From doug.miles at bpxinternet.com Wed Jan 8 09:40:29 2003 From: doug.miles at bpxinternet.com (Doug Miles) Date: Thu Aug 5 00:16:54 2004 Subject: Phoenix.pm: mod_perl internal redirect question Message-ID: <3E1C466D.4010306@bpxinternet.com> Hi all! This is a question for all you mod_perl experts out there. I am attempting to do an internal redirect in mod_perl, preserving any POST request parameters that might be kicking around. I couldn't get the example code I found here: http://www.esc.auckland.ac.nz/Docs/mod_perl/snippets.html#Redirect_a_POST_request_forward to work. It appeared to do the redirect, but wouldn't pass any of the POST parameters. To complicate matters, I'm using Apache::Request (which is a subclass of Apache) instead of the default Apache object. I did finally get the redirect to work, but only by building up my own query string and appending it to the URL being redirected to. This works fine, but I'd really like to know why the other code doesn't work. The commented out code is the code in question. The only other change is to remove the appending of the parameters. Any ideas are appreciated. # redirect ############################################################# sub redirect { use Apache::Constants qw(M_GET); my $self = shift; my %tag_parameter = @_; my @parameters; foreach my $parameter ($self->request->param) { push ( @parameters, uri_escape($parameter) . '=' . uri_escape($self->request->param($parameter)) ); } my $parameters = join('&', @parameters); # Don't know why this doesn't work. Appending parameters to URL below does. ??? # my $content = $self->request->content; # $self->request->method('GET'); # $self->request->method_number(M_GET); # $self->request->headers_in->unset("Content-length"); # $self->request->args($parameters); # my $content = $self->request->content; $self-> request-> internal_redirect ( $self->normalize_absolute_path("/pcms/$tag_parameter{'url'}") . "?$parameters" ); } # END: redirect From intertwingled at qwest.net Wed Jan 8 10:51:31 2003 From: intertwingled at qwest.net (intertwingled) Date: Thu Aug 5 00:16:54 2004 Subject: Phoenix.pm: mod_perl internal redirect question In-Reply-To: <3E1C466D.4010306@bpxinternet.com> Message-ID: <3.0.6.32.20030108095131.007e5180@pop.phnx.qwest.net> Wow. Phoenix.pm still exists! =) At 08:40 AM 1/8/03 -0700, you wrote: >Hi all! > >This is a question for all you mod_perl experts out there. I am >attempting to do an internal redirect in mod_perl, preserving any POST >request parameters that might be kicking around. I couldn't get the >example code I found here: > >http://www.esc.auckland.ac.nz/Docs/mod_perl/snippets.html#Redirect_a_POST_r equest_forward > >to work. It appeared to do the redirect, but wouldn't pass any of the >POST parameters. To complicate matters, I'm using Apache::Request >(which is a subclass of Apache) instead of the default Apache object. I >did finally get the redirect to work, but only by building up my own >query string and appending it to the URL being redirected to. This >works fine, but I'd really like to know why the other code doesn't work. > The commented out code is the code in question. The only other change >is to remove the appending of the parameters. Any ideas are appreciated. > ># redirect ############################################################# > >sub redirect >{ > use Apache::Constants qw(M_GET); > > my $self = shift; > > my %tag_parameter = @_; > > my @parameters; > > foreach my $parameter ($self->request->param) > { > > push > ( > @parameters, > uri_escape($parameter) . '=' . > uri_escape($self->request->param($parameter)) > ); > > } > > my $parameters = join('&', @parameters); > ># Don't know why this doesn't work. Appending parameters to URL below >does. ??? > ># my $content = $self->request->content; ># $self->request->method('GET'); ># $self->request->method_number(M_GET); ># $self->request->headers_in->unset("Content-length"); ># $self->request->args($parameters); ># my $content = $self->request->content; > > $self-> > request-> > internal_redirect > ( > $self->normalize_absolute_path("/pcms/$tag_parameter{'url'}") . > "?$parameters" > ); > >} # END: redirect > > > -- even the safest course is fraught with peril From phx-pm-list at grueslayer.com Wed Jan 8 10:47:09 2003 From: phx-pm-list at grueslayer.com (David A. Sinck) Date: Thu Aug 5 00:16:54 2004 Subject: Phoenix.pm: mod_perl internal redirect question References: <3E1C466D.4010306@bpxinternet.com> Message-ID: <15900.22029.174032.802169@magnitude.righthandgraphics.com> \_ SMTP quoth Doug Miles on 1/8/2003 08:40 as having spake thusly: \_ \_ Hi all! \_ \_ This is a question for all you mod_perl experts out there. I am \_ attempting to do an internal redirect in mod_perl, preserving any POST \_ request parameters that might be kicking around. I couldn't get the \_ example code I found here: You say POST, yet your code only m/GET/. Perhaps these aren't the droid parameters you're looking for? Dunno, just putting up a balloon. David From scott at illogics.org Wed Jan 8 10:42:44 2003 From: scott at illogics.org (Scott Walters) Date: Thu Aug 5 00:16:54 2004 Subject: Phoenix.pm: mod_perl internal redirect question In-Reply-To: from "Doug Miles" at Jan 08, 2003 08:40:29 AM Message-ID: <200301081642.h08GgjaO016016@slowass.net> Doug and folks, Sorry, I don't have time to read the docs and I haven't delt with this before, but I thought I would make a few general statements. CGI.pm is going to read the post off of STDIN. This is something that cannot be undone. You don't have the browser-caching-the-get-string problem, so if you don't have large fields, this is an acceptable solution. It should also be noted that if you don't use CGI to read the post request, it will still be waiting there in STDIN for the CGI application you redirect to to read. In other words, don't read the POST data and it will still be POST data. I'm assuming you don't want to modify the source of the application you're posting to, but I still want to note that using something like Apache::Session (but not Apache::Session) is cleaner - stick the user data in a session-specific container and let things access it there. Finally, I'm going to note why I don't use mod_perl - I hate having a small number of threads/processes, and each costs a lot using Apache/mod_perl. I'm in love with thttpd (which runs as a single process which services any number - even thousands of concurrent requests) and plain old CGI or redirects to a FastCGI-like server. I'm sure mod_perl can't be changed in your project, but I thought I would throw stones at it anyway. I have a copy of Programming Apache Modules with Perl and C if you'd like to barrow it, or perhaps I'll make time in a bit here... Good to see that you aren't shackled up in the dungeon full time and they let you out =) -scott > > > Hi all! > > This is a question for all you mod_perl experts out there. I am > attempting to do an internal redirect in mod_perl, preserving any POST > request parameters that might be kicking around. I couldn't get the > example code I found here: > > http://www.esc.auckland.ac.nz/Docs/mod_perl/snippets.html#Redirect_a_POST_request_forward > > to work. It appeared to do the redirect, but wouldn't pass any of the > POST parameters. To complicate matters, I'm using Apache::Request > (which is a subclass of Apache) instead of the default Apache object. I > did finally get the redirect to work, but only by building up my own > query string and appending it to the URL being redirected to. This > works fine, but I'd really like to know why the other code doesn't work. > The commented out code is the code in question. The only other change > is to remove the appending of the parameters. Any ideas are appreciated. > > # redirect ############################################################# > > sub redirect > { > use Apache::Constants qw(M_GET); > > my $self = shift; > > my %tag_parameter = @_; > > my @parameters; > > foreach my $parameter ($self->request->param) > { > > push > ( > @parameters, > uri_escape($parameter) . '=' . > uri_escape($self->request->param($parameter)) > ); > > } > > my $parameters = join('&', @parameters); > > # Don't know why this doesn't work. Appending parameters to URL below > does. ??? > > # my $content = $self->request->content; > # $self->request->method('GET'); > # $self->request->method_number(M_GET); > # $self->request->headers_in->unset("Content-length"); > # $self->request->args($parameters); > # my $content = $self->request->content; > > $self-> > request-> > internal_redirect > ( > $self->normalize_absolute_path("/pcms/$tag_parameter{'url'}") . > "?$parameters" > ); > > } # END: redirect > > From doug.miles at bpxinternet.com Wed Jan 8 11:15:41 2003 From: doug.miles at bpxinternet.com (Doug Miles) Date: Thu Aug 5 00:16:54 2004 Subject: Phoenix.pm: mod_perl internal redirect question References: <3E1C466D.4010306@bpxinternet.com> <15900.22029.174032.802169@magnitude.righthandgraphics.com> Message-ID: <3E1C5CBD.9080103@bpxinternet.com> David A. Sinck wrote: > > \_ SMTP quoth Doug Miles on 1/8/2003 08:40 as having spake thusly: > \_ > \_ Hi all! > \_ > \_ This is a question for all you mod_perl experts out there. I am > \_ attempting to do an internal redirect in mod_perl, preserving any POST > \_ request parameters that might be kicking around. I couldn't get the > \_ example code I found here: > > You say POST, yet your code only m/GET/. Perhaps these aren't the droid > parameters you're looking for? > > Dunno, just putting up a balloon. > > David > Yeah, its confusing. The idea is that you convert the POST to a GET on the redirect. At least that's the theory. I wouldn't know for sure. :) From doug.miles at bpxinternet.com Wed Jan 8 11:22:39 2003 From: doug.miles at bpxinternet.com (Doug Miles) Date: Thu Aug 5 00:16:54 2004 Subject: Phoenix.pm: mod_perl internal redirect question References: <3.0.6.32.20030108095131.007e5180@pop.phnx.qwest.net> Message-ID: <3E1C5E5F.6040808@bpxinternet.com> intertwingled wrote: > Wow. Phoenix.pm still exists! =) > The reports of its death have been greatly exaggerated. From phx-pm-list at grueslayer.com Wed Jan 8 15:11:02 2003 From: phx-pm-list at grueslayer.com (David A. Sinck) Date: Thu Aug 5 00:16:54 2004 Subject: Phoenix.pm: mod_perl internal redirect question Message-ID: <15900.37862.189537.974415@magnitude.righthandgraphics.com> Does anyone have hints (preferably live code :-) about how to make Term::ReadLine have a default value at the prompt so one can edit it? Sadly, the best I've come up with is: my $term = new Term::ReadLine('a'); $term->addhistory('lame default'); my $input = $term->readline('ctrl-p for default> '); and then pressing ctrl-p when I get hte prompt. tfm isn't clear, the cookbook doesn't have useful examples, and I'm hungry for a solution. TIA, David From doug.miles at bpxinternet.com Wed Jan 8 15:32:08 2003 From: doug.miles at bpxinternet.com (Doug Miles) Date: Thu Aug 5 00:16:54 2004 Subject: Phoenix.pm: Job: Short term to possible? permanent MS SQL project Message-ID: <3E1C98D8.6090602@bpxinternet.com> Hey all, I have a lead here for anyone who has MS SQL Server experience and some C++ experience. It is a short term project that might become a full time job if everything works out. Send your resume to Dennis Foster (dennisf@dfsaz.com), and let him know that you found out about this from me (Doug Miles). From scott at illogics.org Tue Jan 14 17:04:20 2003 From: scott at illogics.org (Scott Walters) Date: Thu Aug 5 00:16:54 2004 Subject: Phoenix.pm: Meeting 1/17/2003 Message-ID: <200301142304.h0EN4MOh007212@slowass.net> We'll be having a Phoenix.pm meeting Thursday, January 17th at 7:00PM. It will be held at Bowne, which is located at 1500 N. Central Avenue, which is on the Southwest corner of Central and McDowell. The parking lot is gated, so just press the button on the intercom, and tell the receptionist that you are there for the Perl meeting. Park in the lot that is straight ahead from the entrance on the South side of McDowell. Park in any uncovered, non-reserved space. Proceed to the main lobby, which is on the Northeast side of the parking lot. Scott will briefly present ActiveWikiPages, a minimal content management system with user scripting, then meeting goers will be broken into teams and assigned laptops for a fun and informal programming competition. The winning entry will be selected by popular vote. Everyone is encouraged to participate - if you're learning Perl, Doug, Kurt and Scott will be available to help. Rules are few and will be announced at the meeting. Bring your laptop if you like. I was asked to send the announcement if everything was go, so I hope I didn't screw it up, especially the date ;) From scott at illogics.org Tue Jan 14 17:36:08 2003 From: scott at illogics.org (Scott Walters) Date: Thu Aug 5 00:16:54 2004 Subject: Phoenix.pm: Re: Perl Meetings (fwd) In-Reply-To: from "Kurt Von Tiehl" at Jan 14, 2003 04:27:29 PM Message-ID: <200301142336.h0ENa9vV007547@slowass.net> > > > > > *grin* Whatever works ;) Btw, Thursday is the 16th ;) > > -Kurt Ack! I typed "cal 2002" into my xterm window... sorry folks... -scott From doug.miles at bpxinternet.com Wed Jan 15 13:13:03 2003 From: doug.miles at bpxinternet.com (Doug Miles) Date: Thu Aug 5 00:16:54 2004 Subject: Phoenix.pm: Apache::Session::Lock::File under mod_perl Message-ID: <3E25B2BF.9010305@bpxinternet.com> Anyone out there using Apache::Session::File and Apache::Session::Lock::File under mod_perl? The problem I am having is that Apache::Session::Lock::File is aparently not releasing the exclusive lock when the tied hash goes out of scope. I can get a simple test program to work from the command line, but when I try it under mod_perl, it locks on the tie to the existing session. If I delete the lock file, it works fine. Any ideas? From aj at exiledplanet.org Wed Jan 15 14:26:58 2003 From: aj at exiledplanet.org (Andrew Johnson) Date: Thu Aug 5 00:16:54 2004 Subject: Phoenix.pm: Apache::Session::Lock::File under mod_perl In-Reply-To: <3E25B2BF.9010305@bpxinternet.com> References: <3E25B2BF.9010305@bpxinternet.com> Message-ID: <3E25C412.6060801@exiledplanet.org> Doug Miles wrote: > Anyone out there using Apache::Session::File and > Apache::Session::Lock::File under mod_perl? The problem I am having is > that Apache::Session::Lock::File is aparently not releasing the > exclusive lock when the tied hash goes out of scope. I can get a simple > test program to work from the command line, but when I try it under > mod_perl, it locks on the tie to the existing session. If I delete the > lock file, it works fine. Any ideas? > Sounds like the lock is persisting because the Apache process that created the lock still exists. Are you sure the part of the code where you're calling release_write_lock() or release_read_lock() is actually being run? From scott at illogics.org Wed Jan 15 14:51:25 2003 From: scott at illogics.org (Scott Walters) Date: Thu Aug 5 00:16:54 2004 Subject: Phoenix.pm: Apache::Session::Lock::File under mod_perl In-Reply-To: from "Doug Miles" at Jan 15, 2003 12:13:03 PM Message-ID: <200301152051.h0FKpPul014495@slowass.net> Ahhh, I distinctly remember the 18th time that Apache::Session failed on Contact Designs. We were sitting in our attorney's office in a downtown Phoenix highrise, arbitrating with an non-paying client who was alleging that we didn't have the experience needed to write an e-commerce site, so they could step out of the contract and use Yahoo! sites, which look like garbage - this after cating to demand after demand by this client. Laptop and Ricochet wireless Internet on the confrence table, I prepare to demonostrate the check out process on another clients site - when Apache::Session decides that another processes hasn't released the SHM lock. This is an absolutely true story. Apache::Session caused constant problems in each of its modes - locking on SHM, locking on a database, serializing to file, serializing to database, etc. I've flamed Apache::Session in this channel before, and I'll do it again. Use require, flock, and Storable, just like God intended. Apache::Session is one of those things that looks cool, but fails erratically under real-world conditions. Sorry to be patronizing, but I told you so, nanana! -scott > > > Anyone out there using Apache::Session::File and > Apache::Session::Lock::File under mod_perl? The problem I am having is > that Apache::Session::Lock::File is aparently not releasing the > exclusive lock when the tied hash goes out of scope. I can get a simple > test program to work from the command line, but when I try it under > mod_perl, it locks on the tie to the existing session. If I delete the > lock file, it works fine. Any ideas? > > From doug.miles at bpxinternet.com Wed Jan 15 15:22:41 2003 From: doug.miles at bpxinternet.com (Doug Miles) Date: Thu Aug 5 00:16:54 2004 Subject: Phoenix.pm: Apache::Session::Lock::File under mod_perl References: <3E25B2BF.9010305@bpxinternet.com> <3E25C412.6060801@exiledplanet.org> Message-ID: <3E25D121.4070603@bpxinternet.com> Andrew Johnson wrote: > Doug Miles wrote: > >> Anyone out there using Apache::Session::File and >> Apache::Session::Lock::File under mod_perl? The problem I am having >> is that Apache::Session::Lock::File is aparently not releasing the >> exclusive lock when the tied hash goes out of scope. I can get a >> simple test program to work from the command line, but when I try it >> under mod_perl, it locks on the tie to the existing session. If I >> delete the lock file, it works fine. Any ideas? >> > > Sounds like the lock is persisting because the Apache process that > created the lock still exists. Are you sure the part of the code where > you're calling release_write_lock() or release_read_lock() is actually > being run? Yeah, the process still exists, but I was executing a release_all_locks like this: my $lock = new Apache::Session::Lock::File; $lock->release_all_locks($session_id); after I finished "writing" to the session hash. Now I have an unlink in the same place, and that works. That kind of defeats the purpose of having a lock though. :) From doug.miles at bpxinternet.com Wed Jan 15 15:25:17 2003 From: doug.miles at bpxinternet.com (Doug Miles) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: Apache::Session::Lock::File under mod_perl References: <200301152051.h0FKpPul014495@slowass.net> Message-ID: <3E25D1BD.6010602@bpxinternet.com> Scott Walters wrote: > > Ahhh, I distinctly remember the 18th time that Apache::Session failed on > Contact Designs. We were sitting in our attorney's office in a downtown > Phoenix highrise, arbitrating with an non-paying client who was alleging > that we didn't have the experience needed to write an e-commerce site, > so they could step out of the contract and use Yahoo! sites, which look > like garbage - this after cating to demand after demand by this client. > Laptop and Ricochet wireless Internet on the confrence table, I prepare > to demonostrate the check out process on another clients site - when > Apache::Session decides that another processes hasn't released the SHM > lock. > > This is an absolutely true story. Apache::Session caused constant > problems in each of its modes - locking on SHM, locking on a database, > serializing to file, serializing to database, etc. > > I've flamed Apache::Session in this channel before, and I'll do it again. > Use require, flock, and Storable, just like God intended. > > Apache::Session is one of those things that looks cool, but fails > erratically under real-world conditions. Sorry to be patronizing, but I told > you so, nanana! > > -scott Well in my case it isn't erratic. It's frustratingly consistent. :) > > >> >>Anyone out there using Apache::Session::File and >>Apache::Session::Lock::File under mod_perl? The problem I am having is >>that Apache::Session::Lock::File is aparently not releasing the >>exclusive lock when the tied hash goes out of scope. I can get a simple >>test program to work from the command line, but when I try it under >>mod_perl, it locks on the tie to the existing session. If I delete the >>lock file, it works fine. Any ideas? >> >> > > > From doug.miles at bpxinternet.com Wed Jan 15 15:28:36 2003 From: doug.miles at bpxinternet.com (Doug Miles) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: Apache::Session::Lock::File under mod_perl References: <200301152051.h0FKpPul014495@slowass.net> Message-ID: <3E25D284.7050405@bpxinternet.com> Scott Walters wrote: > > Ahhh, I distinctly remember the 18th time that Apache::Session failed on > Contact Designs. We were sitting in our attorney's office in a downtown > Phoenix highrise, arbitrating with an non-paying client who was alleging > that we didn't have the experience needed to write an e-commerce site, > so they could step out of the contract and use Yahoo! sites, which look > like garbage - this after cating to demand after demand by this client. > Laptop and Ricochet wireless Internet on the confrence table, I prepare > to demonostrate the check out process on another clients site - when > Apache::Session decides that another processes hasn't released the SHM > lock. > > This is an absolutely true story. Apache::Session caused constant > problems in each of its modes - locking on SHM, locking on a database, > serializing to file, serializing to database, etc. > > I've flamed Apache::Session in this channel before, and I'll do it again. > Use require, flock, and Storable, just like God intended. > > Apache::Session is one of those things that looks cool, but fails > erratically under real-world conditions. Sorry to be patronizing, but I told > you so, nanana! > > -scott > Did you dive into the code any to see what the problem was? I'm trying to decide if it's worth the effort. > >> >>Anyone out there using Apache::Session::File and >>Apache::Session::Lock::File under mod_perl? The problem I am having is >>that Apache::Session::Lock::File is aparently not releasing the >>exclusive lock when the tied hash goes out of scope. I can get a simple >>test program to work from the command line, but when I try it under >>mod_perl, it locks on the tie to the existing session. If I delete the >>lock file, it works fine. Any ideas? >> >> > > > From scott at illogics.org Wed Jan 15 17:20:22 2003 From: scott at illogics.org (Scott Walters) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: Apache::Session::Lock::File under mod_perl In-Reply-To: from "Doug Miles" at Jan 15, 2003 02:28:36 PM Message-ID: <200301152320.h0FNKM5W015679@slowass.net> > Doug incantonated: > Did you dive into the code any to see what the problem was? I'm trying > to decide if it's worth the effort. No. I opened it in vi once and noticed that it was a few thousands more lines that it should be. I sighed and decided it would be easier to replace it than try to fix, undestand, clean, or decruft it. Hrm. I should write a module that takes another module and puts it through the unit tests taking out each line of code one by one. Each line that can be removed will be removed, and the module output in its new, decrufted form. Not as good as a hand polish job, but it'd be interesting to see how much the average module reduces. Reading most peoples code, lines that don't do anything are scattered all over the place. Anyway, in all seriousness, require, flock/SQL's lock row, and Storable worked fine, and is essentially what Apache::Session is doing under all of the cruft anyway. Perl has SYSV SHM primitives built in, but I recommend against this - a core fault or incomplete exception handling code could leave the semaphore stuck. You'll have to write a little object wrapper if you want to release the lock based on scope... -scott > > > > Scott Walters wrote: > > > > Ahhh, I distinctly remember the 18th time that Apache::Session failed on > > Contact Designs. We were sitting in our attorney's office in a downtown > > Phoenix highrise, arbitrating with an non-paying client who was alleging > > that we didn't have the experience needed to write an e-commerce site, > > so they could step out of the contract and use Yahoo! sites, which look > > like garbage - this after cating to demand after demand by this client. > > Laptop and Ricochet wireless Internet on the confrence table, I prepare > > to demonostrate the check out process on another clients site - when > > Apache::Session decides that another processes hasn't released the SHM > > lock. > > > > This is an absolutely true story. Apache::Session caused constant > > problems in each of its modes - locking on SHM, locking on a database, > > serializing to file, serializing to database, etc. > > > > I've flamed Apache::Session in this channel before, and I'll do it again. > > Use require, flock, and Storable, just like God intended. > > > > Apache::Session is one of those things that looks cool, but fails > > erratically under real-world conditions. Sorry to be patronizing, but I told > > you so, nanana! > > > > -scott > > > > > > >> > >>Anyone out there using Apache::Session::File and > >>Apache::Session::Lock::File under mod_perl? The problem I am having is > >>that Apache::Session::Lock::File is aparently not releasing the > >>exclusive lock when the tied hash goes out of scope. I can get a simple > >>test program to work from the command line, but when I try it under > >>mod_perl, it locks on the tie to the existing session. If I delete the > >>lock file, it works fine. Any ideas? > >> > >> > > > > > > > > > From doug.miles at bpxinternet.com Thu Jan 16 11:04:05 2003 From: doug.miles at bpxinternet.com (Doug Miles) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: Apache::Session::Lock::File under mod_perl References: <200301152320.h0FNKM5W015679@slowass.net> Message-ID: <3E26E605.6030302@bpxinternet.com> Scott Walters wrote: >>Doug incantonated: >>Did you dive into the code any to see what the problem was? I'm trying >>to decide if it's worth the effort. > > > No. I opened it in vi once and noticed that it was a few thousands more > lines that it should be. I sighed and decided it would be easier to replace > it than try to fix, undestand, clean, or decruft it. Hrm. I should write a > module that takes another module and puts it through the unit tests > taking out each line of code one by one. Each line that can be removed > will be removed, and the module output in its new, decrufted form. Not > as good as a hand polish job, but it'd be interesting to see how much > the average module reduces. Reading most peoples code, lines that don't > do anything are scattered all over the place. Anyway, in all seriousness, > require, flock/SQL's lock row, and Storable worked fine, and is essentially > what Apache::Session is doing under all of the cruft anyway. Perl has > SYSV SHM primitives built in, but I recommend against this - a core fault > or incomplete exception handling code could leave the semaphore stuck. > You'll have to write a little object wrapper if you want to release the > lock based on scope... > > -scott OK, thanks! I'll probably write my own too... I'm just surprised that I couldn't find more about it on Google groups. :( > > > >> >> >>Scott Walters wrote: >> >>>Ahhh, I distinctly remember the 18th time that Apache::Session failed on >>>Contact Designs. We were sitting in our attorney's office in a downtown >>>Phoenix highrise, arbitrating with an non-paying client who was alleging >>>that we didn't have the experience needed to write an e-commerce site, >>>so they could step out of the contract and use Yahoo! sites, which look >>>like garbage - this after cating to demand after demand by this client. >>>Laptop and Ricochet wireless Internet on the confrence table, I prepare >>>to demonostrate the check out process on another clients site - when >>>Apache::Session decides that another processes hasn't released the SHM >>>lock. >>> >>>This is an absolutely true story. Apache::Session caused constant >>>problems in each of its modes - locking on SHM, locking on a database, >>>serializing to file, serializing to database, etc. >>> >>>I've flamed Apache::Session in this channel before, and I'll do it again. >>>Use require, flock, and Storable, just like God intended. >>> >>>Apache::Session is one of those things that looks cool, but fails >>>erratically under real-world conditions. Sorry to be patronizing, but I told >>>you so, nanana! >>> >>>-scott >>> >> >>>>Anyone out there using Apache::Session::File and >>>>Apache::Session::Lock::File under mod_perl? The problem I am having is >>>>that Apache::Session::Lock::File is aparently not releasing the >>>>exclusive lock when the tied hash goes out of scope. I can get a simple >>>>test program to work from the command line, but when I try it under >>>>mod_perl, it locks on the tie to the existing session. If I delete the >>>>lock file, it works fine. Any ideas? >>>> >>>> >>> >>> >>> >> >> > > From doug.miles at bpxinternet.com Thu Jan 16 11:55:44 2003 From: doug.miles at bpxinternet.com (Doug Miles) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: Reminder: Meeting 1/16/2003 Message-ID: <3E26F220.10403@bpxinternet.com> Please RSVP... We'll be having a Phoenix.pm meeting Thursday, January 16th at 7:00PM. It will be held at Bowne, which is located at 1500 N. Central Avenue, which is on the Southwest corner of Central and McDowell. The parking lot is gated, so just press the button on the intercom, and tell the receptionist that you are there for the Perl meeting. Park in the lot that is straight ahead from the entrance on the South side of McDowell. Park in any uncovered, non-reserved space. Proceed to the main lobby, which is on the Northeast side of the parking lot. Scott will briefly present ActiveWikiPages, a minimal content management system with user scripting, then meeting goers will be broken into teams and assigned laptops for a fun and informal programming competition. The winning entry will be selected by popular vote. Everyone is encouraged to participate - if you're learning Perl, Doug, Kurt and Scott will be available to help. Rules are few and will be announced at the meeting. Bring your laptop if you like. From mpjbell at softhome.net Thu Jan 16 11:56:05 2003 From: mpjbell at softhome.net (Marty Bell) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: Reminder: Meeting 1/16/2003 In-Reply-To: <3E26F220.10403@bpxinternet.com> Message-ID: <3.0.5.32.20030116115605.00a35b70@pop.SoftHome.net> See you there! Sorry I don't have a lap top to bring. Marty At 10:55 AM 1/16/2003 -0700, you wrote: >Please RSVP... > >We'll be having a Phoenix.pm meeting Thursday, January 16th at 7:00PM. >It will be held at Bowne, which is located at 1500 N. Central Avenue, >which is on the Southwest corner of Central and McDowell. The parking >lot is gated, so just press the button on the intercom, and tell the >receptionist that you are there for the Perl meeting. Park in the lot >that is straight ahead from the entrance on the South side of McDowell. >Park in any uncovered, non-reserved space. Proceed to the main lobby, >which is on the Northeast side of the parking lot. > >Scott will briefly present ActiveWikiPages, a minimal content management >system with user scripting, then meeting goers will be broken into teams >and assigned laptops for a fun and informal programming competition. >The winning entry will be selected by popular vote. Everyone is encouraged >to participate - if you're learning Perl, Doug, Kurt and Scott will be >available >to help. Rules are few and will be announced at the meeting. Bring your >laptop if you like. > > > From aj at exiledplanet.org Thu Jan 16 14:21:15 2003 From: aj at exiledplanet.org (Andrew Johnson) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: Reminder: Meeting 1/16/2003 In-Reply-To: <3E26F220.10403@bpxinternet.com> References: <3E26F220.10403@bpxinternet.com> Message-ID: <3E27143B.9080105@exiledplanet.org> Doug Miles wrote: > Please RSVP... > > We'll be having a Phoenix.pm meeting Thursday, January 16th at 7:00PM. I'll be there. From scott at illogics.org Thu Jan 16 14:46:46 2003 From: scott at illogics.org (Scott Walters) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: Re: Whoops - need cables In-Reply-To: from "Kurt Von Tiehl" at Jan 16, 2003 10:39:36 AM Message-ID: <200301162046.h0GKkmB5022309@slowass.net> I've got piles of cables, back from before all of my machines have worn down > > > Just realised something... I was packing up my hub for tonight, and realised I > really only have 1 cable I can conveniently yank for it (well, I also have a > 1-footer for the upload to Scott's hub). Any o' youse guys think you can bring > some along? Otherwise, I'll just hit Fry's on my way down there and get a half > dozen of 'em, but I'd like to know how many should. > > > > -Kurt > > > From aj at exiledplanet.org Thu Jan 16 15:29:26 2003 From: aj at exiledplanet.org (Andrew Johnson) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: Re: Whoops - need cables In-Reply-To: <200301162046.h0GKkmB5022309@slowass.net> References: <200301162046.h0GKkmB5022309@slowass.net> Message-ID: <3E272436.10109@exiledplanet.org> Scott Walters wrote: > I've got piles of cables, back from before all of my machines have worn down > > >> >>Just realised something... I was packing up my hub for tonight, and realised I >>really only have 1 cable I can conveniently yank for it (well, I also have a >>1-footer for the upload to Scott's hub). Any o' youse guys think you can bring >>some along? Otherwise, I'll just hit Fry's on my way down there and get a half >>dozen of 'em, but I'd like to know how many should. >> >> >> >> -Kurt >> Cat 5 10BaseT? I can bring several. From mpjbell at softhome.net Thu Jan 16 14:35:58 2003 From: mpjbell at softhome.net (Marty Bell) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: Re: Whoops - need cables In-Reply-To: <200301162046.h0GKkmB5022309@slowass.net> References: Message-ID: <3.0.5.32.20030116143558.00a35b70@pop.SoftHome.net> I've got one I'll bring (it's 3 feet, only good for near the hub). Marty At 12:46 PM 1/16/2003 -0800, you wrote: > >I've got piles of cables, back from before all of my machines have worn down > >> >> >> Just realised something... I was packing up my hub for tonight, and realised I >> really only have 1 cable I can conveniently yank for it (well, I also have a >> 1-footer for the upload to Scott's hub). Any o' youse guys think you can bring >> some along? Otherwise, I'll just hit Fry's on my way down there and get a half >> dozen of 'em, but I'd like to know how many should. >> >> >> >> -Kurt >> >> >> > > From doug.miles at bpxinternet.com Thu Jan 16 15:56:39 2003 From: doug.miles at bpxinternet.com (Doug Miles) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: Re: Whoops - need cables References: <200301162046.h0GKkmB5022309@slowass.net> Message-ID: <3E272A97.80102@bpxinternet.com> Scott Walters wrote: > I've got piles of cables, back from before all of my machines have worn down That'll be fine if you want to bring them. Saves me from having to roll them back up. :) > >> >>Just realised something... I was packing up my hub for tonight, and realised I >>really only have 1 cable I can conveniently yank for it (well, I also have a >>1-footer for the upload to Scott's hub). Any o' youse guys think you can bring >>some along? Otherwise, I'll just hit Fry's on my way down there and get a half >>dozen of 'em, but I'd like to know how many should. >> >> >> >> -Kurt >> >> >> > > > From doug.miles at bpxinternet.com Fri Jan 17 11:04:54 2003 From: doug.miles at bpxinternet.com (Doug Miles) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: Banner Message-ID: <3E2837B6.3040805@bpxinternet.com> It was brought to my attention last night by Tim Ayers that my stock install of RedHat 7.3 was incomplete because it was lacking banner. I would like everyone to know that I have rectified this situation by downloading a Perl version of banner from: http://www.perl.com/language/ppt/ I hope Tim can find it in his heart to forgive me. :) (If you don't get this, you had to be at the meeting last night.) From scott at illogics.org Fri Jan 17 11:23:30 2003 From: scott at illogics.org (Scott Walters) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: Banner In-Reply-To: from "Doug Miles" at Jan 17, 2003 10:04:54 AM Message-ID: <200301171723.h0HHNWiv028766@slowass.net> Thanks to everyone that showed up. I hope I struck a good balance between interesting and understandable. As always, if there is something you didn't understand, don't blame yourself, blame me. I'll be happy to clarify =) It should be noted that Tim took the show with his entry. I later found my bug. It was a stupid one, of course. It was cool seeing so many laptops - thanks to Heather and Bill for the spares. Finally, I'd like to thank my agent, my producer, my manager, my handler, my supplier, my shrink, my ... Hrm. elm has no "reply to all" command. When I reply to someone on the list, I have to manually cc it to phoneix-pm-list@happyfunball.pm.org. At first I found it annoying, but now I'm starting to think that all mailers should work this way =) -scott > > > It was brought to my attention last night by Tim Ayers that my stock > install of RedHat 7.3 was incomplete because it was lacking banner. I > would like everyone to know that I have rectified this situation by > downloading a Perl version of banner from: > > http://www.perl.com/language/ppt/ > > I hope Tim can find it in his heart to forgive me. :) > > (If you don't get this, you had to be at the meeting last night.) > > From doug.miles at bpxinternet.com Fri Jan 17 11:44:35 2003 From: doug.miles at bpxinternet.com (Doug Miles) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: Presentation Idea Message-ID: <3E284103.3030805@bpxinternet.com> Great meeting last night! A future presentation that I'd like to see is covering new features of Perl. I just realized that there are probably features in 5.6.x (let alone 5.8.x) that I don't know anything about. So if anyone wants to take this on,let us know. From Mark.Pease at motorola.com Fri Jan 17 11:52:51 2003 From: Mark.Pease at motorola.com (Mark Pease) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: Presentation Idea References: <3E284103.3030805@bpxinternet.com> Message-ID: <3E2842F3.445EDF8C@motorola.com> Doug Miles wrote: > > Great meeting last night! A future presentation that I'd like to see is > covering new features of Perl. I just realized that there are probably > features in 5.6.x (let alone 5.8.x) that I don't know anything about. > So if anyone wants to take this on,let us know. Heh... Take a look at: http://www.perl.com/pub/a/2003/01/16/whatsnew.html This went up late yesterday on perl.com. -- Mark Pease Mark.Pease@motorola.com Motorola DigitalDNA(tm) Laboratories perl@perl.sps.mot.com 2100 E. Elliot Rd. Phone: (480) 413-3919 Mail Stop: AZ34 EL741 Tempe, AZ 85284 Pager: (800) 381-3304 Fax: (480) 413-7918 Co-Author (with Carl Dichter) of "Software Engineering with Perl" From doug.miles at bpxinternet.com Fri Jan 17 12:35:57 2003 From: doug.miles at bpxinternet.com (Doug Miles) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: Presentation Idea References: <3E284103.3030805@bpxinternet.com> <3E2842F3.445EDF8C@motorola.com> Message-ID: <3E284D0D.8070108@bpxinternet.com> Mark Pease wrote: > Doug Miles wrote: > >>Great meeting last night! A future presentation that I'd like to see is >>covering new features of Perl. I just realized that there are probably >>features in 5.6.x (let alone 5.8.x) that I don't know anything about. >>So if anyone wants to take this on,let us know. > > > Heh... Take a look at: > > http://www.perl.com/pub/a/2003/01/16/whatsnew.html > > This went up late yesterday on perl.com. > Ooooo... spooky. :) From scott at illogics.org Sat Jan 18 14:53:04 2003 From: scott at illogics.org (Scott Walters) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: TinyWiki presentation - notes Message-ID: <200301182053.h0IKr6FV007652@slowass.net> For those of you who missed the presentation (and those of you who were absolutely enthralled, I've placed the notes up on... http://wiki.slowass.net/?TinyWikiPresentation ...and of course the code is available from... http://wiki.slowass.net/?self Yes, I know, I've been obsessing about Wiki for a long time now. Sorry about that =)~ -scott From djmilesfamily at earthlink.net Sat Jan 18 19:55:50 2003 From: djmilesfamily at earthlink.net (Doug and Julie Miles) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: TPJ for December Message-ID: <5.1.1.6.0.20030118185118.03fcc070@mail.earthlink.net> Does anyone have the Perl Journal PDF for December, or know how to find it on the TPJ site? I wasn't paying attention, and missed it. I contacted TPJ, but got no response. Thanks! From intertwingled at qwest.net Sun Jan 19 12:10:01 2003 From: intertwingled at qwest.net (intertwingled) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: TinyWiki presentation - notes In-Reply-To: <200301182053.h0IKr6FV007652@slowass.net> Message-ID: <3.0.6.32.20030119111001.008b8e90@pop.phnx.qwest.net> You are tweaky about Wiki At 12:53 PM 1/18/03 -0800, you wrote: > >For those of you who missed the presentation (and those of you who were >absolutely enthralled, I've placed the notes up on... > >http://wiki.slowass.net/?TinyWikiPresentation > >...and of course the code is available from... > >http://wiki.slowass.net/?self > >Yes, I know, I've been obsessing about Wiki for a long time now. >Sorry about that =)~ > >-scott > > -- even the safest course is fraught with peril From scott at illogics.org Tue Jan 21 12:32:22 2003 From: scott at illogics.org (Scott Walters) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: "AWK has to be better at some things" Message-ID: <200301211832.h0LIWOjR007976@slowass.net> Last meeting, I mentioned that I had an awk script that I hadn't found a good perl equivilent for. I had started to rewrite in Perl, but decided that awk was just doing too good of a job. This script sorts through a directory full of files, each file representing one player on WeeHours, http://weehours.net . #!/bin/sh LINES=17 PLAYERS=/home/ah/players/ #while true;do cd $PLAYERS echo "------------------------------------------------------------------------" echo -n 'Number of active players: ' echo `ls -1 *.o | wc | awk '{ print $1}'` echo 'List of top players as of: ' `date` echo " Level 20 with 10 quests solved is Wizard" echo "------------------------------------------------------------------------" echo "" echo "Name Lev Hours Name Lev Hours Name Lev Hours" echo "" egrep '^level |^cap_name |^age ' *.o | sort -r | sed -e 's/\.o:/ /' -e 's/\"//g' | awk ' $2=="level" { level = $3; } $2=="cap_name" { name = $3; } $2=="age" { if ( level < 21 ) printf("%-11s %2d %3d\n", name, level, $3/30/60); age = 0; name = "XX"; }' | sort -r -b -n -k 2 | grep -v -e '[A-Z][a-z]*test ' | pr -l$LINES -t -3 | head -$LINES The format of the player file is something like (just a brief section of it): money 10038 name "phaedrus" is_npc 0 brief 0 level 10000 armour_class 3 hit_point 20131 race "elf" max_hp 20131 max_sp 20154 experience 37493980 msgin "arrives" msgout "leaves" mmsgout "disappears in a puff of smoke" Each variable that isn't declared static in a player object is saved to file in this way. The awk script knows which order the variables appear in (and hence what order they are written to file in). It looks for level, name, and then age, in that order. Actually, it only required that age appear last. sort and pr do most of the actual work in making the report look pretty: ------------------------------------------------------------------------ Number of active players: 705 List of top players as of: Tue Jan 21 10:18:18 PST 2003 Level 20 with 10 quests solved is Wizard ------------------------------------------------------------------------ Name Lev Hours Name Lev Hours Name Lev Hours Xell 20 189 Nadja 18 333 Diablos 12 197 X 20 65 Milenko 18 91 Sedriiche 11 51 Twoe 20 144 Glitz 18 476 Scrottie 11 548 Solo 20 340 Yllek 17 49 Maveric 11 60 Simester 20 83 Cas 17 63 Grover 11 5 Simedrus 20 59 Nosejob 16 52 Arnin 11 52 Run 20 584 Loki 16 31 Plob 10 17 Prometheus 20 419 Seth 14 47 Ireland 10 0 Marduk 20 36 Tacobelldog 13 116 Deathtoall 10 123 Hargrove 20 113 Shade 13 37 Bartuc 10 36 Darkone 20 101 Reflection 13 46 Wynn 9 83 Darian 20 58 Jule 13 32 Skyfox 9 8 Vanilla 19 136 Demitri 13 13 Kitten 9 14 Urethra 19 134 Chath 13 82 Fall 9 51 Tetrahagael 19 39 Aguila 13 2225 Darkness 9 12 Kelly 19 72 Roloc 12 18 Rapture 8 37 Stimpy 18 409 Lestat 12 15 Kodiac 8 32 I'm not suggesting that anyone rewrite this script for me, and it really doesn't matter - the game is already composed of hundreds of missmatched parts and years of kludges. It has survived a few major philosophical transitions in programming. There might be a better Perl solution, but awk, sh, etc should not be discounted. -scott From phx-pm-list at grueslayer.com Mon Jan 27 09:26:27 2003 From: phx-pm-list at grueslayer.com (David A. Sinck) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: humor Message-ID: <15925.20387.99065.407444@magnitude.righthandgraphics.com> Thought y'all'd enjoy this. A few nights ago, my eldest comes to me at 2 AM or so in the morning complaining about roaming nightmare shadows. Being the dutiful father, I stumble out of bed and go look for said thing. son: "It was about this big." [gesture] me: "mmmph" son: "It looked like Randal." me: [WTFO, he knows Schwartz?] [0.5 second passes] me: [OH. That Randal. From Monsters Inc, the purple iguana monster.] me: Ok, look, Randal's not here. [...] I'd like to point out that not until much later did it occur to me that it didn't bother me in the least to think that Schwartz being the source of nightmares. I mean, who hasn't woken up screaming their first few nights alone with a creepy regex? David From intertwingled at qwest.net Mon Jan 27 10:18:05 2003 From: intertwingled at qwest.net (intertwingled) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: humor In-Reply-To: <15925.20387.99065.407444@magnitude.righthandgraphics.com> Message-ID: <3.0.6.32.20030127091805.00911a50@pop.phnx.qwest.net> Randal's head IS the size and shape of a pumpkin... kinda scary. Tony At 08:26 AM 1/27/03 -0700, you wrote: > >Thought y'all'd enjoy this. > >A few nights ago, my eldest comes to me at 2 AM or so in the morning >complaining about roaming nightmare shadows. Being the dutiful >father, I stumble out of bed and go look for said thing. > >son: "It was about this big." > [gesture] >me: "mmmph" >son: "It looked like Randal." >me: [WTFO, he knows Schwartz?] > [0.5 second passes] >me: [OH. That Randal. From Monsters Inc, the purple iguana > monster.] >me: Ok, look, Randal's not here. [...] > >I'd like to point out that not until much later did it occur to me >that it didn't bother me in the least to think that Schwartz being the >source of nightmares. > >I mean, who hasn't woken up screaming their first few nights alone >with a creepy regex? > >David > > > > > > -- even the safest course is fraught with peril From intertwingled at qwest.net Mon Jan 27 10:18:05 2003 From: intertwingled at qwest.net (intertwingled) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: humor In-Reply-To: <15925.20387.99065.407444@magnitude.righthandgraphics.com> Message-ID: <3.0.6.32.20030127091805.00911a50@pop.phnx.qwest.net> Randal's head IS the size and shape of a pumpkin... kinda scary. Tony At 08:26 AM 1/27/03 -0700, you wrote: > >Thought y'all'd enjoy this. > >A few nights ago, my eldest comes to me at 2 AM or so in the morning >complaining about roaming nightmare shadows. Being the dutiful >father, I stumble out of bed and go look for said thing. > >son: "It was about this big." > [gesture] >me: "mmmph" >son: "It looked like Randal." >me: [WTFO, he knows Schwartz?] > [0.5 second passes] >me: [OH. That Randal. From Monsters Inc, the purple iguana > monster.] >me: Ok, look, Randal's not here. [...] > >I'd like to point out that not until much later did it occur to me >that it didn't bother me in the least to think that Schwartz being the >source of nightmares. > >I mean, who hasn't woken up screaming their first few nights alone >with a creepy regex? > >David > > > > > > -- even the safest course is fraught with peril From doug.miles at bpxinternet.com Mon Jan 27 11:07:42 2003 From: doug.miles at bpxinternet.com (Doug Miles) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: humor References: <15925.20387.99065.407444@magnitude.righthandgraphics.com> Message-ID: <3E35675E.4050700@bpxinternet.com> David A. Sinck wrote: > Thought y'all'd enjoy this. > > A few nights ago, my eldest comes to me at 2 AM or so in the morning > complaining about roaming nightmare shadows. Being the dutiful > father, I stumble out of bed and go look for said thing. > > son: "It was about this big." > [gesture] > me: "mmmph" > son: "It looked like Randal." > me: [WTFO, he knows Schwartz?] > [0.5 second passes] > me: [OH. That Randal. From Monsters Inc, the purple iguana > monster.] > me: Ok, look, Randal's not here. [...] > > I'd like to point out that not until much later did it occur to me > that it didn't bother me in the least to think that Schwartz being the > source of nightmares. > > I mean, who hasn't woken up screaming their first few nights alone > with a creepy regex? > > David As a father and a Perl hacker, thanks for the laugh! From doug.miles at bpxinternet.com Wed Jan 29 15:43:08 2003 From: doug.miles at bpxinternet.com (Doug Miles) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: Apache::Session::Lock::File under mod_perl References: <3E25B2BF.9010305@bpxinternet.com> Message-ID: <3E384AEC.2020404@bpxinternet.com> Doug Miles wrote: > Anyone out there using Apache::Session::File and > Apache::Session::Lock::File under mod_perl? The problem I am having is > that Apache::Session::Lock::File is aparently not releasing the > exclusive lock when the tied hash goes out of scope. I can get a simple > test program to work from the command line, but when I try it under > mod_perl, it locks on the tie to the existing session. If I delete the > lock file, it works fine. Any ideas? > > As a follow-up to this question... I just replaced Apache::Session with CGI::Session, and I think it is experiencing the same problem with different symptoms. When I would try to read in the session, it would create a new one as if I was passing it an undef. I was able to get it to work by explicitly doing a close on the session before doing the internal redirect. I failed to mention the internal redirect in the above problem, and I think that it is the crucial piece of information. I'm guessing that the process that launches the redirect process must stick around until the redirect process is completed. Any thoughts? From doug.miles at bpxinternet.com Wed Jan 29 16:00:57 2003 From: doug.miles at bpxinternet.com (Doug Miles) Date: Thu Aug 5 00:16:55 2004 Subject: Phoenix.pm: tie problem under mod_perl Message-ID: <3E384F19.60206@bpxinternet.com> I'm having a strange problem with tie under mod_perl. I don't know that it is a mod_perl problem, but here is what's happening: I have a hash in package 'buildpage' that I need to tie to Tie::IxHash. The tieing is done from package 'BuildPage::Template'. This appears to work. When I try to access the hash from package 'BuildPage::Tag', it is empty, and tied returns a false value. In fact, I found that it doesn't seem to matter which module I am tieing to. But, if I comment out the tie, the hash becomes visible again. I have tried setting up the same scenerio from the command line, and it works perfectly. Any ideas?