Blogs

Apache Solr + Drupal

Download the apachesolr module from drupal.org, and put it in sites/all/modules:

http://drupal.org/project/apachesolr

download solr-php-client, version 22

http://code.google.com/p/solr-php-client/downloads/list

and uncompress it in the apachesolr module directory.

Ready to activate the module, but we will did not installed the Apache server, so the module will not work.

Now, as you guessed, it's time to install the apache solr server. Download one of the latests releases:

http://www.apache.org/dyn/closer.cgi/lucene/solr/

Remember to download the binary files, not the src directory. You'll know that you have the correct tar if you have an example directory with a .jar file.

 

Now, time to move all these files to a .bak version in conf directory

apache-solr-4.0.0/example/solr/collection1/conf

 

  • schema.xml,
  • solrconfig.xml,
  • protwords.xml

 

then, copy these files from your drupal module directory.

 

 

 

 

 

categorias: 

paging results in Drupal

Paginating the results in Drupal is incredible incredible easy. You just have to remember the two "magic functions":

 

  • pager_query($sql, $count);
  • theme('pager',10);

 

pager_query substitutes the typical db_query. Just something like that: 

 

  • // $resource = db_query($sql);  // -> old sentence without pager
  • $resource = pager_query($sql, $count); // -> pager

 

Then, we simple come back to the function that is called by our hook_menu, and we change:

  • return theme('list_products')

with

  • return theme('list_products') . theme('pager',10);

 

It couldn't be easier, could it? :-). In fact it is harder to undertand the documentation and the different howtos and tutorials which you can find than the process itself.

archlinux and vbox: sharing folders

This days i'm discovering a new Linux flavour, which it remembers me a lot to another old friend: Gentoo. It has a really similar way of installing, it's very very customizable like Gentoo, and it's quite minimalistic. I'm speaking about Archlinux.

I have to say that this days I was thinking in changing from more than 3 or 4 idilic years with Debian/Ubuntu to a more customizable system. I really don't like the last change which Ubuntu has made on the Desktop, and I really loved when I had the possibility of using last releases of software when using Gentoo (5 or 6 years ago).

So, the possibilities were mainly Gentoo... and my recent Archlinux discovery. Anyway, this is another story.

The fact is that I thought that working with Arhclinux was going to be more harder. Far from the reality. The installation is faster and not too much complicated, unlike Slackware, it has a powerfull package system with dependency resolution, and it is very customizable, so I can build it as fast as I want or as heavy and nice graphic looking as i prefer.

An example of how easy it is to do some things is shaing a folder with the vbox, (something that a fully advice against, unless you have no choice).

In the guest / Arhclinux side it was as simple as installing the module and activating:

  • pacman -S virtualbox-guest-modules
  • pacman -S virtualbox-guest-utils
And finally:
  • modprobe -a vboxguest vboxsf vboxvideo
  • systemctl start vboxservice.service
Enjoy :-)
categorias: 

ERROR 2013 (HY000): Lost connection to MySQL server

Have you ever had a big (and when i say big I mean BIG) database which you need to import in a sql file?

If so, you'll probably have found this issue (or even you are having the problem just now and you are reading this looking for a solution):

ERROR 2013 (HY000): Lost connection to MySQL server

This uses to happen when trying to import the dump with the mysql -u command:

mysql -u root -p database < databasetoimport.sql

You can try to play with the my.ini settings, but a best secure solution is just to do the import from the mysql own console. How? easy.

1. go to the place where your sql is. Then log in mysql:

mysql -uroot -p

2. create the database if needed:

create database mynewdb

3. Change to it:

use mynewdb;

4. and finally, import it:

source databasetoimport.sql

Wait and hopefully the problem will not appear again.

The reason is simple. Mysql is designed to have this behabiour of closing the connection dealing with external scripts, like php or the own mysql command. If you are executing the import from inside the console, the problem will not appear.

Installing capistrano in Centos

Very very easy in CentOS, just like it would be in a Debian system:

# yum install ruby -y

# yum install rubygems -y

# gem install capistrano

Really, it couldn't be easier. Next step, just having fun with Capistrano :-)

categorias: 

Upgrading two git repositories automatically

There is a simple way of upgrading or executing a command immediatly after a push has been done.

We just need to go to the server repository (the one which is going to receive the push), and enter in hooks (.git/hooks or hooks if it's a bare repository).

Then, we rename the post-receive.sample to post-receive, without extension, and add this content:

echo "********************"

echo "Post receive hook: Updating User Staging"

echo "********************"

cd ../live

GIT_DIR='.git'

git pull origin master

echo "********************"

echo "Post receive hook: Updating User Staging"

echo "********************"

You'll just have to change the ../live path with the place where your live git code lives, and the next time you make a commit, the live directory will be automatically updated with a git pull.

It is not the perfect way of going live, is my case this is a staging server, so it's fine. If you need to go to "real" live, Capistrano is what you are looking for. We'll speak about it next weeks.

categorias: 

Capistrano config.rb

set :application, "bounty app"

#set :repository,  "localhost:/var/www/git/bountyrepo"

set :repository,  "localhost:/var/www/git/blessed"


set :scm, :git # You can set :scm explicitly or Capistrano will make an intelligent guess based on known version control directory names

# Or: `accurev`, `bzr`, `cvs`, `darcs`, `git`, `mercurial`, `perforce`, `subversion` or `none`

#



default_run_options[:pty] = true

set :use_sudo, false


#role :web, "localhost"                          # Your HTTP server, Apache/etc

#role :app, "localhost"                          # This may be the same as your `Web` server

#role :db,  "your primary db-server here", :primary => true # This is where Rails migrations will run

#role :db,  "your slave db-server here"

#

#set :deploy_to, "/var/www/html/staging"



task :staging do

  role :web, "localhost"

  set :deploy_to, "/var/www/html/staging"

  #set :deploy_via, :remote_cache

  #after('deploy:symlink', 'cache:clear')

end

task :live do
  role :web, "localhost"
  set :deploy_to, "/var/www/html/live"
  #set :deploy_via, :remote_cache
  #after('deploy:symlink', 'cache:clear')
end
task :live2 do
  role :web, "localhost"
  set :deploy_to, "/var/www/html/live2"
  #set :deploy_via, :remote_cache
  #after('deploy:symlink', 'cache:clear')
end
# if you're still using the script/reaper helper you will need
# If you are using Passenger mod_rails uncomment this:
# namespace :deploy do
#   task :start do ; end
#   task :stop do ; end
#   task :restart, :roles => :app, :except => { :no_release => true } do
#     run "#{try_sudo} touch #{File.join(current_path,'tmp','restart.txt')}"
#   end
# end
#
#
namespace:deploy do
    task:start do
    end
    task:stop do
    end
    task:finalize_update do
        run "chmod -R g+w #{release_path}"
    end
    task:restart do
    end
 after "deploy:restart" do
   #add any tasks in here that you want to run after the project is deployed
     run "rm -rf #{release_path}.git"
  end
end

 

categorias: 

Developing under Linux Apache in windows

Contradictory? Yes, a bit, but you cannot allways choose the systems and architecture in which you are working.

solution? Runing apache under linux instead of runing wamp or some weird solutions like that (i don't like at all, sorry).

First, install vbox: https://www.virtualbox.org/

Second, download the iso of your preffered linux distro, and install it.

Third, when you machine is running, go to Devices>install guest additions.

Now you have to come back to your linux machine, and install a few things, kernel upgrades included (if needed).

 

  • yum install gcc -y
  • yum install kernel sources -y
  • yum install kernel-devel -y

 

 

Reboot your system, and mount:

 

  • mkdir /media/VirtualBoxGuestAdditions
  • mount -t vboxsf /dev/cdrom /media/VirtualBoxGuestAdditions

 

 

 

categorias: 

Centos tips

 

 

  • Configuring your network to start when booting: chkconfig network on (via http://nixcraft.com/centos-rhel-fedora/18643-centos-linux-6-2-eth0-netwo...)
  • configuring services to start apache (or any other daemon) when system boots: chkconfig --level 2345 httpd on
  • ... more coming
  • Override Apache permissions problem: "You don't have permission to access ... on this server". Simply execute: setenforce 0

 

categorias: 

Accessing centos apache/httpd from vbox host

This is the scenario. You installed vbox in your Mac, windows or Linux computer. Then you´ve installed Centos or Red Hat (or any other Red Hat flavor) in this virtual box.

Next step, installed httpd (apache2) and... even it is running and httpd status confirms it with a "running" message, it cannot be accesible from your host machine.

The problem is on iptables. Red Hat by default denies access to this machine from other (external) machines. Solution? Very easy, open iptables file:

vim /etc/sysconfig/iptables

(vim is my favourite editor for a loooong long time)

Look for this line:

-A INPUT -m state –state NEW -m tcp -p tcp –dport 22 -j ACCEPT

and after that, add this one:

-A INPUT -m state –state NEW -m tcp -p tcp –dport 80 -j ACCEPT

Restart your services and everything done:

service iptables restart

Thank to http://www.fliquidstudios.com/2009/06/18/creating-a-virtual-development-...

categorias: