Re: Apache : mod_perl vs fastcgi
Dana Powers [EMAIL PROTECTED] writes: FastCGI with perl may or may not be that fast. I could be totally wrong here but I believe that FastCGI will be able to cache the perl runtime, but perl will still have to reload and recompile any scripts during every hit. Not true. You put some code in your CGI to "get next CGI request". Your fully compiled program does this for as many times as it needs to. See the paper linked off my homepage *plug* *plug* ;-) -- Dave Hodgkinson, http://www.hodgkinson.org Editor-in-chief, The Highway Star http://www.deep-purple.com Apache, mod_perl, MySQL, Sybase hired gun for, well, hire -
Re: What is *.xs file?
James G Smith wrote: I would start with `man h2xs' (only a start), and get _Programming Perl_, which talks about it some, as does _The Perl Cookbook_, I believe. Then Advanced Perl programming talks about it also. -- - frankie -
Re: Hang in $r-print w/ POST
"Darren" == darren chamberlain [EMAIL PROTECTED] writes: Darren Apache::Request retrieves the POSTed content, so if you try to get Darren it, the server will hang, waiting for input. With Apache::Request, Darren just call the Darren $r-param method to fetch the data, and don't try to call it yourself. Yes I know that, I didn't try to read something (POST data from stdin) I try to write (print to stdout). BTW my code act the same if I use Apache::Request or not... And I forgot to say that I use CGI module in my handler, and that no other handler eat r-content ... Apache v1.3.9 mod_perl v1.21 CGI.pm v2.62 This piece of code illustrate the pb (thanks to give me any hint): #!/usr/local/bin/perl # # Test.pm: Apache::CENA::Test # package Apache::CENA::Test; use diagnostics -verbose; use strict; use Apache::Constants qw(:common); use CGI qw(:html :form :cgi); sub handler { my $r = shift; # Do not bother with HEAD requests return DECLINED if $r-header_only(); # Do not bother with internal sub-requests return DECLINED unless $r-is_main(); # Get POST content GET args $r-log-info( "Getting args ... " ); my %Args = ($r-args(), $r-content()); my $this = $r-subprocess_env('SCRIPT_URL'); $r-log-info( "Sending headers ... " ); $r-content_type( 'text/html' ); $r-send_http_header(); $r-log-info( "Sending start_html ... " ); # Hangs here in POST methods!!! $r-print( start_html(-title= $this )); $r-print( h1( $this ), hr()); map{ $r-print( "$_ = ", strong($Args{$_}), br());} keys %Args; $r-print( hr(), a({-href= "$this?get-arg1=1get-arg2=2"}, 'GET' ), start_form(-method = 'POST', -action = "$this?get-arg3=3get-arg4=4"), hidden( -name = 'post-content1', -default = 5 ), hidden( -name = 'post-content2', -default = 5 ), submit( 'POST', 'POST' ), end_form(), end_html() ); $r-log-info( "Done" ); return OK; } 1; # Test.pm ends here -- / From the last station before the end of the neT \ \ Centre d'Etudes de la Navigation Aerienne, div SSS / / 7, av edouard Belin - 31055 Toulouse CEDEX - France \ \ Tel:(+33|0)5.62.25.95.22 | Fax:(+33|0)5.62.25.95.99 /
Re: Apache : mod_perl vs fastcgi
In addition, you can read the CGI.pm documentation on the use of Perl/CGI.pm with FastCGI and you'll be good to go with that get next request technique. At 08:14 AM 7/7/00 +0100, David Hodgkinson wrote: Dana Powers [EMAIL PROTECTED] writes: FastCGI with perl may or may not be that fast. I could be totally wrong here but I believe that FastCGI will be able to cache the perl runtime, but perl will still have to reload and recompile any scripts during every hit. Not true. You put some code in your CGI to "get next CGI request". Your fully compiled program does this for as many times as it needs to. See the paper linked off my homepage *plug* *plug* ;-) -- Dave Hodgkinson, http://www.hodgkinson.org Editor-in-chief, The Highway Star http://www.deep-purple.com Apache, mod_perl, MySQL, Sybase hired gun for, well, hire -
Re: Hang in $r-print w/ POST
heddy Boubaker ([EMAIL PROTECTED]) said something to this effect: "Darren" == darren chamberlain [EMAIL PROTECTED] writes: Darren Apache::Request retrieves the POSTed content, so if you try to get Darren it, the server will hang, waiting for input. With Apache::Request, Darren just call the Darren $r-param method to fetch the data, and don't try to call it yourself. Yes I know that, I didn't try to read something (POST data from stdin) I try to write (print to stdout). BTW my code act the same if I use Apache::Request or not... And I forgot to say that I use CGI module in my handler, and that no other handler eat r-content ... But using CGI's methods invoke CGI's self_or_default meethod, which will either use the object passed in or create one internally ($Q) and use that to perform its output. So, even though you are not explicitly creating a CGI object, you are creating one implicitly, which is what is grabbing the POSTed data, which is why the script hangs right where you first try to use a CGI method. If you need the HTML generation methods (it looks like this is all you use CGI for), look into HTML::StickyForm for the forms, and extend it with your own package for other HTML elements, or use a templating system like HTML::Template or TemplateToolkit. I prefer to use HTML::Template, or HTML::Mason when possible, for exactly this reason. Another option, of course, is to use CGI in an object oriented fashion, and preload it with -compile to have the entirety of it reside in the parent process. Recommendation: Abondon CGI, and go with Apache::Request (or Apache::RequestNote), HTML::StickyForm, and HTML::Template. (darren) -- The goal of technology is to build something that will last at least until we finish building it.
Re: AIX, dlopen problem?
On Thu, Jul 06, 2000 at 07:08:03PM -0500, John Marquart wrote: Environment: AIX 4.3.3 xlc apache 1.3.12 mod_perl 1.22 perl 5.00503 (compiled w/ INSTALL patch, details below) I have had various difficulties trying to get mod_perl to compile successfully using perl 5.6.0, also w/ mod_perl 1.24. However, with the above configuration I have had no compilation problems. When I try to include a module in httpd.conf w/ a "PerlModule Apache::IU::AuthenKerberos" statement then run apachectl start I get the following error message: bash-2.01# ./apachectl start Syntax error on line 191 of /usr/local/apache_perl.4/conf/httpd.conf: Can't load '/usr/local/lib/perl5/site_perl/5.005/aix/auto/Apache/IU/AuthenKerberos/AuthenKerberos.so' for module Apache::IU::AuthenKerberos: dlopen: /usr/local/lib/perl5/site_perl/5.005/aix/auto/Apache/IU/AuthenKerberos/AuthenKerberos.so: 30 /usr/local/lib/perl5/site_perl/5.005/aix/auto/Apache/IU/AuthenKerberos/AuthenKerberos.so36 ap_add_module 20 httpd36 ap_null_cleanup 21 httpd36 ap_palloc 22 httpd36 ap_register_cleanup 23 httpd at /usr/local/lib/perl5/5.00503/aix/DynaLoader.pm line 169. This suspicially looks like you have some imports in the AuthenKerberos wrong. Modules that are plug-ins to Perl but at the same time need to reference Apache symbols need some more ld flags to compile. The Apache::src modules knows which: $ perl -MApache::src -le 'print Apache::src-new-otherldflags;' -bI:/usr/local/lib/perl5/site_perl/5.005/aix/auto/Apache/mod_perl.exp -bI:/usr/local/apache/libexec/httpd.exp Add these to the link line of the AuthenKerberos module. If that still does not work, I would need to examine the exact link command line and the output of "dump -nv AuthenKerberos.so". -- Jens-Uwe Mager HELIOS Software GmbH Steinriede 3 30827 Garbsen Germany Phone: +49 5131 709320 FAX:+49 5131 709325 Internet: [EMAIL PROTECTED]
Re: Hang in $r-print w/ POST
"Darren" == darren chamberlain [EMAIL PROTECTED] writes: Darren But using CGI's methods invoke CGI's self_or_default meethod, which Darren will either use the object passed in or create one internally ($Q) Darren and use that to perform its output. So, even though you are not Darren explicitly creating a CGI object, you are creating one implicitly, Darren which is what is grabbing the POSTed data, which is why the script Darren hangs right where you first try to use a CGI method. Agh! You'r right this is THE reason !! the CGI-init method eat all the POST data !! Is this documented anywhere ? Couldn't this be in the mod_perl Guide or FAQ ? Darren If you need the HTML generation methods (it looks like this is all Darren you use CGI for), look into HTML::StickyForm for the forms, and Darren extend it with your own package for other HTML elements, or use a Darren templating system like HTML::Template or TemplateToolkit. I prefer to Darren use HTML::Template, or HTML::Mason when possible, for exactly this Darren reason. Another option, of course, is to use CGI in an object Darren oriented fashion, and preload it with -compile to have the entirety Darren of it reside in the parent process. I liked the CGI functions to generate HTML ... it seems there is no simple way to tell CGI module not to eat POST data. HTML::StickyForms could be an option for forms ... the others (Template, Mason) are not for what I'm trying to do. I'm just looking for a simple but complete html generator (instead of hard coding all html tags in my code), does such a beast exists or have I to write my own? thanks a lot for you answers. -- / From the last station before the end of the neT \ \ Centre d'Etudes de la Navigation Aerienne, div SSS / / 7, av edouard Belin - 31055 Toulouse CEDEX - France \ \ Tel:(+33|0)5.62.25.95.22 | Fax:(+33|0)5.62.25.95.99 /
New mod_perl site goes live...
Well, the site's old, but the mod_perl bit of it is brand spanking new. Thanks to sterling work by Ged Haywood and a little bit of tweaking by me: http://www.bargainholidays.com/ Is now mod_perl. Well, the most important script on the site it. Response times are now roughly ten times better than they were, but the most impressive thing is that the system load now looks like a dead person's cardiogram as opposed to that of someone in bad ventricular fibrillation. It ate today's lunchtime peak without breaking a sweat. There are still a couple of CPU wastes in the perl itself but I'll fix those in due course. Thanks for a great tool! -- Dave Hodgkinson, http://www.hodgkinson.org Editor-in-chief, The Highway Star http://www.deep-purple.com Apache, mod_perl, MySQL, Sybase hired gun for, well, hire -
Re: Apache/mod_perl
"PS" == Pramod Sokke [EMAIL PROTECTED] writes: PS We are running Netscape Enterprise server with cgis written in perl and C. PS I'm looking at moving over to Apache and start using mod_perl. How [ .. ] PS over to Apache/mod_perl going to be a simple plug-in or would it involve PS re-writing lots of stuff? The C stuff will probably not be worth rewriting, but that depends on what it does. The perl stuff will need to be "cleaned" if it is sloppy code. That is, if it is clean running in Perl under "-w" and "use strict" you're most likely going to have little difficulty with them. But what you should do is use the two-server performance enhancement (using mod_proxy and mod_rewrite) and have your legacy apps run on the front-end server, and then migrate your perl to the mod_perl backend one at a time. -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Vivek Khera, Ph.D.Khera Communications, Inc. Internet: [EMAIL PROTECTED] Rockville, MD +1-301-545-6996 GPG MIME spoken herehttp://www.khera.org/~vivek/
Re: Apache/mod_perl
Vivek Khera [EMAIL PROTECTED] writes: "PS" == Pramod Sokke [EMAIL PROTECTED] writes: PS We are running Netscape Enterprise server with cgis written in perl and C. PS I'm looking at moving over to Apache and start using mod_perl. How [ .. ] PS over to Apache/mod_perl going to be a simple plug-in or would it involve PS re-writing lots of stuff? The C stuff will probably not be worth rewriting, but that depends on what it does. The perl stuff will need to be "cleaned" if it is sloppy code. That is, if it is clean running in Perl under "-w" and "use strict" you're most likely going to have little difficulty with them. But what you should do is use the two-server performance enhancement (using mod_proxy and mod_rewrite) and have your legacy apps run on the front-end server, and then migrate your perl to the mod_perl backend one at a time. My first thought would be to do some 80/20 analysis and see which bits are REALLY important. If it's the C code, do the fastcgi thing (or Velocigen if you want actual support). If it's the perl code, then as Vivek says. -- Dave Hodgkinson, http://www.hodgkinson.org Editor-in-chief, The Highway Star http://www.deep-purple.com Apache, mod_perl, MySQL, Sybase hired gun for, well, hire -
Re: Apache::ASP and clock times
We are not sure why but it appears that the time problem may end up being specific to a DSO enabled, debian distribution of apache. When we recompiled a static version, the time that the ASP processes reported remained the same as the system clock (no changes from PDT to GMT). It looks like this is a bad-compilation problem or an effect of DSO. On Fri, 7 Jul 2000, Vivek Khera wrote: "JC" == Joshua Chamas [EMAIL PROTECTED] writes: JC It may be possible that the time calculation is based JC on some TZ* environment variable that is not consistent, It has been my experience that the first time any of the system date/time routines are called and TZ is set, that becomes the permanent time zone until the process goes away. So it could be something totally other than ASP setting the time zone for the process.
Re: Apache::ASP
I just meant, if you build/rebuild your modperl httpd, make sure you copy it to your normal httpd place, like /usr/local/apache/bin or something. People often forget this step so that even when they build it and passes all the tests, they still get the PerlSetVar not defined error from apache, since they are using their old apache binary still. Check out a related documented FAQ for more info: http://www.nodeworks.com/asp/faq.html#Apache%20errors%20on%20the%20PerlHandler%20directive%20%3F --Joshua Vincent Bruijnes wrote: How do you mean with copy it to over to where my httpd resides? my apache resides in /www and where and what file should i place in /www/ ? btw i don't use DSO Sincerely Vincent Bruijnes At 00:57 6-7-00 -0700, you wrote: Vincent Bruijnes wrote: I got this error always when running /site/eg/index.html /www/htdocs/site/eg/.htaccess: Invalid command 'PerlSetVar', perhaps mis-spelled or defined by a module not included in the server configuration. Is there a clue ? Install mod_perl, or make sure its active. If using DSO, make sure the module is built and commented in. If you haven't built mod_perl, build it, statically link it with Apache, and make sure you copy it over to where your httpd resides on disk ( common error ). If you have any questions about the above, be sure to read the guide http://perl.apache.org/guide -- Joshua
Re: R. Berjon's Mod_perl site (was Re: Coldfusion vs. apache/mod_perl)
On Fri, 7 Jul 2000, Robin Berjon wrote: No need to apologize, we've all got a lot to do. Is it just a matter of committing the cvs or is there more to be done ? I have a little bit of time between now and the middle of the month, I can help if need be. jim took a snapshot of a subset of the site and applied his changes to it. i need to incorporate changes that have been made since the snapshot, and apply the template to the rest of the site. it'll be a fair amount of work, but perhaps i can do it in chunks, and i'll certainly commit the template, so perhaps folks can help with that part.
Re: Hang in $r-print w/ POST
"heddy" == heddy Boubaker [EMAIL PROTECTED] writes: I liked the CGI functions to generate HTML ... it seems there is no simple way to tell CGI module not to eat POST data. HTML::StickyForms could be an option for forms ... the others (Template, Mason) are not for what I'm trying to do. I'm just looking for a simple but complete html generator (instead of hard coding all html tags in my code), does such a beast exists or have I to write my own? From the CGI docs: To create an empty query, initialize it from an empty string or hash: $empty_query = new CGI(""); -or- $empty_query = new CGI({}); This gives you a CGI object that you can use to generate your html and leave the query content alone. HTHYO. Peace.
Re: Apache::ASP and post-POST redirect
Uh, I found what it was! After receiving Joshua's email, I experimented some more and while trying to get the simplest possible configuration disabled the 'Filter' flag (which I had turned on because I used Apache::SSI in conjunction with ASP). And this did the trick. As soon as filtering was gone, redirects started working again. Hmm, I don't think I ever mentioned that ASP post-POST redirects just all of a sudden stopped working for me. A browser would send a POST request and never get anything back. The logs showed that the request was processed to the point of $Response-Redirect, but nothing would ever come out from the other side. The server was simply closing the connection without sending any data back to the client. I did find a hack that made redirects work by replacing ASP's Redirect with the following lines: Apache-print("HTTP/1.0 301 Moved Permanantly\n"); Apache-print("Location: http://www.ucsd.edu\n"); Apache-print("\n"); $Response-Redirect('http://mill.ucsd.edu/index.html'); (where the last line was simply used to force apache to close the connection) This worked as expected. I swear this used to work before with filters on. Did something change in Apache::Filter? Thanks Dmitry PS I'm running Apache/1.3.12 and mod_perl/1.24 (all hand-compiled) on RedHat 6.2. Apache::ASP v.0.19 (I did try the newest version, it behaved the same), Apache::Filter v.1.011 and Apache::SSI v.2.13 At 08:26 AM 7/7/00, Joshua Chamas wrote: I imagine that if you just do: $Response-Clear(); $Response-Redirect(); you will get what you are going for. A POST should not follow a redirect. A redirect at the top of your scripts should likely not need the clear, which is how I tend to use it. Note with the latest release there is also a $Server-Transfer() which is faster than a redirect, and maybe useful for your needs. [...] Dmitry Beransky wrote: [...] In a mod_perl module, if I want to return a redirect after processing a POST, I need to make sure to reset the method to 'GET' and content length to 0. How does Apache::ASP handle this case or, rather, how do I handle after POST redirects in ASP?
Re: Apache::ASP and post-POST redirect
Dmitry Beransky wrote: Uh, I found what it was! After receiving Joshua's email, I experimented some more and while trying to get the simplest possible configuration disabled the 'Filter' flag (which I had turned on because I used Apache::SSI in conjunction with ASP). And this did the trick. As soon as filtering was gone, redirects started working again. Hmm, I don't think I ever mentioned that ASP post-POST redirects just all of a sudden stopped working for me. A browser would send a POST request and never get anything back. The logs showed that the request was processed to the point of $Response-Redirect, but nothing would ever come out from the other side. The server was simply closing the connection without sending any data back to the client. I did find a hack that made redirects work by replacing ASP's Redirect with the following lines: Apache-print("HTTP/1.0 301 Moved Permanantly\n"); Apache-print("Location: http://www.ucsd.edu\n"); Apache-print("\n"); $Response-Redirect('http://mill.ucsd.edu/index.html'); (where the last line was simply used to force apache to close the connection) This worked as expected. I bet its an ASP-Apache::Filter issue, because in general there has been quite a lot with of issues coordinating between filtered modules on the headers. Note that in recent versions of Apache::ASP, I have started to return the 302 status for redirects from the handler which might be affecting this behavior. Also the new $Server-Transfer() feature might make this issue go away for you too, as it is an internal redirect. Until I can look at what's causing this, you might want to just override the *Apache::ASP::Response::Redirect sub in your global.asa or so, so you don't have to directly hacking Apache::ASP sub Apache::ASP::Response::Redirect { my($self, $location) = @_; Apache-print("HTTP/1.0 301 Moved Permanantly\n"); Apache-print("Location: $location\n"); Apache-print("\n"); $self-End(); } --Joshua At 08:26 AM 7/7/00, Joshua Chamas wrote: I imagine that if you just do: $Response-Clear(); $Response-Redirect(); you will get what you are going for. A POST should not follow a redirect. A redirect at the top of your scripts should likely not need the clear, which is how I tend to use it. Note with the latest release there is also a $Server-Transfer() which is faster than a redirect, and maybe useful for your needs. [...] Dmitry Beransky wrote: [...] In a mod_perl module, if I want to return a redirect after processing a POST, I need to make sure to reset the method to 'GET' and content length to 0. How does Apache::ASP handle this case or, rather, how do I handle after POST redirects in ASP?
Re: Apache::Session::Object
Perrin- modifying Apache::Session to support both interfaces and sending Jeffrey the patch. This is a good suggestion. I'll try modifying Apache::Session first and sending Jeff the patch. If he doesn't want to integrate it I'll package it as a separate module. -Nate
RE: Apache::ASP and post-POST redirect
-Original Message- From: Joshua Chamas [mailto:[EMAIL PROTECTED]] Sent: Friday, July 07, 2000 12:46 PM To: Dmitry Beransky Cc: [EMAIL PROTECTED]; Ken Williams Subject: Re: Apache::ASP and post-POST redirect I bet its an ASP-Apache::Filter issue, because in general there has been quite a lot with of issues coordinating between filtered modules on the headers. I haven't really been following this thread, since I don't use asp, but Apache::Filter (1.09?) introduced a new mechanism for sending the headers. basically, it waits to see the first print to STDOUT from the last filter before calling $r-send_http_headers(), whereas previously it was sending headers right at the call to $r-filter_input by the last filter. I don't see how this would make a difference to anyone (except to the positive), though, but I thought I'd bring it up just in case... HTH --Geoff
Re: Apache/mod_perl
Has anybody used stronghold? I'm considering using stronghold for SSL support since ours is a commercial application. Would mod_perl and all related modules work as fine with Stronghold as with plain Apache? Thanks, Pramod At 10:24 AM 7/7/00 -0400, Vivek Khera wrote: "PS" == Pramod Sokke [EMAIL PROTECTED] writes: PS We are running Netscape Enterprise server with cgis written in perl and C. PS I'm looking at moving over to Apache and start using mod_perl. How [ .. ] PS over to Apache/mod_perl going to be a simple plug-in or would it involve PS re-writing lots of stuff? The C stuff will probably not be worth rewriting, but that depends on what it does. The perl stuff will need to be "cleaned" if it is sloppy code. That is, if it is clean running in Perl under "-w" and "use strict" you're most likely going to have little difficulty with them. But what you should do is use the two-server performance enhancement (using mod_proxy and mod_rewrite) and have your legacy apps run on the front-end server, and then migrate your perl to the mod_perl backend one at a time. -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Vivek Khera, Ph.D.Khera Communications, Inc. Internet: [EMAIL PROTECTED] Rockville, MD +1-301-545-6996 GPG MIME spoken herehttp://www.khera.org/~vivek/
Re: Apache/mod_perl
Can't speak much to Stronghold, got a copy and ultimately put it on the shelf. I rather like Raven (covalent.com), works like a peach with mod_perl etc. Just wish Apache didn't have to be patched to install the damn stuff... -Alan -Original Message- From: Pramod Sokke [EMAIL PROTECTED] To: Vivek Khera [EMAIL PROTECTED]; [EMAIL PROTECTED] [EMAIL PROTECTED] Date: Friday, July 07, 2000 11:03 AM Subject: Re: Apache/mod_perl Has anybody used stronghold? I'm considering using stronghold for SSL support since ours is a commercial application. Would mod_perl and all related modules work as fine with Stronghold as with plain Apache? Thanks, Pramod At 10:24 AM 7/7/00 -0400, Vivek Khera wrote: "PS" == Pramod Sokke [EMAIL PROTECTED] writes: PS We are running Netscape Enterprise server with cgis written in perl and C. PS I'm looking at moving over to Apache and start using mod_perl. How [ .. ] PS over to Apache/mod_perl going to be a simple plug-in or would it involve PS re-writing lots of stuff? The C stuff will probably not be worth rewriting, but that depends on what it does. The perl stuff will need to be "cleaned" if it is sloppy code. That is, if it is clean running in Perl under "-w" and "use strict" you're most likely going to have little difficulty with them. But what you should do is use the two-server performance enhancement (using mod_proxy and mod_rewrite) and have your legacy apps run on the front-end server, and then migrate your perl to the mod_perl backend one at a time. -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Vivek Khera, Ph.D.Khera Communications, Inc. Internet: [EMAIL PROTECTED] Rockville, MD +1-301-545-6996 GPG MIME spoken herehttp://www.khera.org/~vivek/
Re: Apache/mod_perl
--On Friday, July 07, 2000 11:09 AM -0700 Alan Sparks [EMAIL PROTECTED] wrote: Can't speak much to Stronghold, got a copy and ultimately put it on the shelf. I rather like Raven (covalent.com), works like a peach with mod_perl etc. Just wish Apache didn't have to be patched to install the damn stuff... -Alan If you can patiently wait till September 21 when the RSA patents expire covering the SSL algorithms, then you can use mod_ssl without fear of patent violations or lawsuits. In fact, I recollect reading somewhere (might even been on this list) that Stronghold is going to or is now using mod_ssl as the underlying Apache/SSL interface (along with openSSL for the RSA stuff). -- Rob _ _ _ _ __ _ _ _ _ /\_\_\_\_\/\_\ /\_\_\_\_\_\ /\/_/_/_/_/ /\/_/ \/_/_/_/_/_/ QUIDQUID LATINE DICTUM SIT, /\/_/__\/_/ __/\/_//\/_/ PROFUNDUM VIDITUR /\/_/_/_/_/ /\_\ /\/_//\/_/ /\/_/ \/_/ /\/_/_/\/_//\/_/ (Whatever is said in Latin \/_/ \/_/ \/_/_/_/_/ \/_/ appears profound) Rob Tanner McMinnville, Oregon [EMAIL PROTECTED]
More on the set_handlers() / push_handlers() bug...
Hello, again. Here's a quick summary of the problem: upon clearing the stack of handlers (with set_handlers(PerlAuthzHandlers = undef)) the server appears to not recognize when it has found a DirectoryIndex file and returns the contents of a directory instead of the index file. Since last writing on this topic, I upgraded to all of the newest versions (apache-1.3.12, perl-5.6, mod_perl-1.24) and configured mod_perl with PERL_TRACE=1. Looking at the error_log with the handler tracing on and httpd running in single-server mode, I discovered that the server does appear to cycle through the directory indexes, but doesn't realize when it has found one. I've attached two error logs for the request http://dw3.tivoli.com:81/home/passwd/group/: one with the cache enabled, one without. I've also attached a copy of the document returned by the server for the with-authz-cache request. The document is the auto-generated contents of the directory. As you can see, index.phtml does exist in the directory and is the first file searched for by the server. Does anyone have any ideas here? Can anyone reproduce the problem? I've also included the source for the AuthzCache and AuthzLDAP. If anyone could use at least the Cache code and see if you could reproduce the problem, I'd really appreciate it. I noticed that the error log with the cache enabled shows that subsequent internal requests (such as those you see when requesting a directory instead of a named file) don't reset the handler stack to what is in the configuration. See about line 65 of each log file for this difference. Regards, Christian - Christian Gilmore Infrastructure Tools Team Lead Web Multimedia Development Tivoli Systems, Inc. AuthzCache.pm AuthzLDAP.pm Title: Index of /home/passwd/group Index of /home/passwd/group NameLast modified Size Description Parent Directory18-May-2000 11:25 - display_group_info.p.. 07-Jun-2000 10:4511k group.phtml 08-Jun-2000 17:3818k index.phtml 05-Jun-2000 12:13 7k Apache/1.3.12 Server at dw3.tivoli.com Port 81 error_log_with_cache error_log_without_cache
Re: Apache/mod_perl
So if i didn't want to wait, how would I go about getting an RSA liscense? to have a mod_ssl server before then. Also, are the CA's willing to sign for this time of server? Scott On 7 Jul 2000, at 11:34, Rob Tanner wrote: If you can patiently wait till September 21 when the RSA patents expire covering the SSL algorithms, then you can use mod_ssl without fear of patent violations or lawsuits.
Re: Apache/mod_perl
Scott Alexander wrote: So if i didn't want to wait, how would I go about getting an RSA liscense? to have a mod_ssl server before then. Also, are the CA's willing to sign for this time of server? When I checked into this Raven SSL seemed to be the solution. It is basically mod_ssl with an RSA license. Some people also suggested that if you buy the Red-Hat server package (or whatever it is called) which includes a RSA license, you could just put it into a corner and use mod_ssl. But I am not sure if that kind of 'license transfer' is really legal. Gerd
Re: Apache/mod_perl
"PS" == Pramod Sokke [EMAIL PROTECTED] writes: PS Has anybody used stronghold? I'm considering using stronghold for SSL PS support since ours is a commercial application. Would mod_perl and all PS related modules work as fine with Stronghold as with plain Apache? Yes. There's even a file called INSTALL.simple.stronghold in the mod_perl source directory. I haven't built stronghold with mod_perl in nearly a year, so I couldn't tell you how accurate those instructions are. -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Vivek Khera, Ph.D.Khera Communications, Inc. Internet: [EMAIL PROTECTED] Rockville, MD +1-301-545-6996 GPG MIME spoken herehttp://www.khera.org/~vivek/
Re: Apache/mod_perl
"SA" == Scott Alexander [EMAIL PROTECTED] writes: SA So if i didn't want to wait, how would I go about getting an RSA SA liscense? to have a mod_ssl server before then. You don't get a license from RSA cheaply. This is why you buy Stronghold or Raven or some equivalent. The cost of Stronhold and Raven is not that much in the grand scheme of things if you have stuff important enough to be protecting with SSL in the first place. SA Also, are the CA's willing to sign for this time of server? Yes, they will. -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Vivek Khera, Ph.D.Khera Communications, Inc. Internet: [EMAIL PROTECTED] Rockville, MD +1-301-545-6996 GPG MIME spoken herehttp://www.khera.org/~vivek/
Re: Apache/mod_perl
"GK" == Gerd Knops [EMAIL PROTECTED] writes: GK includes a RSA license, you could just put it into a corner and use GK mod_ssl. But I am not sure if that kind of 'license transfer' is really GK legal. Buying a licensed RSA implementation doesn't give you a license to use RSA, it gives you a license to use that RSA implementation in the context which you bought it. Consult a lawyer, as I am not one, nor would I want to be. -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Vivek Khera, Ph.D.Khera Communications, Inc. Internet: [EMAIL PROTECTED] Rockville, MD +1-301-545-6996 GPG MIME spoken herehttp://www.khera.org/~vivek/
Re: Apache/mod_perl
The problem is cost. If licensing were cheap, Stronghold, Raven, and others would have a somewhat hard time staying in business (an issue they'll all have to address soon enough). The SSL vendors have an OEM license of some sort and so what you pay for a right-to-use license is a much smaller hunk of change especially because it's spread across a large customer base. So, if you can't wait till September, don't want to risk being in violation of a license and getting caught, and don't want to fork over $3-4K, the Stronghold/Raven route is a much better choice. -- Rob --On Friday, July 07, 2000 12:42 PM -0700 Scott Alexander [EMAIL PROTECTED] wrote: So if i didn't want to wait, how would I go about getting an RSA liscense? to have a mod_ssl server before then. Also, are the CA's willing to sign for this time of server? Scott On 7 Jul 2000, at 11:34, Rob Tanner wrote: If you can patiently wait till September 21 when the RSA patents expire covering the SSL algorithms, then you can use mod_ssl without fear of patent violations or lawsuits. _ _ _ _ __ _ _ _ _ /\_\_\_\_\/\_\ /\_\_\_\_\_\ /\/_/_/_/_/ /\/_/ \/_/_/_/_/_/ QUIDQUID LATINE DICTUM SIT, /\/_/__\/_/ __/\/_//\/_/ PROFUNDUM VIDITUR /\/_/_/_/_/ /\_\ /\/_//\/_/ /\/_/ \/_/ /\/_/_/\/_//\/_/ (Whatever is said in Latin \/_/ \/_/ \/_/_/_/_/ \/_/ appears profound) Rob Tanner McMinnville, Oregon [EMAIL PROTECTED]
RE: More on the set_handlers() / push_handlers() bug...
Forgive me for the obvious and probably stupid questions, (especially since I didn't see the earlier posts in which you probably answered them =o) but just for the record -- No problem. I appreciate your trying to help! Could you be misspelling "PerlAuthzHandler"? I note you said "PerlAuthzHandlers" with an 's' below Is that a feature I should kick myself for not being aware of? Or maybe just an email typo? Or a code error? (I do see it's a stack of handlerS, but) Heh. No, that was just a typo in my mail. I have it right in the handler code, which I had attached. Why undefine the handlers? Could you just use $r-set_handlers( PerlAuthzHandler = [ \OK ] ); or something? Could that be part of the problem? (Though I don't see how.I'm still pretty new to this.) Well, setting it to undef is what is recommended in the "Apache Modules" book (page 466), but I have tried setting it to \OK. Setting it to \OK turns out to do nothing at all. The server ignores your attempted change (even though it shows that it did change when you look at get_handlers) and continues as if you never even called set_handlers. Please, if you have the time and opportunity, give the AuthzCache a try. I'd really like to know that others can reproduce the problem. It can wrap around any perl Authz handler as follows: PerlAuthzHandler Tivoli::Apache::AuthzCache your authz handler here Tivoli::Apache::AuthzCache::manage_cache Regards, Christian - Christian Gilmore Infrastructure Tools Team Lead Web Multimedia Development Tivoli Systems, Inc.
Re: Apache/mod_perl
The issue has less to do with the cost of the whole thing, then with Strong hold having horrible support as well as a laundry list of other problems. Our as we convert our code base into completely mod_perl modules, the stronghold 2.4.2 has to be replaced by either 3.x or going to some other solutions based on apache/mod_perl/ssl component. Stronghold on solaris won't run the perlhandler that we want it to. The question were facing is whether it is worth it to go after the 3.x upgrade for strong hold or some other route. I'm going to try the raven 30day trial and see if i can make that work the way we want it to. Scott On 7 Jul 2000, at 16:42, Vivek Khera wrote: You don't get a license from RSA cheaply. This is why you buy Stronghold or Raven or some equivalent. The cost of Stronhold and Raven is not that much in the grand scheme of things if you have stuff important enough to be protecting with SSL in the first place
Re: Apache::ASP and post-POST redirect
Dmitry Beransky wrote: Hmm, I don't think I ever mentioned that ASP post-POST redirects just all of a sudden stopped working for me. A browser would send a POST request and never get anything back. The logs showed that the request was processed to the point of $Response-Redirect, but nothing would ever come out from the other side. The server was simply closing the connection without sending any data back to the client. I did find a hack that made redirects work by replacing ASP's Redirect with the following lines: I don't know a thing about Apache::Redirect, but I've had this same problem before: a handler returning REDIRECT after reading POST data would freeze the apache process (it got stuck trying to read the data again). The work-around I used is to set $r-header_in("Content-length", 0) just after reading the POST data. Then, if some cleanup function in apache tries to read it again, it finds that there is none to read. -- Roger Espel Llima, [EMAIL PROTECTED] http://www.iagora.com/~espel/index.html
Re: Hang in $r-print w/ POST
Le Fri, Jul 07, 2000 at 02:18:29PM +0200, heddy Boubaker a dit: to do. I'm just looking for a simple but complete html generator (instead of hard coding all html tags in my code), does such a beast exists or have I to write my own? I'm not sure what you mean by complete, but CGI::FastTemplate always suited my needs. Of course there are many other solutions, like others have told you : HTML::Mason (seems to be very popular), etc... HTH -- Patrick. Because if life has a meaning, we should already know it.
best encryption module
whats the best encryption module for use with mod perl? i want to encrypt passwords store in a db and then be able to check what a users inputs against it
Re: Apache/mod_perl
Stronghold would give you professional tech support, but if all you want is SSL, check out Ralph Engelschall's mod_ssl (www.modssl.org) based on the OpenSSL protocol (www.openssl.org). It works well with vanilla Apache, and it's *free* -- thanks again, Ralph. =o) --- Pramod Sokke [EMAIL PROTECTED] wrote: Has anybody used stronghold? I'm considering using stronghold for SSL support since ours is a commercial application. Would mod_perl and all related modules work as fine with Stronghold as with plain Apache? Thanks, Pramod At 10:24 AM 7/7/00 -0400, Vivek Khera wrote: "PS" == Pramod Sokke [EMAIL PROTECTED] writes: PS We are running Netscape Enterprise server with cgis written in perl and C. PS I'm looking at moving over to Apache and start using mod_perl. How [ .. ] PS over to Apache/mod_perl going to be a simple plug-in or would it involve PS re-writing lots of stuff? The C stuff will probably not be worth rewriting, but that depends on what it does. The perl stuff will need to be "cleaned" if it is sloppy code. That is, if it is clean running in Perl under "-w" and "use strict" you're most likely going to have little difficulty with them. But what you should do is use the two-server performance enhancement (using mod_proxy and mod_rewrite) and have your legacy apps run on the front-end server, and then migrate your perl to the mod_perl backend one at a time. -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Vivek Khera, Ph.D.Khera Communications, Inc. Internet: [EMAIL PROTECTED] Rockville, MD +1-301-545-6996 GPG MIME spoken herehttp://www.khera.org/~vivek/ = "Seize the wildness of the moment, Feel the movement of the moon -- Swans fly with wings wide open to the sky." -- B-52's - Real friends are those whom, when you inconvenience them, it bothers you more than them. -- me. =o) - "There are trivial truths and there are great Truths. The opposite of a trival truth is obviously false. The opposite of a great Truth is also true." -- Neils Bohr - TEMPVS PECVDEM COLLARE EST - It's time to thin the herd. - [http://www.catfishforbreakfast.com/letgod.html] - __ Do You Yahoo!? Send instant messages get email alerts with Yahoo! Messenger. http://im.yahoo.com/
Re: Apache/mod_perl
On Fri, 7 Jul 2000, Paul wrote: Stronghold would give you professional tech support, but if all you want is SSL, check out Ralph Engelschall's mod_ssl (www.modssl.org) based on the OpenSSL protocol (www.openssl.org). It works well with vanilla Apache, and it's *free* -- thanks again, Ralph. =o) Let me echo all of the above - I just installed mod_ssl'ed apache on linux and solaris with no problems, and it works great. Thanks Ralph, and everyone else involved. --- Pramod Sokke [EMAIL PROTECTED] wrote: Has anybody used stronghold? I'm considering using stronghold for SSL support since ours is a commercial application. Would mod_perl and all related modules work as fine with Stronghold as with plain Apache? Thanks, Pramod At 10:24 AM 7/7/00 -0400, Vivek Khera wrote: "PS" == Pramod Sokke [EMAIL PROTECTED] writes: PS We are running Netscape Enterprise server with cgis written in perl and C. PS I'm looking at moving over to Apache and start using mod_perl. How [ .. ] PS over to Apache/mod_perl going to be a simple plug-in or would it involve PS re-writing lots of stuff? The C stuff will probably not be worth rewriting, but that depends on what it does. The perl stuff will need to be "cleaned" if it is sloppy code. That is, if it is clean running in Perl under "-w" and "use strict" you're most likely going to have little difficulty with them. But what you should do is use the two-server performance enhancement (using mod_proxy and mod_rewrite) and have your legacy apps run on the front-end server, and then migrate your perl to the mod_perl backend one at a time. -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Vivek Khera, Ph.D.Khera Communications, Inc. Internet: [EMAIL PROTECTED] Rockville, MD +1-301-545-6996 GPG MIME spoken herehttp://www.khera.org/~vivek/ = "Seize the wildness of the moment, Feel the movement of the moon -- Swans fly with wings wide open to the sky." -- B-52's - Real friends are those whom, when you inconvenience them, it bothers you more than them. -- me. =o) - "There are trivial truths and there are great Truths. The opposite of a trival truth is obviously false. The opposite of a great Truth is also true." -- Neils Bohr - TEMPVS PECVDEM COLLARE EST - It's time to thin the herd. - [http://www.catfishforbreakfast.com/letgod.html] - __ Do You Yahoo!? Send instant messages get email alerts with Yahoo! Messenger. http://im.yahoo.com/ -- Blue Lang Unix Systems Admin QSP, Inc., 3200 Atlantic Ave, Ste 100, Raleigh, NC, 27604 Home: 919 835 1540 Work: 919 875 6994 Fax: 919 872 4015
Newbie: make test failure!
Config: mod_perl-1.24 perl v5.6.0 apache 1.3.12 RedHat linux 6.1, kernel 2.2.12 Using: [/usr/local/apahce/build/mod_perl-1.24]# perl Makefile.PL \ EVERYTHING=1 \ APACHE_PREFIX=/usr/local/apache everything seems to look ok (to me) for both this and 'make' but when I run 'make test' I get: ... httpd listening on port 8529 will write error_log to: t/logs/error_log letting apache warm up...\c done /usr/bin/perl t/TEST 0 still waiting for server to warm up.not ok server failed to start! (please examine t/logs/error_log) at t/TEST line 95. make: *** [run_tests] Error 9 which isn't so good I figure. t/logs/error_log says this: [Sat July 8 13:34:16 2000] [crit] (98)Addresses already in use: make_sock: could not bind to port 8529 Although the very first time I ran this it fell over with a different error (which I rather stupidly failed to note and subsequently hosed the log - it's great being new at this...) I can also reproduce this behaviour (perhaps not surprisingly) by running 'make start_http' followed by 't/TEST -v modules' Any insight appreciated! Cheers, Greg.
cvs commit: modperl-site/embperl Changes.pod.1.html
richter 00/07/07 14:56:26 Modified:embperl Changes.pod.1.html Log: Embperl Webpages - Changes Revision ChangesPath 1.157 +13 -4 modperl-site/embperl/Changes.pod.1.html Index: Changes.pod.1.html === RCS file: /home/cvs/modperl-site/embperl/Changes.pod.1.html,v retrieving revision 1.156 retrieving revision 1.157 diff -u -r1.156 -r1.157 --- Changes.pod.1.html2000/06/22 12:05:25 1.156 +++ Changes.pod.1.html2000/07/07 21:56:25 1.157 @@ -18,7 +18,7 @@ blockquote [a href="index.html"HOME/a]nbsp;nbsp; [a href="Changes.pod.cont.html"CONTENT/a]nbsp;nbsp; [a href="Changes.pod.cont.html"PREV (Revision History - Content)/a]nbsp;nbsp; [a href="Changes.pod.2.html"NEXT (1.3b3 (BETA) 25.04.2000)/a]nbsp;nbsp; brhr P -Last Update: Thu Jun 22 14:05:40 2000 (MET) +Last Update: Fri Jul 7 23:56:49 2000 (MET) P NOTE: This version is only available via A HREF="CVS.pod.1.html#INTRO""CVS"/A @@ -35,9 +35,18 @@ - Characters between 128 and 159 are all HTML escaped now to avoid problems with buggy browser, which were reported to treat the chars 139 and 141 as lt; and gt;. Spotted by Dirk Lutzebaeck. -/PRE -P -PRE + - If a requested file is not found when using EmbperlObject as handler, + the file given by Clt;EMBPERL_OBJECT_FALLBACKgt; is displayed instead. + If Clt;EMBPERL_OBJECT_FALLBACKgt; isn't set a staus 404, NOT_FOUND is + returned as usual. + - quot;perl Makefile.PL debugquot; will build debugging information for + gdb/ms-vc++ into Embperl library. + - test.pl can take a bunch of new options for debugging Embperl itself. + See make test TESTARGS=quot;--helpquot;. + - Embperl 1.x and 2.x share now the same Makefile.PL and test.pl + - Added new debug flag dbgObjectSerach which logs the EmbperlObjects + work when searching the correct file. + /PRE p[a href="index.html"HOME/a]nbsp;nbsp; [a href="Changes.pod.cont.html"CONTENT/a]nbsp;nbsp; [a href="Changes.pod.cont.html"PREV (Revision History - Content)/a]nbsp;nbsp; [a href="Changes.pod.2.html"NEXT (1.3b3 (BETA) 25.04.2000)/a]nbsp;nbsp; br font color="#808080"___br