Virt-Factory Roadmap
Note: this content is intended to change a lot. Ideas/comments welcome. The last released milestone was 0.0.2, which was released on 5/1/2007. We are currently working on 0.0.3.
Table of Contents
0.0.3
- Further test infrastructure upgrades
- ORM usage and move to Postgres (possibly also support MySQL?). Database init scripts also. Migration support.
- Work on new cobbler features.
- Incorporate shinier and better CSS/HTML/artwork
- Refactor vf_import to use puppet modules
- Implement basis for localization framework
0.0.4
- Use busrpc(AMQP) to communicate with nodes
- Status tracking -- heartbeat from node daemons and associated infrastructure. async notification of changes to Xen state show up in WUI, use AMQP to do this.
- Keeping VM's (re)started when host nodes restart.
- Seperate WUI from central server so they can optionally be installed on seperate machines.
- Write the command line app ("ampm")
0.0.5
- Upload and import profiles via the WUI without using vf_import
- Batch management via tasks and tag selection. (Do X to all machines tagged "foo" and "bar", look at all machines tagged "foo" and "bar", etc).
- A per host task view. Slice by hosts, see what each are doing. See what is going on in a specific host when you visit that page for the individual host. Tracking of who/what initiated tasks when. (Audit logging). Show this when clicking on the task.
- Better distro import interface (allow for running this at any time, not just startup).
- Improve registration of virtual guests so hostnames and MAC data are always up to date. Possibly register on each node daemon start?
0.0.6
- Tag management tools should be able to store the tags required to run a virtual machine, i.e. "lab4 storagepool5 fullvirtcapable" and be able to have a button to click on for a mahcine, that will allow all virtual deployments running on it to move to similarly tagged guests. These should be selected on an individual basis and can be overriden with non-similarly tagged guests as needed. Full Automation will come later.
- Knowledge of basic storage information needed for migration.
- Integration with basic monitoring information (something like SPECTRE) needed to develop automated "into the cloud" deployment heuristics.
Future / To Be Filed
- Automate cloud deployment and migration based on tags and basic resource availability.
- Guard against unauthorized registration in the case of PXE (requires Cobbler voodoo).
- Create more sample profiles -- (should move this up in priority)
- API cleanup -- create higher level API more suitable for ISVs and sysadmins who want to integrate with VF. It's a bit low-level as it stands now, but workable.
- Integrated Backup features
- Make it easy to deploy RHX-type content as virtual appliance apps