01
keep the disk, drop the machine
snapshots capture the disk state of a stopped sandbox.
save the state you care about, remove the running machine, and start again later from the same bytes.
back to microsandbox
features / snapshots
save the disk state. start clean from it.
capture a stopped sandbox's disk state, archive it, and start a new sandbox from it.
02
what a snapshot is, and is not
a snapshot is the disk of a stopped sandbox. it is not a live memory image, and there is no fork, pause, or resume. what you save is the filesystem, and you restore it into a fresh microvm.
how it works
capture and restore
01
stop the sandbox
snapshots are taken from a stopped sandbox, so the disk is consistent.
02
save it
with the msb cli: snapshot create --from, then snapshot save, optionally with the image so it is self-contained.
03
restore later
run --from-snapshot boots a fresh microvm with that disk state.
see it work
create, archive, and run from a snapshot
microsandbox cli 0.6.8bash
msb stop eval-base
msb snapshot create swebench-base --from eval-base
msb snapshot save swebench-base ./swebench-base.tar.zst --with-image
msb run --from-snapshot swebench-base --name task-0441 -- pytest -q
msb rm --force task-0441common questions
what is a good use for it?
a warm base image for evals or repeated tasks: set up once, snapshot, then start every run from the same disk.