Dear all,
I have quite basic questions, but I couldn't manage to find an answer so far... Please, could you help? Just upgraded from OpenNebula 3.2 to 4.4, Ubuntu 12.04 based controllers and hosts, KVM-based. I create my images, for the future templates, using Virtual Machine Manager, based on Ubuntu official iso files. In OpenNebula 4.4 docs, there are three guidelines on setting up contextualization: basic, advanced and cloud-init usage. 1. But... can one combine those strategies? For example, can I enable basic contextualization in order to benefit from network parameters and then use advanced contextualization (init script context file) in order to setup, for example, a NFS mount point? 2. In order to benefit from contextualization variables, is it a pre-requisite that one installs one-context package to use each of the contextualization strategies or only for a specific scenario (basic, advanced or cloud-init)? 3. Suppose I want to use only the information on basic contextualizion (DEV_IP, DEV_DNS, etc...) to setup, for instance, virtual machine instance IP, gateway, DNS... What should be the correct steps? Should I... a. Install the one-context package before anything? b. Forget about one-context package and only enable network contextualization in the template and that´s done; or use both the one-context package and enable the network contextualization in the template? c. Do the above and also use a init script context file, using the DEV_IP, DEV_DNS, etc... variables? 4. In this sense, are the one-context scripts (00-network, vmcontext, etc...) used only with advanced contextualization, or also with the basic one? 5. At last, about the swap partition... I have an image with a swap partition defined, tried to use it in a template, but I am facing several boot hangups associated to swap file creation (it did not happen in OpenNebula 3.2). I read in the docs ( http://docs.opennebula.org/stable/user/virtual_machine_setup/bcont.html) that "...The contextualization package will also mount any partition labeled swap as swap..." . Does that mean we should at first create an image without swap and then setup in OpenNebula template configuration, which will use this image, a volatile disk, swap type? Is that the correct/suggested way to work? I really appreciate both your help and OpenNebula, I feel just a bit confused on those contextualization strategies and swap issues... Best regards, Nelson Peixoto Kotowski Filho
_______________________________________________ Users mailing list [email protected] http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
