Use the right cow image and apply the second layer to provide read-write access. The correct setup is: - base image + base cow -> read-only snapshot (base changes "cached" until committed) - read-only snapshot + VM cow -> read-write snapshot (changes discarded after VM shutdown) This way, even VM without Qubes-specific startup scripts will can benefit from Template VMs, while VMs with Qubes-specific startup scripts may still see original root.img content (for possible signature verification, when storage domain got implemented). QubesOS/qubes-issues#2256 |
||
|---|---|---|
| .. | ||
| __init__.py | ||
| domain.py | ||
| file.py | ||
| kernels.py | ||
| lvm.py | ||
| README.md | ||
WNI File storage
Before v3.1 there existed a draft wni storage. You can find it in the git history
(it was in /core/storage directory, now gone)