Dontyoudarenottobeaware

After I have changed my debian server from rvm to use rbenv I gave it a shot for my local system: These commands should work: git clone https://github.com/sstephenson/rbenv.git ~/.rbenv echo 'export PATH="$HOME/.rbenv/bin:$PATH"' >> ~/.bashrc echo 'eval "$(rbenv init -)"' >> ~/.bashrc " type rbenv" should show something like "rbenv is a function" git clone https://github.com/sstephenson/ruby-build.git ~/.rbenv/plugins/ruby-build git clone https://github.com/sstephenson/rbenv-gem-rehash.git ~/.rbenv/plugins/rbenv-gem-rehash rbenv install 2.1.1 -k rbenv global 2.1.1 gem install rails   But unfortunately first ruby and then rails were not working as good on manjaro/arch as on debian. So I had to find solutions: CONFIGURE_OPTS="--with-readline-dir=/usr/include/readline" rbenv install 2.1.1 OR (could not get console working, tried following) curl -fsSL https://gist.github.com/mislav/a18b9d7f0d...

Diese Zusammenfassung ist nicht verfügbar. Klicke hier, um den Post aufzurufen.

After I fixed one problem with my zurb foundation gem (downgrading to 5.1.1.0 worked) I ran into other problems when deploying into production environment onto my debian server. Deploying with capistrano or mina is both depending on a working ruby system with all the gems and of course bundler working in perfect harmony. Unfortunately my rvm per user installation did not work that good with my capistrano version 3. And after finding several forks on github I decided to look for alternatives, which meant other users were not happy with the original. I gave mina a try and found it different and interesting enough to use it. And mina works great except bundler was not "showing up to the party". My reason to switch back though was the ability of capistrano to roll back releases. In the end everything pointed to one black sheep: rvm. Right now I am installing rbenv which sounds promising because focussing on little things like it was intend by the UNIX founders with al...

After trying the "Redirect permanent / https://....." for serveral hours and reading almost all of the internet I tried my other browser midori and my configuration just works fine. So, my conclusion: firefox browser sucks!!!!! Keeping this in mind (and I emptied my cache etc) I will not rely on firefox anymore when working with my servers apache configuration. Well, if you want to use more than one virtual host with ONE wildcard ssl certificate then you have to redirect via vhost files: <VirtualHost *:80> ServerName sub.domain.com Redirect permanent / https://...... ..... </VirtualHost> <VirtualHost *:443>  SSLEngine On .... ServerName sub.domain.com .... </VirtualHost>

Its a long time ago I had enough time to set up my mail server, so I will summarize my steps, or better: my problems: [on debian 7] install postfix install postfixadmin install postgresql make them work together as they all should use virtual users install dovecot and make it the default SSL handler for postfix, main.cf master.cf etc. Well, this is bull because its of no help to anyone, the devil resides in the details: postfixadmin sets up the database postfix should use queries for to look for mails in the database, so virtual_mailboxes etc have to be set up dovecot should be the MDA for the MTA postfix dovecot and postfixadmin should use a simple and same encryption, because both look into the database; is this step is done (and it took me a while, CRAM-MD5?) you can set up mailboxes via the web interface postfixadmin

This could be a huge post, but I am cutting it short: configure pg_hba.conf right, e.g. set host stuff to password md5 try to connect to your database locally to test it, e.g. psql -U user "psql -U user dbname -W" and type in the password install phppgadmin, e.g. on debian I could not find a /etc/phppgadmin/apache.conf but a /etc/apache2/conf.d/phppgadmin your apache2 should be able to serve SSL, otherwise look a my post from yesterday: SSL virtualhost after you setup a SSH connection through rsa keys and always redirect to it in order to serve multiple site using SSL you should change to *:443 instead of _default_ also you have to tell /etc/apache2/ports.conf to use NameVirtualhost when listening on 443; otherwise apache will tell you about conflicting virtualhosts. apache2: now its time to copy the content of conf.d/phppgadmin into a sites-available file and dont forget to delete the alias line and adjust names and folders. From now on I followed these steps: Step...

After migrating my server from one account to another, basically setting it up again, I ran into some problems with my etherpad lite installation. Since my first post on this I wanted to add some more users for more control whose group is accessing what pad. But there was always something wrong! My conclusions: use "service etherpad-lite restart" instead of the /admin restart button make sure that apache is running or restart it often enough change the bloody JSON file trough the /admin interface since its telling you about corrupt syntax Last point on the list was my major problem for I forgot to put down some more blanks in order to have the JSON hierarchy. After that was sorted my basic authentication ALMOST worked: dont use "requiresession = true" for it messes up some stuff somehow Now my installation works like a charm with my postgresql database.

wagn is something cool, and this is how I installed it on debian 6.0.7: already had my server setup with apache2, passenger, rvm and postgresql ( howto ) I also installed imagemagick (in case rmagick would work with it) got to your webroot sudo git clone https://github.com/wagn/wagn.git now you have a wagn folder change permissions to have write access, e.g.: sudo chown -R foobar:www-data wagn cd wagn (or whatever foldername you want) I tried, but no, rmagick is not working, so: outcomment rmagick in Gemfile bundle install --without mysql:memcache:test:debug:development:assets bundle exec rake wagn:install change your database.yml to fit your postgresql settings and of course have a working database already set up; wagn does dropping, creating, loading schema so your database user should be granted createdb privileges bundle exec rake wagn:create and you should see the db setting up create a /etc/apache2/sites-available/wagn.domain.com ...with e.g. a virtual host: ...

My centOS vserver broke down after a "yum update" and even the centOS community forum could not give an answer within 24h. In my books that is poor for the almighty and stable centOS. Bare in mind that I am just a normal user of linux, no professional and definitely no expert when it comes to server, but I can read, try and share my findings: debian 6 installation steps via ssh secure your server: adduser foobar for security purposes  change /etc/sshd/sshd_conf to PermitRootLogin no now you can login as foobar and change to root via su - visudo for sudo command permissions and put this at the bottom: foobar   ALL=(ALL) ALL change hostname change hostname in /etc/hostname and in /etc/hosts reboot login via rsa key rather than password locally do ssh-keygen ssh-copy-id -i foobar@hostname... ssh now works without passwords but with keys (for easy deployment) install ruby and rails via rvm login as root always good to have: sudo apt-get install...