Apache::Session problems, film at 11:00 ...
All:
I'm getting very odd behavior out of Apache::Session, with
serious problems using both the MySQL and File variants.
Yes, I know I've come here with this problem before. Sigh.
I even fixed it, although it was one of those things where I
didn't quite know why it started working. Anyway, it stopped
working about a week ago, and, as usual, I have no clue. Hence
this plea for help:
With Apache::Session::File, this code creates a new session id
with every request. The lock file for each session remains in
the lock directory. I ran a 'chmod -R 777 ' on both
the session store and lock directories.
With Apache::Session::MySQL, this code behaves more normally:
it reuses the session id, the way [I believe] it should, except
$session{state} never seems to make it into the database. I
say that because I look at the contents of the sessions table
between transactions, and it looks like this:
mysql> select * from sessions;
+--+---+
| id |
a_session |
+--+---+
| 4def39f4e8144aede90532951232c040
| |
+--+---+
1 row in set (0.00 sec)
I did make sure that the right privileges existed for the
database user accessing the sessions table.
I tried uninstalling Apache::Session ('rm -rf
/usr/local/lib/perl5/site_perl/5.6.0/Apache/Session*'),
and reinstalled it using CPAN, on the theory that I may have
diddled it while checking out its code. But that didn't help.
Here's the (relevant) code, with short, annotated, log extract
following:
##
## Physemp::Search
##
package Physemp::Search;
use strict;
use Apache;
use Apache::Request;
use Apache::Constants qw( :common );
use CGI::Cookie;
use Apache::Session::MySQL;
use DBI;
use Data::Dumper;
my (%states, %_CACHE);
sub handler ($$) {
my ($class, $q) = @_;
my $self = $class->new(my $r = Apache::Request->new($q));
my $html = '';
$self->get_session($r);
my $coderef =
$self->{make}->{$self->frame}->{$self->page}->{$self->command}
|| \&unimplemented;
$html = $self->$coderef($r);
$r->content_type('text/html');
$self->put_or_del_session($r);
$r->send_http_header;
print $html;
return OK;
}
sub get_session {
my ($self, $r) = @_;
my %session;
my $cookie_str = $r->header_in('Cookie');
my %cookies = $cookie_str eq '' ? ( ) :
CGI::Cookie->parse($cookie_str);
if (exists $cookies{SessionID}) {
(my $session_id = $cookies{SessionID}->value) =~ s/([0-9a-f]+)/$1/;
eval {
tie %session, $self->{tieclass}, $session_id, $self->{tieattrs};
};
if ($@) {
$r->log_error($@);
$r->log_error("get_session: No session data found.");
$self->{state} = { };
$self->{session_id} = '';
} else {
$r->log_error("get_session: Session data found.");
$r->log_error("get_session: \$session{state} is \n", Dumper
$session{state});
$session{state} = { account => {} } unless exists $session{state};
$self->{session_id} = $session{_session_id};
$self->{state} = $session{state};
}
undef %session;
} else {
$r->log_error("get_session: No Session ID cookie.");
$self->{state} = { };
$self->{session_id} = '';
}
$r->log_error("get_session: Session ID is '$self->{session_id}'.");
$r->log_error("get_session: State is \n", Dumper $self->{state});
}
sub put_or_del_session {
my ($self, $r) = @_;
my (%session, $cookie);
if ($self->command eq 'make' or $self->page eq 'action') {
eval {
tie %session,
$self->{tieclass},
($self->{session_id} eq '' ? undef : $self->{session_id}),
$self->{tieattrs};
};
if ($@) {
$r->log_error("put_or_del_session: $@");
eval { tie %session, $self->{tieclass}, undef, $self->{tieattrs};
};
if ($@) {
$r->log_error("put_or_del_session: $@");
return; # WTH, we can't do any good here
}
}
if ($self->command eq 'logout') {
$r->log_error("put_or_del_session: deleting session.");
$cookie = CGI::Cookie->new( -name=> 'SessionID',
-path=> $self->{uri},
-domain => '.physemp.com',
-expires => '-10m',
-value => '' );
tied(%session)->delete;
} else {
$r->log_error("put_or_del_session: updating session.");
$session{state} = $self->{state};
$session{changes}++;
$r->log_error("put_or_del_session: Session ID is
'$session{_session_id}'.");
$r->log_error("put_or_del_session: State is \n", Dumper
$session{state});
$co
Re: an unusual [job request] + taking mod_perl to the commercial world
I would like to restate that while I think these engines are cool and useful, that they are not the things that bring the masses to your platform. This was the point I was making. I am not naysaying projects like Enhydra, but just stated that they are not as directly useful for bringing the masses to the platform. While it is true that an "Enhydra" type of engine makes writing application easier, what you really still always need in order to gain a critical mass is something more concrete that the masses can hook onto. I am not talking about techies loving mod_perl or Enhydra or AxKit. But everyday webmasters and CIOs saying "XYZ platform has so many applications for it I can see them demoed, my tech staff can install them within a day" so let's use it. There are just certain things that are harder to market than others. Applications make platforms easier to market because it shows off the power. I was not at the meeting, but I heard Stas convinced one of our clients to go with mod_perl by showing them a site he created called SinglesHeaven in CGI and then in mod_perl. "Look how fast it is and you can see it's a real application". Showing the same people benchmarks of hello world and template renderings generally do not have the same effect. At 11:06 AM 4/28/01 -0400, Bakki Kudva wrote: >On Sat, 28 Apr 2001 09:14:10 +0100 (BST) >Matt Sergeant <[EMAIL PROTECTED]> wrote: > >Amen to that and there is Enhydra on the Java side. To get the >functionality of these two frameworks I'd have to integrate many many CPAN >modules, keep track of various versions, make sure each is active etc etc. >A nice application framework like Enhydra or zope on mod_perl which is >maintained perhaps by all the authors of individual modules would be a >great start. >bakki > > > Actually there's an exception to this rule. Look at Zope. > >
Re: an unusual [job request] + taking mod_perl to the commercial world
At 09:14 AM 4/28/01 +0100, Matt Sergeant wrote: >On Sat, 28 Apr 2001, Gunther Birznieks wrote: > > > As I think I mentioned, it's great that the people like you on this list > > have a passion for delivering cool software. > > >[snipped] > > > > People rarely look at toolkits like payment gateways and messaging servers > > unless there is an application that fits their needs that they can use > > which happens to use these backend components. > >Actually there's an exception to this rule. Look at Zope. But Zope has an application? -- content management. A template engine is not an application, but a content management tool built upon templates surely is? I thought you would recognize this as you are building something to allow this on AxKit? Of course, I guess you could consider AxKit an application because presumably it comes with scripts to allow aggregration of news content in RSS format? I consider this a really nice application. But it's also a bit difficult to tell that this is what AxKit does. You might consider separating AxKit the engine from AxKit the applications to allow people to find your site looking for applications (eg news, content management) so that they want to use AxKit. And then when they want to use AxKit, they will want to use mod_perl.
Re: an unusual [job request] + taking mod_perl to the commercial world
At 09:06 AM 4/28/01 -0400, barries wrote: > > But then this gets into the Collabnet SourceXChange model which Collabnet > > just shut down due to lack of capability of making money (I guess). So > I am > > not sure if this would bode well for this type of model on mod_perl. > >Not sure either, except that this would be more targeted towards >specific research projects, as opposed to the build a venue and they >will come SourceXchange model. I guess my understanding of SourceXChange is that it was also for research projects and additions to existing open source projects. So the idea is to precisely bring together open source developers and people who want to give grants for adding or creating specific open source projects. Anyway, I am not saying it is impossible. Anyway, I think what Stas may want to do is write up a list of things that he wants to do with mod_perl if he were given 6 months or a year of salary to do whatever he wanted. This would make such a request to have someone pay for his work to be concrete and possibly then endear others to help in a more concrete way. It is also quite possible that people do not realize what improvements should be made to mod_perl (or they would be doing it) and so if Stas has a list of ideas of what to do with mod_perl, then he may inadvertently spark ideas in other people's minds as to what can be done and they would do it. I guess off the top of my head, there are things that Doug has stated he expects other people to write as soon as mod_perl 2.0 is released. So perhaps that could be worked on in parallel with mod_perl 2.0 so that when Apache 2.0 is released, mod_perl 2.0 can be released in a very full-featured way. Later, Gunther
Re: cookies work for some browsers, not for others... ?
On Sun, Apr 29, 2001 at 12:21:33AM +0200, Robin Berjon wrote:
> At 17:17 28/04/2001 -0500, will trillich wrote:
> >so i guess what you're saying is, some browsers look for
> >a "redirect:" header and then charge off to the new location
> >without handling any "set-cookie:" headers in the meantime?
>
> Precisely. And some also don't report the cookie before the second page
> after the redirect (presumably because they consider it to be the same
> request). I think that behaviour only happens with permanent redirects though.
>
> One thing that helps (often, not always) is to make sure that your
> Set-Cookie header is sent before the Location header of the redirect.
here's the code, direct from the modperl book, and downloaded in
person from modperl.com:
package Apache::TicketAccess;
use strict;
use Apache::Constants qw(:common);
use Apache::TicketTool ();
sub handler {
my $r = shift;
my $ticketTool = Apache::TicketTool->new($r);
my($result, $msg) = $ticketTool->verify_ticket($r);
unless ($result) {
$r->log_reason($msg, $r->filename);
my $cookie = $ticketTool->make_return_address($r);
$r->err_headers_out->add('Set-Cookie' => $cookie);
return FORBIDDEN;
}
return OK;
}
1;
__END__
i suppose i'd need to change
return FORBIDDEN;
to
print htmlheaders-including-meta-refresh, brief-html-stuff;
return OK;
right?
--
don't visit this page. it's bad for you. take my expert word for it.
http://www.salon.com/people/col/pagl/2001/03/21/spring/index1.html
[EMAIL PROTECTED]
http://sourceforge.net/projects/newbiedoc -- we need your brain!
http://www.dontUthink.com/ -- your brain needs us!
Re: $dbh->disconnect with Apache::DBI? (was Re: Failed requests inbenchmark)
Philip Mak wrote: > > I noticed something weird in my database logs, though: > > 010427 22:36:41 Aborted connection 2544 to db: 'animelyrics' user: > 'animel' host: `localhost' (Got an error reading communication packets) > 010427 22:37:14 Aborted connection 2546 to db: 'animelyrics' user: > 'animel' host: `localhost' (Got an error reading communication packets) > 010427 22:54:11 Aborted connection 2601 to db: 'animelyrics' user: > 'animel' host: `localhost' (Got an error reading communication packets) > > Reading some other mailing list messages suggests that I did not do > $dbh->disconnect() properly. But I'm using Apache::DBI, so should I need > to do that? > I would expect to see these messages if your web scripts were issuing database calls and the httpd processes got killed with a apachectl stop/start. There might then be aborted connections at the database end. There is probably a graceful way to issue a full stop, before a clean start ( not apachectl graceful ), but I don't know what it is. --Josh _ Joshua Chamas Chamas Enterprises Inc. NodeWorks >> free web link monitoring Huntington Beach, CA USA http://www.nodeworks.com1-714-625-4051
Re: cookies work for some browsers, not for others... ?
At 17:17 28/04/2001 -0500, will trillich wrote: >so i guess what you're saying is, some browsers look for >a "redirect:" header and then charge off to the new location >without handling any "set-cookie:" headers in the meantime? Precisely. And some also don't report the cookie before the second page after the redirect (presumably because they consider it to be the same request). I think that behaviour only happens with permanent redirects though. One thing that helps (often, not always) is to make sure that your Set-Cookie header is sent before the Location header of the redirect. ___ Robin Berjon <[EMAIL PROTECTED]> -- CTO k n o w s c a p e : // venture knowledge agency www.knowscape.com --- "Many people would sooner die than think. In fact, they do." - Bertrand Russell
Re: cookies work for some browsers, not for others... ?
On Sat, Apr 28, 2001 at 12:54:17PM -0700, Jim Winstead wrote: > in general, your problem with some browsers that otherwise support > cookies may be with issuing redirects and cookies on the same request, > which has been known to trip up some browsers. the easy workaround is > to use a refresh to do the redirection. pooh. i'll look into that. hmm. they all do the redirect properly, but when they arrive at the redirected url, they don't seem to have (or at least report) the cookies they've been given. so i guess what you're saying is, some browsers look for a "redirect:" header and then charge off to the new location without handling any "set-cookie:" headers in the meantime? > fmt: w70: No such file or directory hmm? -- [EMAIL PROTECTED] http://sourceforge.net/projects/newbiedoc -- we need your brain! http://www.dontUthink.com/ -- your brain needs us!
Help -- Internal Server Error after mod_perl and new apache install on HP-UX 10.20.
Greetings. I've spent 3 days now (please don't laugh too hard) trying to get ANY web pages to work on my HP-UX 10.20 box, after installing the latest mod_perl and apache server and compiling from source. Below are the typical error messages in error_log, the output from "httpd -l", and my httpd.conf. I'm trying to open up the access to my box (within my firewall) for testing to force something to work -- unfortunately I'm not getting anywhere. I think I have tried every build combination of the modules, but I always have a problem with the mod_perl test (also shown below). Is there a way to get a higher level of tracing in the error_log so I can know where the "internal error" occurs? What I'm trying to implement is LXR (Linux Cross Reference) which I think would benefit from mod_perl. Any suggestings appreciated. Thanks, ronan +++ [Sat Apr 28 09:57:18 2001] [notice] Apache/1.3.19 (Unix) mod_perl/1.25 configure d -- resuming normal operations [Sat Apr 28 09:57:18 2001] [info] Server built: Apr 28 2001 01:01:40 [Sat Apr 28 09:57:23 2001] [error] [client 15.17.158.220] internal error: / [Sat Apr 28 09:57:27 2001] [error] [client 15.17.158.220] internal error: /LXR/ [Sat Apr 28 10:08:59 2001] [error] [client 15.17.158.220] internal error: /LXR [Sat Apr 28 10:09:03 2001] [error] [client 15.17.158.220] internal error: / [Sat Apr 28 10:14:30 2001] [error] [client 15.17.158.220] internal error: /LXR/ [Sat Apr 28 10:14:37 2001] [error] [client 15.17.158.220] internal error: / [Sat Apr 28 10:14:44 2001] [error] [client 15.17.158.220] internal error: /tmp [Sat Apr 28 10:14:48 2001] [error] [client 15.17.158.220] internal error: /tmp/l xr [Sat Apr 28 10:15:16 2001] [notice] child pid 28610 exit signal Bus error (10) --- # ../bin/httpd -l Compiled-in modules: http_core.c mod_so.c mod_perl.c suexec: disabled; invalid wrapper /opt/apache/bin/suexec ---
Re: brochureware perl.apache.org?!
At 04:41 PM 4/28/01 -0400, Philip Mak wrote: >On Sat, 28 Apr 2001, Drew Taylor wrote: > > > I agree 100%. If I might throw my $.02 in, IMHO a part of this marketing > > should be a more "brochureware" perl.apache.org. > >If you guys do redesign perl.apache.org, please, PLEASE take usability >into account. http://www.useit.com/ is a great resource about how to make >a usable website. Agreed! I assumed usability would be taken into account in any new design. :-) Two jobs ago I worked for a web development shop whose primary focus was on user-centered design, so I know all about keeping the end user in mind. Drew Taylor mailto:[EMAIL PROTECTED] http://www.drewtaylor.com/
brochureware perl.apache.org?!
On Sat, 28 Apr 2001, Drew Taylor wrote: > I agree 100%. If I might throw my $.02 in, IMHO a part of this marketing > should be a more "brochureware" perl.apache.org. If you guys do redesign perl.apache.org, please, PLEASE take usability into account. http://www.useit.com/ is a great resource about how to make a usable website. I've seen too many corporate websites that load slow, and is hard to find any *useful* information in them. A bad offender that I can think of off the top of my head is http://www.advertising.com/. -Philip Mak ([EMAIL PROTECTED])
Re: an unusual [job request] + taking mod_perl to the commercialworld
On Sat, 28 Apr 2001, Bakki Kudva wrote: > On Sat, 28 Apr 2001 09:14:10 +0100 (BST) > Matt Sergeant <[EMAIL PROTECTED]> wrote: > > Amen to that and there is Enhydra on the Java side. To get the > functionality of these two frameworks I'd have to integrate many many CPAN > modules, keep track of various versions, make sure each is active etc etc. > A nice application framework like Enhydra or zope on mod_perl which is > maintained perhaps by all the authors of individual modules would be a > great start. Actually there are probably more than one project of this kind, but we're working on something like Zope (with more of a focus on content management) at AxKit. You can see some information about it on http://axkit.com/products/ -- /||** Founder and CTO ** ** http://axkit.com/ ** //||** AxKit.com Ltd ** ** XML Application Serving ** // ||** http://axkit.org ** ** XSLT, XPathScript, XSP ** // \\| // ** mod_perl news and resources: http://take23.org ** \\// //\\ // \\
Re: cookies work for some browsers, not for others... ?
in general, your problem with some browsers that otherwise support cookies may be with issuing redirects and cookies on the same request, which has been known to trip up some browsers. the easy workaround is to use a refresh to do the redirection. fmt: w70: No such file or directory On Sat, Apr 28, 2001 at 02:44:29PM -0500, will trillich wrote: > ??? > what's that trailing zero for (or from), by the way? and that > "cf" that preceeds ??? chunked encoding. http://www.w3.org/Protocols/rfc2616/rfc2616-sec3.html#sec3.6.1 jim
Re: an unusual [job request] + taking mod_perl to the commercial world
barries wrote: > > Anyway, this seems promising. Where I know we wouldn't pay money to fund an > > entire year of Stas developing mod_perl solely, I certainly know that there > > are probably features I would seriously consider sponsoring. > > Any others out there that might be interested, let's hear from you :-). > My employer's on the rocks, so no easy money there I'm finding (while doing consulting work) that, if I already have an "in" with a firm, I can usually leverage that to work in time training the staff in how to use mod_perl indirectly, by teaching them Embperl first. Embperl ends up being the draw. I find I can take average Perl programmers and get them up and run- ning with Embperl, basically in two sessions. And once they figure out what all they can do - and how quickly all their Perl scripts can transfer - they buy right in. The main hurdle turns out to be working a build of mod_perl into the servers they're using. Often quite a number of developers are all at work, and they don't all merit the kind of trust that mod_perl requires. Mod_perl, though, is a pretty trusting module. -- Richard Goerwitz [EMAIL PROTECTED] tel: 401 438 8978
cookies work for some browsers, not for others... ?
i've been tinkering with the modperl book examples for Apache::Ticket*.pm (as described p305-322)... it works for linux/konqueror linux/netscape win/explorer it doesn't work for linux/lynx mac/netscape the ones that do work get to the login page with two textfields (username and password) with a login button; those that don't work get a "your browser doesn't accept cookies" page. where can i find some pointers on the differences between various browsers' adherence to standards, and hints on workarounds? (or, is this something else i've stumbled into?) -- here's a telnet trace of the situation, so apparently the program logic is working in all instances; but some browsers just won't play nice-- $ telnet sample-from-modperl-book.com 80 Trying ##.##.##.##... Connected to sample-from-modperl-book.com. Escape character is '^]'. GET /try HTTP/1.1 Host: sample-from-modperl-book.com HTTP/1.1 302 Found Date: Sat, 28 Apr 2001 19:26:48 GMT Server: Apache/1.3.9 Set-Cookie: request_uri=http%3A%2F%2Fsample-from-modperl-book.com%2Ftry%3F; domain=.sample-from-modperl-book.com; path=/ Location: http://www.sample-from-modperl-book.com/login Transfer-Encoding: chunked Content-Type: text/html; charset=iso-8859-1 cf 302 Found Found The document has moved http://www.sample-from-modperl-book.com/login";>here. 0 Connection closed by foreign host. ??? what's that trailing zero for (or from), by the way? and that "cf" that preceeds Log In Please Log In Name Password http://sample-from-modperl-book.com/try?";>Note: You must set your browser to accept cookies in order for login to succeed.You will be asked to log in again after some period of time has elapsed. 0 Connection closed by foreign host. ??? and here it's bracketed with "294" in front, and "0" again taking up the rear. what's up with that? ??? -- don't visit this page. it's bad for you. take my expert word for it. http://www.salon.com/people/col/pagl/2001/03/21/spring/index1.html [EMAIL PROTECTED] http://sourceforge.net/projects/newbiedoc -- we need your brain! http://www.dontUthink.com/ -- your brain needs us!
Re: an unusual [job request] + taking mod_perl to the commercial world
At 14:04 28/04/2001 -0400, Drew Taylor wrote: >I remember seeing some proofs done by Robin Berjon (I'm sure I'm not >spelling it right!) long ago that I really liked. But they were never used >AFAIK. I also registered modperlnews.(org|com) a while back with the >intention of doing something useful with them, but I have not yet done >anything. I also have some time on my hands while I'm searching for a new >job, so I am volunteering my time if it's needed or wanted. I'll bet that >Template Toolkit (thanks Andy!) would work wonderfully for putting a new >face on the site. If the powers that be are interested, I am available to >help. You got the spelling right, and if you hadn't I don't think it would have been a big deal at all. In fact, my (geek) friends call me Berfon since that is how my name appeared in the first Perl Poetry contest that I was credited in ;-) The proofs never made it to the perl.apache.org site for many reasons, one of which being that that's when I changed continents again, another being that there were other people working on it too, and there are probably many other reasons at hand too. Anyway, the proof eventually got used for take23.org which I'm very happy about (well, it was just a proof and there are quite a few things I'd like to change about it, but that's another story). I also helped with axkit.org. I don't think it's professional level as is, but it's probably better than "programmer style" and until I or someone else has the time to make it better I hope it does the job ok. I'm not going to promise to do things I don't have the time to do, a few years spent in the Open Source community have taught me (the hard way) about that. However, now that my company has grown (and keeps growing despite rumours of a slump -- modperl + xml definitely help) and that I have internal resources other than myself I can offer to create a new proof. The limits of what I can offer are a design proof + templates (in xslt). I don't have enough people yet to take over the maintenance of the whole site. If there's a way we can have some XML publishing on that server then it would be done in little time (note that xml means pod too, thanks to pod2xml). If someone can solve the publishing side of the issue, then I can do the design/templating side. ___ Robin Berjon <[EMAIL PROTECTED]> -- CTO k n o w s c a p e : // venture knowledge agency www.knowscape.com --- "What I like about deadlines is the lovely whooshing sound they make as they rush past." --Douglas Adams
Re: an unusual [job request] + taking mod_perl to the commercial world
On Sat, 28 Apr 2001 09:14:10 +0100 (BST) Matt Sergeant <[EMAIL PROTECTED]> wrote: Amen to that and there is Enhydra on the Java side. To get the functionality of these two frameworks I'd have to integrate many many CPAN modules, keep track of various versions, make sure each is active etc etc. A nice application framework like Enhydra or zope on mod_perl which is maintained perhaps by all the authors of individual modules would be a great start. bakki > Actually there's an exception to this rule. Look at Zope. > -- .-.| Bakki Kudva__Open Source EDMS__ oo|| Navaco ph: 814-833-2592 /`'\| 420 Pasadena Drive fax: 603-947-5747 (\_;/) | Erie, PA 16505 http://www.navaco.com/
Re: an unusual [job request] + taking mod_perl to the commercial world
At 10:41 PM 4/27/01 -0400, barries wrote: >On Fri, Apr 27, 2001 at 11:44:49PM +0800, Stas Bekman wrote: > > > > Since mod_perl is an open source, it's a tough quest. Basically what I > > want is get some company that will benefit from me working on open source > > project full time and pay me a salary. Of course it's probably hard to get > > a full time open source position, so probably some compromising offer, > > where we do some 50-75% of the time mod_perl development and the rest > > doing something else, if it makes the company more happy. > >As you and others have mentioned here now and in the past, mod_perl >needs PR and working apps (which are both good PR and good reasons for >others to start using it as a means to an end: like the new Slash code, >for instance). I agree 100%. If I might throw my $.02 in, IMHO a part of this marketing should be a more "brochureware" perl.apache.org. The content there is top notch and I can't say enough good things about it. The amount of documentation is simply amazing and there are many commercial products that aren't as well documented. However, the current site looks like it was designed by a programmer. And I mean no offense, because that is also how my personal site currently looks. :-) I think that perl.apache.org should present a more "professional" face to corporate people who are looking at using perl/mod_perl. For an example, look at the various java.apache.org sites. They have navigation and are split into multiple sections and pages. Someone took the time to section off the various topics and put together graphics and navigation. I'm sure the amount of data/documentation is the same between say Tomcat and mod_perl, but I prefer the packaging of the java sites. If we are to get more corporate interest in mod_perl, the community should work to present the information about mod_perl in such a way that the managers/business people who make the decisions can reason that mod_perl is a technology they should use. We already know mod_perl is great, we just have to make the rest of the world realize it. Marketing is as much about appearance as capabilities. Just look at the marketing depts that dotcoms put together to hype a product that didn't yet exist! I remember seeing some proofs done by Robin Berjon (I'm sure I'm not spelling it right!) long ago that I really liked. But they were never used AFAIK. I also registered modperlnews.(org|com) a while back with the intention of doing something useful with them, but I have not yet done anything. I also have some time on my hands while I'm searching for a new job, so I am volunteering my time if it's needed or wanted. I'll bet that Template Toolkit (thanks Andy!) would work wonderfully for putting a new face on the site. If the powers that be are interested, I am available to help. And if I'm off track, then tell me and I'll go back to lurking. :-) >Such an organization could do (aside from the feature development listed >above): > >- general advocacy: press releases, reference cards, publish > articles and sell a qr/this (article|research) funded by > (ActiveState|VMWare|Covalent|IBM S/390 Marketing Division|.*)/ credit Drew Taylor mailto:[EMAIL PROTECTED] http://www.drewtaylor.com/
Re: Child keeps seg faulting
Hi there, On Fri, 27 Apr 2001, Chuck Carson wrote: > I have no problems on Linux, my exact same build procedure gives me this > error on Solaris 2.7. > > What is the recommended combo for Solaris 2.7? I am currently using this: > apache 1.3.19 > mod_perl 1.25 > php 4.04pl1 > perl 5.6 Er - which error? Segfault? Definitely try static, and searching archives. I get the feeling that PHP and mod_perl on Solaris is an unhappy mix. Anyone care to comment? 73, Ged.
Re: Environment variables in startup.pl
Hi there,
On Fri, 27 Apr 2001, Philip Mak wrote:
> On Fri, 27 Apr 2001, Scott Alexander wrote:
>
> > Should this work in a startup.pl file
> >
> > my $hostname = $ENV{"HOSTNAME"} ;
> > from the prompt I can write echo $HOSTNAME and get the correct hostname of the
>server.
> > But from within startup.pl I don't get it.
>
> Try this:
>
> my $hostname = `/bin/hostname`;
Or better still:
http://perl.apache.org/guide
73,
Ged.
Re: an unusual [job request] + taking mod_perl to the commercial world
On Sat, Apr 28, 2001 at 12:16:18PM +0800, Gunther Birznieks wrote: > > If the person being sponsored by grants is a foreigner to the US, but > wishes to be in the US Agreed, but I don't think that's a roadblock, just a situation to be dealt with. mod_perl infrastructure seems to be the kind of project that can be done on a distributed basis. Geographic proximity helps, no doubt, but that shouldn't stop work. We might need US staff, but I'm not sure where the fingers tap on the keys matters as long as they're talented fingers in the context of a project and community that has already proven to work well in a distributed manner. > Of course, the location of the person or person(s) being sponsored may make > a big difference in being able to be sponsored. Doug, would you sponser Stas a spot on your couch for, say, a 5 year work commitment from him? ;-). > Or would be actually apply a set of grants... I would like a company > to sponsor this XYZ feature. That's what I meant by "feature purchases" below, not that it was at all clear :-/. Though I envisioned it more like talks might be opened with a company and a mutually beneficial feature set might emerge from that. For instance, I've been wondering about using fam/imon[1] to make it so that Apache::Reload and templating and caching systems can have their cake and eat it to: no need to repeatedly stat() the filesystem to know if your source files have change or cached content has become outdated. > But then this gets into the Collabnet SourceXChange model which Collabnet > just shut down due to lack of capability of making money (I guess). So I am > not sure if this would bode well for this type of model on mod_perl. Not sure either, except that this would be more targeted towards specific research projects, as opposed to the build a venue and they will come SourceXchange model. > Anyway, this seems promising. Where I know we wouldn't pay money to fund an > entire year of Stas developing mod_perl solely, I certainly know that there > are probably features I would seriously consider sponsoring. Any others out there that might be interested, let's hear from you :-). My employer's on the rocks, so no easy money there - Barrie [1] http://oss.sgi.com/projects/fam/
Re: an unusual [job request] + taking mod_perl to the commercialworld
On Sat, 28 Apr 2001, Gunther Birznieks wrote: > As I think I mentioned, it's great that the people like you on this list > have a passion for delivering cool software. > > I may have missed the intent of these two posts (micropayment and messaging > engines), but unfortunately, I don't really consider either of these items > to be application-level items. I consider them infrastructure. > > eg it's nice that SmartWorker (as a toolkit) is open source, but opendesk > is closed source. And it's the applications (ironically) in opendesk that > make smart worker worth taking a look at. But it cripples (hope I don't > upset anyone) SmartWorker's marketing to have opendesk closed source > because people prefer downloading apps and then they learn about the > framework it was developed in. Rarely is it the other way around. > > People rarely look at toolkits like payment gateways and messaging servers > unless there is an application that fits their needs that they can use > which happens to use these backend components. Actually there's an exception to this rule. Look at Zope. -- /||** Founder and CTO ** ** http://axkit.com/ ** //||** AxKit.com Ltd ** ** XML Application Serving ** // ||** http://axkit.org ** ** XSLT, XPathScript, XSP ** // \\| // ** mod_perl news and resources: http://take23.org ** \\// //\\ // \\
