Using Fedora/RHEL as Hadoop Node Template in Serenget 1.0 GA

[Jesse Hu]

Hi Users,

Fedora as Serengeti Node Template is not officially supported by Serengeti yet. However you can follow the steps below to use Fedora 18：
1) Download and deploy VMware vSphere Big Data Extensions 1.0 GA (i.e. Enterprise Serengeti) from http://www.vmware.com/go/download-bigdataextensions
2) Login Serengeti Server as user serengeti, download attached fedora.diff, then execute: cd /opt/serengeti/cookbooks/ ; patch -p1 < fedora.diff ; knife cookbook upload hadoop_common hive -V
3) Follow guide Install CentOS 6.x and VMware Tools on the Hadoop Template Virtual Machine
to create your Fedora 18 (or other version) VM Template.  There might be an simple error when executing "./installer.sh /usr/java/default", you can fix it by yourself or post it here for help.
4) use the standard CLI command to create cluster.

BTW, although RHEL is also not officially supported by Serengeti, you can use RHEL 5.x/6.x as the Serengeti Node Template by following the same steps above. Have a try if you need Fedora or RHEL :)

Thanks
Jesse Hu