Now that we have full control of Capstan let us try to improve it a little.
Here is my wish list or things I would like to do in short-term:
- Add ability to pull kernel and mpm packages from github
- Add diagnostic option to run (-d?) or enhance -v (verbose) that would:
- show full command line when invoking given hypervisor
- list networking ports exposed
- show warning when not using accelerator (KVM, etc)
- possibly add --dry-run
- Add support of running with firecracker
- Add support of running with Intel HAXM accelerator (would help on Mac
and Windows)
- Improved VirtualBox support
- For example fix this annoying bug -
https://github.com/mikelangelo-project/capstan/issues/71
- Address at least some issues from old mikelangelo capstan
- Support to build images for multiple platforms -
https://github.com/mikelangelo-project/capstan/issues/93
- use qemu-img convert?
- vt acceleration options for qemu -
https://github.com/mikelangelo-project/capstan/issues/94
- Add ability to add arbitrary directory from host
- Ad ability to add specific executable SO from host along with all
dependencies
- find dependencies using ldd like we do in python3x app
- Add ability to find collisions/conflicts - same file but from two mpms
and possibly different version etc
Also there is a little bit of dichotomy between 'capstan build' and
'capstan package build' where the 1st one uses Capstan file and other
creates image from composing it from mpms. Is there a value in keeping
both? Maybe remove Capstanfile option altogether? I do not remember ever
using it, what are the advantages? On other hand it would be nice to have
capstan package build have an option to call make or something before
composing all files into an image?
Thoughts?
Waldek
--
You received this message because you are subscribed to the Google Groups "OSv
Development" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
