Hello,
I am running sahara service in my openstack POC infrastructure. How should I scale in and scale out my hadoop cluster size. I am using apache vannila plugin for deploying hadoop cluster. Can anyone help me to scale my cluster.
Best Regards,
Radhakrishnan Rk
The Sahara documentation has a paragraph about scaling via the GUI: https://docs.openstack.org/sahara/lat....
The CLI has a command dataprocessing cluster scalehttps://docs.openstack.org/python-sah....
Of course, you can also use the API: https://developer.openstack.org/api-r....
Disclaimer: I am by no means an expert in big data or Sahara.
Thank for the information. I have checked the doc. I am not sure will it add or remove the cluster nodes by proper service decommission. If it tries to decommission a node manager or data node of a hadoop cluster. It should backup all the data and stop the daemon gracefully. Let me check here.
Some more info here: https://docs.openstack.org/sahara/oca.... It doesn't specifically mention data backup, but "Decommissioning a Data Node may take some time because Hadoop rearranges data replicas around the Cluster" sounds good.
Asked: 2018-02-04 22:20:17 -0500
Seen: 191 times
Last updated: Feb 06 '18
