Scoop -- the swiss army chainsaw of content management
Front Page Everything News Code Help! Wishlist Project Scoop Sites Dev Notes Latest CVS changes Development Activities
Frequently Asked Questions

This FAQ aims to answer a few questions about Scoop, what is it, whats its for, etc, as well as some install questions. If you think its missing something, or if you would like to add to it (we can always use help with docs :-) then email hurstdog @ kuro5hin.org

General:

Installing:

  • What is CPAN? and how do I get these crazy modules?" CPAN is the Comprehensive Perl Archive Network, its a repository of Perl modules so that perl coders don't have to re-invent the wheel every time they start a project. To get the CPAN modules for scoop, you should use the installer. its in the script/ directory called install.pl It should get everything set up for you. A few caveats though:
    • Make sure when it talks about what to do about dependancies, tell it to 'ask'. Otherwise you might end up with a whole new perl on your system! (sometimes, modules might try that anyway... but they *shouldn't* :(
    • Apache::Session::DBI and DBD::MySQL might fail to install because they can't connect to the database. This is because they don't ask for any test username and password, and the uname and pass they do use rarely exists in mysql installs. So after running install.pl, you might have to go into cpan by hand and install them, with
      perl -MCPAN -e shell
      at the prompt, then at the cpan prompt, run:
      cpan> force install DBD::mSQL
      cpan> force install Apache::Session::DBI
      
      If it wasn't those modules that were causing trouble, then running the above won't hurt anything.

  • I keep getting "Document Contains no Data", whats that mean? This means that somehow apache/scoop is failing without warning and just stopping. One fix I've seen work (it just worked for me) is to go back into cpan, and 'force install' DBD::mSQL, so:
    cpan> force install DBD::mSQL
    and tell it to install the msyql modules only.

  • How do I compile Apache and mod_perl for scoop? First, you should make sure that you have expat version 1.95 or greater installed. That way you don't get headaches with broken RDF's later. Then, get the apache source:
    host:#] lynx --dump http://httpd.apache.org/dist/httpd/apache_1.3.20.tar.gz > apache_1.3.20.tar.gz
    host:#] lynx --dump http://perl.apache.org/dist/mod_perl-1.25.tar.gz > mod_perl-1.25.tar.gz
    
    Now, extract both of those files with 'tar -xzvf filename.tar.gz'. Once you have them extracted you should have 2 more directories, one for each. Now 'cd' into the mod_perl directory with 'cd mod_perl-1.25' and run the following.

    Note: you can also compile apache and mod_perl seperately, though that is outside the scope of this document. If you do, be sure to pass "--disable-rule=EXPAT" to the apache configure script. Otherwise, RDFs and Cron won't work correctly.

    host:#] perl Makefile.PL \
    APACHE_PREFIX=/usr/local/apache \
    APACHE_SRC=../apache-1.3.20/src \
    DO_HTTPD=1 \
    USE_APACI=1 \
    EVERYTHING=1 \
    -->[...perl output...]
    
    host:#] make
    -->[...make output...]
    
    host:#] make test
    -->[...should tell you all tests passed, eventually...]
    
    host:#] make install
    -->[...should say that it's installed in /usr/local/apache...]
    
    And with that, apache and mod_perl will be installed to /usr/local/apache . Now you should just have to run the installer (after MySQL is installed, of course) to finish up the configuration.

Running Scoop

  • Help! RDFs don't work! I can't fetch any from the RDF admin tool! This is because apache, by default, compiles in a modified version of expat (called expat-lite) that differs from the normal expat, which perl's XML parser module uses. Having two different expat's in memory is enough to segfault the apache child. You will see this behavior when running cron with 'op=cron' or re-fetching the RDF feeds from the RDF admin tool. Sadly, the only way to fix this is to recompile apache. Make sure you have expat installed, you can get it from Sourceforge. Then use the instructions here.

  • Cron is segfaulting apache! If when you visit http://yoursite/cron/, apache segfaults, then it is probably an RDF problem. Same as the entry above to fix. Recompile apache and mod_perl after you install expat. See this for help compiling.

  • I keep getting logged out This is because you have an old cookie lying around for your scoop site that is causing trouble. This happens a lot when testing and setting up the site for the first time. To remedy, quit your browser, open up your cookie file and delete all of the entries from your scoop site. Then, log into mysql and delete all of the scoop sessions. So:
    host:#] mysql -u dbuser -p scoopdb
    ...
    mysql> delete from sessions
    ...
    mysql> quit
    host:#]
    
    And then try to log in again.

  • I can't even log in This is probably because of an incorrect cookie_host in your httpd.conf file for scoop. The cookie_host has to have at least 2 periods in it. So 'mysite.com' won't work, but '.mysite.com' would work, it would set the cookie for any site under mysite.com though, so that might not be what you want. Generally what people do, is they set the cookie for 'www.mysite.com' and redirect mysite.com to www.mysite.com in apache, using a directive like the following (note, this is really only a problem if its a vhost):
    <VirtualHost your.ip.addr.>
      ServerName mysite.com
      RedirectPermanant / http://www.mysite.com/
    </VirtualHost>
    


  • I have a user the forgot their password, how do I change it? Make sure you're logged into a Superuser level account, then use the Edit User box in the Admin Tool box to type in their name and click 'prefs'. Scroll down do where you can enter a new password for them. Type in a new password for them, into both boxes. Click [Save], and you're done. Just let them know about the new password.

  • Why do my hotlist links have in them? This happens from time to time, for some random reason, though the fix is easy. Make sure you're logged in as a superuser, then use the box editor to open hotlist_box. Scroll down to near the bottom, where you'll see |rootdir|. Change this to and save it. You can check, but the hotlist links should be okay again.

  • How come my image dir is being appended to the URL This generally only happens with the new URLs, where args are passed differently, and it's caused by having the imagedir var set so that's it's relative, such as "images". In the past, this has worked fine, but now you need to put your rootdir on the front of that. So, if you have a rootdir of "/scoop" and an imagedir of "images", change that to "/scoop/images", or on a vhost install, it'd be more like "/images", even though rootdir is blank.

Menu
create account
faq
search
report bugs
Scoop Administrators Guide
Scoop Box Exchange

Scoop Site Scroller: Get one yourself!
Satanosphere
/* You are not expected to understand this. */

Login
Make a new account
Username:
Password:

Hosted by ScoopHost.com Powered by Scoop
All trademarks and copyrights on this page are owned by their respective companies. Comments are owned by the Poster. The Rest 1999 The Management

create account | faq | search