I based my instructions on previous post on Howto Get Started With Amazon EC2 API Tools, so I won’t give details on some steps. And this post will cover mainly the steps taken to complete my objectives above.

To start, I signed up for an account and enabled EC2 and S3 services, and generate X509 certificate. Next, I selected a test server running CentOS 5.3 with Cpanel and installed java (openjdk 1.6, using yum) as a requirement.

Then, download EC2 API Tools and extract to my working directory at /myhome/.ec2/ and upload your private key and x509 certificates. Don’t forget to follow the filename format of cert-xxx.pem and pk-xx.pem.

Export shell variables (posted on the previous post) and specify the correct private and x509 path. Then run source /myhome/.bashrc or open new terminal to load new environment variables.

Setup EC2 keypair. At first i used the certificate from different account but i got the error below:
Client.AuthFailure: AWS was not able to validate the provided access credentials
I searched for this error and one suggestion is to chmod your certificate and key files to 600 but it didn’t help me. My problem is on our account because one of my teammates changed our account password and probably generated new keys. Anyway, this is where i signed up for a new account and proceeded without issues.

Search for the AMIs to use. Following the steps listed on the instructions, I tried several AMI’s (start/stop processes). I observed some AMI’s took longer to start compare to others but i have no idea why . Btw, you can also search for AMI’s, and start/stop them from Amazon Management Console (EC2 Dashboard).

My next task is to create my private AMIs and here’s a good video tutorial on Customizing an existing AMI and create your own AMI from it. From this part that I need to setup my S3 bucket or directory to store my AMI. There’s a Firefox addon called S3Fox that my friend suggested but unfortunately i can’t install it on my Firefox due to some errors. I found and tried this BucketExplorer for creating my S3 Bucket. Btw, this one is commercial and you can try it for 30 days. I haven’t checked for other apps.

Back to creating my private AMI based on the above video, I ran into issue with ec2-bundle-vol command as it is not included on the AMI that i used, so i search for other AMIs that includes the EC2 Tools and found one from RightScale (CentOS5V1_10.img.manifest.xml).

After this i was able to complete my private AMI and start new instance from it using the above steps without any issues.

Moving Files:

• Copy Drupal file and preserve mode (ownerships, permissions, etc)
  Example: cp -rp drupal_source drupal_destination
  Review your directory permissions on sites/default/files, sites/default/files/civicrm, and other directories.
• Update references to Drupal url, path, and database details (name, user, pass, and host). Sample commands below using grep:
  find /path/to/drupal -type f -exec perl -pi -e "s/example.com/example2.com/g" {} \;
find /path/to/drupal -type f -exec perl -pi -e "s/public_html\/example/public_html\/example2/g" {} \;
find /path/to/drupal -type f -exec perl -pi -e "s/db_name/db_name2/g" {} \;
find /path/to/drupal -type f -exec perl -pi -e "s/db_user/db_user2/g" {} \;
find /path/to/drupal -type f -exec perl -pi -e "s/db_pass/db_pass2/g" {} \;

Moving Database/s:

Case 1: Combined CiviCRM and Drupal Database.

• Create sql dump of source database.
  Example: mysqldump -Q -udb_user -pdbpass db_name > db_name.sql
• Import to destination database.
  Example: mysql -udb_user2 -pdbpass2 db_name2 < db_name.sql
• Update references to Drupal url, path, and database details (name, user, pass, and host) of non-CiviCRM tables. You can use PhpMyAdmin to export this tables, then do the search/replace process on your local editor, and upload back the updates sql. You can also dump the tables from using command line (but you’ll have a long list of tables) and do the grep (same as above) and re-import the updated sql file.
• Update CiviCRM configurations from Drupal Admin section. You need to update the “Resource URLs” and “Directories”.
  CiviCRM Admin Settings: Administer Civicrm > Global Settings > Directories (or use the direct url: /civicrm/admin/setting/path?reset=1)
  CiviCRM Admin Settings: Administer Civicrm > Global Settings > Resource Urls (or use the direct url: /civicrm/admin/setting/url?reset=1)
• Optional: You can empty Sessions and Cache tables if you want.

Case 2: Separate CiviCRM and Drupal Database (recommended install for CiviCRM).
Process for this setup is almost the same as Case 1, the difference is on the import process for databases. I’ll just provide the complete info below.

• Create sql dump of source databases.
  Examples:
  mysqldump -Q -udb_user -pdbpass db_name_drupal > db_name_drupal.sql
mysqldump -Q -udb_user -pdbpass db_name_civicrm > db_name_civicrm.sql

• Import directly the CiviCRM database.
  Example: mysql -udb_user2 -pdbpass2 db_name2_civicrm < db_name_civicrm.sql
• Update CiviCRM configurations from Drupal Admin section. You need to update the “Resource URLs” and “Directories”.
  CiviCRM Admin Settings: Administer Civicrm > Global Settings > Directories (or use the direct url: /civicrm/admin/setting/path?reset=1)
  CiviCRM Admin Settings: Administer Civicrm > Global Settings > Resource Urls (or use the direct url: /civicrm/admin/setting/url?reset=1)
• Update references to Drupal url, path, and database details (name, user, pass, and host) of Drupal database dump.
  perl -pi -e "s/example.com/example2.com/g" db_name_drupal.sql
perl -pi -e "s/public_html\/example/public_html\/example2/g" db_name_drupal.sql
• Import the Drupal database.
  Example: mysql -udb_user2 -pdbpass2 db_name2_drupal < db_name_drupal.sql
• Optional: You can empty Sessions and Cache tables if you want.

That’s All!