can't load 'C:/Strawberry/perl/site/lib/auto/ModPerl/Const/Const.dll' for module 
ModPerl::Const: load_file:Das angegebene Modul wurde nicht gefunden at 
C:/Strawberry/perl/lib/DynaLoader.pm line 193.

  at C:/Strawberry/perl/site/lib/Apache2/Const.pm line 19.
Compilation failed in require at C:/Strawberry/perl/site/lib/Apache2/Const.pm 
line 19.
BEGIN failed--compilation aborted at 
C:/Strawberry/perl/site/lib/Apache2/Const.pm line 19.
Compilation failed in require at 
D:\develop\06_SVN\AP2lib\trunk\modlib\AUTH\UMA2.pm line 28.
BEGIN failed--compilation aborted at D:\develop\06_SVN\AP2lib\trunk\modlib\AUTH\UMA2.pm 
line 28.


Line 28 of UMA2.pm is :

use Apache2::Const -compile => qw(:common :http :methods :override :proxy :log REDIRECT 
AUTHZ_GRANTED AUTHZ_DENIED AUTHZ_DENIED_NO_USER AUTHZ_GENERAL_ERROR);


(wrapping due to email)

C:/Strawberry/perl/site/lib/Apache2/Const.pm line 19 is :

use ModPerl::Const ();

File C:\strawberry\perl\site\lib\ModPerl\Const.pm exists, but has no 
"load_file" in it.

The file C:/Strawberry/perl/site/lib/auto/ModPerl/Const/Const.dll exists.

For non-german-speakers, the above message "load_file:Das angegebene Modul wurde nicht 
gefunden" translates as "load_file:the specified module was not found".


The above module "UMA2.pm" compiles without error on a Linux server with similar httpd, 
perl and mod_perl versions.


Any clue or additional questions welcome.
Thanks

Re: mod_perl, compilation issues, cont

2020-02-03 Thread tomcat/perl

Thank you very much for your reply below. I was getting desperate of getting
any at all.
I will check the permissions issue and post back to the list.

Maybe in the mneantime an additional question to the list : has *anyone at all* succeeded
in installing (and using) Apache 2.4 / perl 5.20+ / mod_perl 2.x on a Windows 10 platform ?
Knowing that someone else has been able to do that, would already help me in focusing my
debugging efforts to something specific to my laptop.

On 31.01.2020 16:45, Bruce Johnson wrote:
I am not a Windows expert by any means, but it it possible that it’s a permissions issue
on that DLL file or enclosing directory(s)? Folder permissions in Windows will let you
grant directory listing rights but not read rights for the process that’s doing the
compilation…

See:
https://www.online-tech-tips.com/computer-tips/set-file-folder-permissions-windows/ under
Permission Types; the specific right is ‘List Folder Contents’

This would account for both dl_findfile working and the ‘Can’t Load file’ error because
the first utilizes the listing rights and the second needs to actually read the file.

On Jan 16, 2020, at 1:34 AM, André Warnier (tomcat/perl) <mailto:a...@ice-sa.com>> wrote:

Can anyone tell me what the error below really means ?
(and if possible, give me a clue as to what may be happening ?)

To get possibly more details, I re-ran the compile command after setting PERL_DL_DEBUG
in the environment (just something I found in Dynaloader.pm). It does give some more
detail, but I am still stumped..

Full config described below.
D:\develop\06_SVN\AP2lib\trunk\modlib\AUTH\UMA2.pm is my module, which compiles just
fine on Linux servers with similar software installed (Apache, mod_perl, perl).

C:\>perl -cw D:\develop\06_SVN\AP2lib\trunk\modlib\AUTH\UMA2.pm
DynaLoader.pm loaded (C:/Strawberry/perl/site/lib C:/Strawberry/perl/vendor/lib
C:/Strawberry/perl/lib, C:\STRAWB~1\c\lib C:\STRAWB~1\c\x86_64-w64-mingw32\lib
C:\STRAWB~1\c\lib\gcc\x86_64-w64-mingw32\7.1.0)

DynaLoader::bootstrap for ModPerl::Const (auto/ModPerl/Const/Const.xs.dll)
dl_findfile(-LC:/Strawberry/perl/site/lib/auto/ModPerl/Const
-LC:/Strawberry/perl/site/lib -LC:/Strawberry/perl/vendor/lib -LC:/Strawberry/perl/lib
Const)

checking in C:/Strawberry/perl/site/lib/auto/ModPerl/Const for Const.xs.dll
checking in C:/Strawberry/perl/site/lib/auto/ModPerl/Const for Const.dll
dl_findfile found: C:/Strawberry/perl/site/lib/auto/ModPerl/Const/Const.dll
Can't load 'C:/Strawberry/perl/site/lib/auto/ModPerl/Const/Const.dll' for module
ModPerl::Const: load_file:Das angegebene Modul wurde nicht gefunden at
C:/Strawberry/perl/lib/DynaLoader.pm line 193.

at C:/Strawberry/perl/site/lib/Apache2/Const.pm line 19.
Compilation failed in require at C:/Strawberry/perl/site/lib/Apache2/Const.pm
line 19.
BEGIN failed--compilation aborted at
C:/Strawberry/perl/site/lib/Apache2/Const.pm line 19.
Compilation failed in require at
D:\develop\06_SVN\AP2lib\trunk\modlib\AUTH\UMA2.pm line 28.
BEGIN failed--compilation aborted at D:\develop\06_SVN\AP2lib\trunk\modlib\AUTH\UMA2.pm
line 28.

OS : Windows 10
Apache httpd 2.4.41
perl (Strawberry) : perl 5, version 26, subversion 2 (v5.26.2) built for
MSWin32-x64-multi-thread

mod_perl : from Steve Hay's mod_perl-2.0.10-strawberryperl-5.26.1.1-64bit.zip,
as per Apache log :
Apache/2.4.41 (Win64) mod_perl/2.0.10 Perl/v5.26.2 configured

Problem :

C:\>perl -cw D:\develop\06_SVN\AP2lib\trunk\modlib\AUTH\UMA2.pm
Can't load 'C:/Strawberry/perl/site/lib/auto/ModPerl/Const/Const.dll' for module
ModPerl::Const: load_file:Das angegebene Modul wurde nicht gefunden at
C:/Strawberry/perl/lib/DynaLoader.pm line 193.

Line 28 of UMA2.pm is :

use Apache2::Const -compile => qw(:common :http :methods :override :proxy :log REDIRECT
AUTHZ_GRANTED AUTHZ_DENIED AUTHZ_DENIED_NO_USER AUTHZ_GENERAL_ERROR);

(wrapping due to email)

C:/Strawberry/perl/site/lib/Apache2/Const.pm line 19 is :

use ModPerl::Const ();

File C:\strawberry\perl\site\lib\ModPerl\Const.pm exists, but has no
"load_file" in it.

The file C:/Strawberry/perl/site/lib/auto/ModPerl/Const/Const.dll exists.

For non-german-speakers, the above message "load_file:Das angegebene Modul wurde nicht
gefunden" translates as "load_file:the specified module was not found".

The above module "UMA2.pm" compiles without error on a Linux server with similar httpd,
perl and mod_perl versions.

Any

Re: mod_perl, compilation issues, cont

2020-02-03 Thread tomcat/perl


Additional info :

I checked the permissions of all below c:/strawberry/perl, and one thing I can say is that 
all directories and files below that point have at least permissions read/write/execute 
for all "authenticated users".


On the other hand, I also ran the following (why did I not think of that before ?), just 
to make sure that the issue would be reproducible with only the "vanilla" installed Apache 
2.4, perl 5.26.3, mod_perl 2.x :


C:\strawberry\perl\site>set PERL_DL_DEBUG=1

C:\strawberry\perl\site>perl -cw C:/Strawberry/perl/site/lib/Apache2/Const.pm
DynaLoader.pm loaded (C:/Strawberry/perl/site/lib C:/Strawberry/perl/vendor/lib 
C:/Strawberry/perl/lib, C:\STRAWB~1\c\lib C:\STRAWB~1\c\x86_64-w64-mingw32\lib 
C:\STRAWB~1\c\lib\gcc\x86_64-w64-mingw32\7.1.0)

DynaLoader::bootstrap for ModPerl::Const (auto/ModPerl/Const/Const.xs.dll)
dl_findfile(-LC:/Strawberry/perl/site/lib/auto/ModPerl/Const -LC:/Strawberry/perl/site/lib 
-LC:/Strawberry/perl/vendor/lib -LC:/Strawberry/perl/lib Const)

 checking in C:/Strawberry/perl/site/lib/auto/ModPerl/Const for Const.xs.dll
 checking in C:/Strawberry/perl/site/lib/auto/ModPerl/Const for Const.dll
dl_findfile found: C:/Strawberry/perl/site/lib/auto/ModPerl/Const/Const.dll
Can't load 'C:/Strawberry/perl/site/lib/auto/ModPerl/Const/Const.dll' for module 
ModPerl::Const: load_file:Das angegebene Modul wurde nicht gefunden at 
C:/Strawberry/perl/lib/DynaLoader.pm line 193.

  at C:/Strawberry/perl/site/lib/Apache2/Const.pm line 19.
Compilation failed in require at C:/Strawberry/perl/site/lib/Apache2/Const.pm 
line 19.
BEGIN failed--compilation aborted at 
C:/Strawberry/perl/site/lib/Apache2/Const.pm line 19.

(Note : to make extra sure, I also ran this in a CMD window "run as administrator"; the 
results are the same.)



Any other clue, anyone ?

>perl -V :

 C:\strawberry\perl\site>perl -V
Summary of my perl5 (revision 5 version 26 subversion 3) configuration:

  Platform:
osname=MSWin32
osvers=10.0.17134.407
archname=MSWin32-x64-multi-thread
uname='Win32 strawberry-perl 5.26.3.1 #1 Sun Dec  2 22:57:44 2018 x64'
config_args='undef'
hint=recommended
useposix=true
d_sigaction=undef
useithreads=define
usemultiplicity=define
use64bitint=define
use64bitall=undef
uselongdouble=undef
usemymalloc=n
default_inc_excludes_dot=define
bincompat5005=undef
  Compiler:
cc='gcc'
ccflags =' -s -O2 -DWIN32 -DWIN64 -DCONSERVATIVE -D__USE_MINGW_ANSI_STDIO 
-DPERL_TEXTMODE_SCRIPTS -DPERL_IMPLICIT_CONTEXT -DPERL_IMPLICIT_SYS -DUSE_PERLIO -fwrapv 
-fno-strict-aliasing -mms-bitfields'

optimize='-s -O2'
cppflags='-DWIN32'
ccversion=''
gccversion='7.1.0'
gccosandvers=''
intsize=4
longsize=4
ptrsize=8
doublesize=8
byteorder=12345678
doublekind=3
d_longlong=define
longlongsize=8
d_longdbl=define
longdblsize=16
longdblkind=3
ivtype='long long'
ivsize=8
nvtype='double'
nvsize=8
Off_t='long long'
lseeksize=8
alignbytes=8
prototype=define
  Linker and Libraries:
ld='g++'
ldflags ='-s -L"C:\STRAWB~1\perl\lib\CORE" -L"C:\STRAWB~1\c\lib"'
libpth=C:\STRAWB~1\c\lib C:\STRAWB~1\c\x86_64-w64-mingw32\lib 
C:\STRAWB~1\c\lib\gcc\x86_64-w64-mingw32\7.1.0
libs= -lmoldname -lkernel32 -luser32 -lgdi32 -lwinspool -lcomdlg32 -ladvapi32 
-lshell32 -lole32 -loleaut32 -lnetapi32 -luuid -lws2_32 -lmpr -lwinmm -lversion -lodbc32 
-lodbccp32 -lcomctl32
perllibs= -lmoldname -lkernel32 -luser32 -lgdi32 -lwinspool -lcomdlg32 -ladvapi32 
-lshell32 -lole32 -loleaut32 -lnetapi32 -luuid -lws2_32 -lmpr -lwinmm -lversion -lodbc32 
-lodbccp32 -lcomctl32

libc=
so=dll
useshrplib=true
libperl=libperl526.a
gnulibc_version=''
  Dynamic Linking:
dlsrc=dl_win32.xs
dlext=xs.dll
d_dlsymun=undef
ccdlflags=' '
cccdlflags=' '
lddlflags='-mdll -s -L"C:\STRAWB~1\perl\lib\CORE" -L"C:\STRAWB~1\c\lib"'


Characteristics of this binary (from libperl):
  Compile-time options:
HAS_TIMES
HAVE_INTERP_INTERN
MULTIPLICITY
PERLIO_LAYERS
PERL_COPY_ON_WRITE
PERL_DONT_CREATE_GVSV
PERL_IMPLICIT_CONTEXT
PERL_IMPLICIT_SYS
PERL_MALLOC_WRAP
PERL_OP_PARENT
PERL_PRESERVE_IVUV
USE_64_BIT_INT
USE_ITHREADS
USE_LARGE_FILES
    USE_LOCALE
USE_LOCALE_COLLATE
    USE_LOCALE_CTYPE
    USE_LOCALE_NUMERIC
USE_LOCALE_TIME
    USE_PERLIO
USE_PERL_ATOF
  Built under MSWin32
  Compiled at Dec  2 2018 23:03:33
  %ENV:
PERL_DL_DEBUG="1"
  @INC:
C:/Strawberry/perl/site/lib
C:/Strawberry/perl/vendor/lib
C:/Strawberry/perl/lib

C:\strawberry\perl\site>

Environment :
C:\

Re: POST large amount of data causes segmentation fault

2020-02-10 Thread tomcat/perl


On 09.02.2020 16:26, Vincent Veyron wrote:

On Sun, 9 Feb 2020 10:57:04 +
Scott Alexander  wrote:


I meant I'm posting data from a form, javascript is collecting a series of numbers 
from a  as a string.


Not really clear what happens here? I guess Javascript writes the content of 
the text input?


Javascript seperates each number with a '/' then I use split within Perl to get 
the list of numbers into an array.



I have a  with list of users and GSM numbers. If that list is about 600 
users it has worked fine. I was testing with about 4000 users when I came across this 
problem.



I just tried the line of code you posted earlier (@number = split /\//, 
$req->param("numbers")) on a form that I use, it processes a text input of 50 
000 ten digit numbers separated by a slash with no problem.

I use Apache/2.4.25 (Debian), libapache2-mod-per 2.0.10-2+deb9u and 
libapache2-request-perl 2.13-7~deb9u1



Might it not have something to do with the *format* used for the POST (from javascript / 
). I mean

Content-type : application/x-www-form-urlencoded
or
Content-type : multipart/form-data
or even
Content-type : text/plain

The parser for that data is certainly different on the server side, and one may be running 
against different limits/quirks.


of course a segfault is not nice anyway..

Re (e.g.9 . 
https://dev.to/sidthesloth92/understanding-html-form-encoding-url-encoded-and-multipart-forms-3lpa

Re: Building mod_perl2.pm

2020-02-13 Thread tomcat/perl


On 12.02.2020 23:01, Ruben Safir wrote:

On Tue, Feb 11, 2020 at 07:04:56PM -0600, Steven Lembark wrote:



Hello,
I am trying to use Apache::DBI module. I got it installed on the SUSE
Linux 11SP4 on my dev box.

It has a dependency of mod_perl2. But when I try to install
mod_perl2, I get

Please repair your Module::CoreList at lib/Apache2/Build.pm line 51.
BEGIN failed--compilation aborted at lib/Apache2/Build.pm line 66.
Compilation failed in require at Makefile.PL line 39.
BEGIN failed--compilation aborted at Makefile.PL line 39.

Going by the code in Build.pm, I think the version between Corelist
and Embed is not matching. But the same is installed on our test
servers and I am not sure what is different. Probably it has
Apache::DBI for Apache 1.x installed? How do I check? and any help to
resolve the above error?


You will probably find that maintaining your own copy of Perl is
less work than depending on the pre-compiled [mess] in most
distro's is less work. Building Perl takes little enough time: just
install it into /opt/perl/ via:

 ./Configure --prefix=/opt/perl -de;



This is really true and this is really a problem.  It would be nice to
get some alternate packages made for the larger distros



Just a comment here : we have been installing and using mod_perl2 for many years on many 
servers with Linux Debian, and installing it using the normal Debian package manager 
commands, without any problems that I can recall.


e.g.

root@kunz:~# uname -a
Linux kunz 4.9.0-11-amd64 #1 SMP Debian 4.9.189-3+deb9u2 (2019-11-11) x86_64 
GNU/Linux

root@kunz:~# apt-cache search mod_perl
[ .. long list, among which ..]

libapache2-mod-perl2 - Integration of perl with the Apache2 web server
libapache2-mod-perl2-dev - Integration of perl with the Apache2 web server - 
development files
libapache2-mod-perl2-doc - Integration of perl with the Apache2 web server - 
documentation

root@kunz:~# apt-get install libapache2-mod-perl2
Reading package lists... Done
Building dependency tree
Reading state information... Done
libapache2-mod-perl2 is already the newest version (2.0.10-2+deb9u1).
0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
root@kunz:~#

So I guess that your issue above does indeed have to do with the specific SUSE packages, 
and you should probably discuss that with them.











(you can see examples of building it in some of my talks on
slideshare.net if you like). After that:

 cd /opt/perl;
 rm *;
 ln -fs /* .;

 [ -d /opt/bin ] || mkdir /opt/bin;
 cd /opt/bin;
 ln -fs ../perl/bin/* .;

and add "/opt/bin" to your path. After that "cpan FooBar" will
install any modules you require, probably with less hassle than
working around O/S issues with library versions.

You could spend a week or so banging your head out deciphering
whatever bone-headed assumptions the distro builders made or
just get it right the first time yourself.

The process may look daunting the first time around but after
a while you'll notice it takes only a few of your favorite
beverages a month to maintain and induces far less annoyance...

Enjoy

--
Steven Lembark3646 Flora Place
Workhorse ComputingSt. Louis, MO 63110
lemb...@wrkhors.com+1 888 359 3508

Apache 2.4, AAA, Shibboleth


Hi experts.

In our Apache 2.4(+mod_perl) setups, we use the following kind of thing :

...
PerlAddAuthzProvider UMA-user AUTH::UMA2->authz_user
...

  AuthName ALUtop
  AuthType shibboleth
  PerlSetVar UMA_AuthType "SAML2"
  ShibRequestSetting requireSession 1
  ShibRequestMapperAuthz Off
  ShibUseHeaders On
  ShibUseEnvironment Off
  PerlSetVar UMA_Debug 3
  
Require UMA-user valid-user
Require shibboleth
  
...


which basically means :
we combine 2 AAA methods :
- one is our own (AUTH::UMA2 above, mod_perl based)
- the other is SAML, via Shibboleth (and all the directives above with "shib" in them, 
correspond to the Shibboleth installation instructions)


(The reason being : the corporate user authentication happens via SAML, but it does not 
provide us enough information about the user. So we run another additional scheme, which 
supplements the user information, in order to feed more complete data to our applications.)


Our own AAA method is such, that once the user has been authenticated once, we set a 
token, such that for subsequent requests we do not need to re-authenticate the user.

But no matter what we do at that level, the Shibboleth authentication runs 
anyway.(*)

And my question is : considering the above setup, would mod_perl provide a way, through 
some mod_perl API, to disable the Shibboleth authentication (on a request-by-request 
base), when our own authz module determines that we do not need it to run anymore for the 
current request ?


(replace Shibboleth by any other authentication that would be configured in addition to 
our own; I'm looking for some "generic" mechanism, not only for Shibboleth).


Or is it so that different authentication methods/modules don't insert themselves in any 
standard way which can easily be interfaced with in order to dynamically disable them ?


Note: Shibboleth itself does caching of its prior authentication, and it is not really a 
big performance hit to re-run it each time, and we can live with it as it is. But in the 
absolute, it is unnecessary for 90% of the accesses to the applications, so it just sounds 
like disabling it would be a nice/efficient thing to do.


I thought of dynamically removing the "Require shibboleth" e.g., but there does not seem 
to be an API to do that.

Re: Apache 2.4, AAA, Shibboleth

On 06.03.2020 17:18, Russell Lundberg wrote:

Andy, could your custom auth handler run before Shiboleth, test for a Shiboleth
token? If not present, use push_handler to run Shiboleth? If the Shiboleth
token or cookie is present, don’t run Shiboleth?

Well, the problem is that Shibboleth is quite "invasive" (no complaint here, it works that
way it is supposed to and documented), in the sense that there are quite a few directives
for it (as you can see below), and I am not quite sure if I can configure a section "with
Shibboleth but not its handler", in order to only push the handler later.
For example, I have the feeling that just indicating "AuthType Shibboleth" (which is
required) already puts the handler in the chain.

On the other hand, if I could just insert the "require shibbolet"
dynamically (instead of in the static config), that might do it.
But there seems to be no mod_perl API to add a "require" on-the-fly. "Requires" are just a
list that the AUTHZ handler gets when called, and apparently each such handler only gets
its own.
For example, the "Require UMA-user valid-user" below, results in our AUTHZ function
"AUTH::UMA2->authz_user" being called, but in the list of requires we just get (our own)
"valid-user".

But thanks for the suggestion. It /is/ the kind of thing I'm looking for.
Only maybe the opposite : how to "pop" the Shibboleth handler when it's there, and just
for the current request.

Russell

Sent from my iPhone

On Mar 6, 2020, at 08:26, André Warnier (tomcat/perl) wrote:

Hi experts.

In our Apache 2.4(+mod_perl) setups, we use the following kind of thing :

...
PerlAddAuthzProvider UMA-user AUTH::UMA2->authz_user
...

AuthName ALUtop
AuthType shibboleth
PerlSetVar UMA_AuthType "SAML2"
ShibRequestSetting requireSession 1
ShibRequestMapperAuthz Off
ShibUseHeaders On
ShibUseEnvironment Off
PerlSetVar UMA_Debug 3

Require UMA-user valid-user
Require shibboleth

...

which basically means :
we combine 2 AAA methods :
- one is our own (AUTH::UMA2 above, mod_perl based)
- the other is SAML, via Shibboleth (and all the directives above with "shib"
in them, correspond to the Shibboleth installation instructions)

(The reason being : the corporate user authentication happens via SAML, but it
does not provide us enough information about the user. So we run another
additional scheme, which supplements the user information, in order to feed
more complete data to our applications.)

Our own AAA method is such, that once the user has been authenticated once, we
set a token, such that for subsequent requests we do not need to
re-authenticate the user.
But no matter what we do at that level, the Shibboleth authentication runs
anyway.(*)

And my question is : considering the above setup, would mod_perl provide a way,
through some mod_perl API, to disable the Shibboleth authentication (on a
request-by-request base), when our own authz module determines that we do not
need it to run anymore for the current request ?

(replace Shibboleth by any other authentication that would be configured in addition to
our own; I'm looking for some "generic" mechanism, not only for Shibboleth).

Or is it so that different authentication methods/modules don't insert
themselves in any standard way which can easily be interfaced with in order to
dynamically disable them ?

Note: Shibboleth itself does caching of its prior authentication, and it is not
really a big performance hit to re-run it each time, and we can live with it as
it is. But in the absolute, it is unnecessary for 90% of the accesses to the
applications, so it just sounds like disabling it would be a nice/efficient
thing to do.

I thought of dynamically removing the "Require shibboleth" e.g., but there does
not seem to be an API to do that.

Re: Apache 2.4, AAA, Shibboleth


Genius !
Yes, I'll try that.
Why did I not think of that myself ?
If anything, it'll be fun to watch the log lines of authz_core and try to make 
sense of them.
I don't even think that I have to extend our module, it should work just as well with 
"require UMA-user valid-user" in both cases.


One more question : why the top-level  ?

On 07.03.2020 00:59, Paul B. Henson wrote:

On Sat, Mar 07, 2020 at 12:22:35AM +0100, André Warnier (tomcat/perl) wrote:


Does anyone have an idea how I could do the equivalent of a "pop the
'require shibboleth' but just for this request" ?


Would you be willing to extend your local custom module to achieve this?

You could do something like:



Require UMA-token

Require shibboleth
Require UMA-user valid-user




Your custom module would get called first, which would check to see if
your custom session token was valid. If so, it would return success and
nothing else would be processed. Otherwise, the shibboleth module would
run, then your custom module which would establish your session token so
on the next access the shib auth would be bypassed.


ShibRequestMapperAuthz Off
ShibUseHeaders On
ShibUseEnvironment Off
PerlSetVar UMA_Debug 3

  Require UMA-user valid-user
  Require shibboleth

...


which basically means :
we combine 2 AAA methods :
- one is our own (AUTH::UMA2 above, mod_perl based)
- the other is SAML, via Shibboleth (and all the directives above with
"shib" in them, correspond to the Shibboleth installation instructions)

(The reason being : the corporate user authentication happens via SAML, but
it does not provide us enough information about the user. So we run another
additional scheme, which supplements the user information, in order to feed
more complete data to our applications.)

Our own AAA method is such, that once the user has been authenticated once,
we set a token, such that for subsequent requests we do not need to
re-authenticate the user.
But no matter what we do at that level, the Shibboleth authentication runs 
anyway.(*)

And my question is : considering the above setup, would mod_perl provide a
way, through some mod_perl API, to disable the Shibboleth authentication (on
a request-by-request base), when our own authz module determines that we do
not need it to run anymore for the current request ?

(replace Shibboleth by any other authentication that would be configured in
addition to our own; I'm looking for some "generic" mechanism, not only for
Shibboleth).

Or is it so that different authentication methods/modules don't insert
themselves in any standard way which can easily be interfaced with in order
to dynamically disable them ?

Note: Shibboleth itself does caching of its prior authentication, and it is
not really a big performance hit to re-run it each time, and we can live
with it as it is. But in the absolute, it is unnecessary for 90% of the
accesses to the applications, so it just sounds like disabling it would be a
nice/efficient thing to do.

I thought of dynamically removing the "Require shibboleth" e.g., but there
does not seem to be an API to do that.

Re: Apache 2.4, AAA, Shibboleth