On Thu, May 02, 2024 at 05:47:34AM +0000, Ryotaro Banno (伴野 良太郎) wrote: > /usr/bin/nsenter -m -u -i -n -p -t 1 /sbin/lvm lvcreate -n 7ee0d142-bffe-459f-87cb-56a355d24d2c -L 104857600b -W y -y ubuntu-vg > /usr/bin/nsenter -m -u -i -n -p -t 1 /sbin/lvm lvremove -f ubuntu-vg/7ee0d142-bffe-459f-87cb-56a355d24d2c Hi, nsenter is the correct way to run the lvm on the host, as opposed to inside the container. I'd try some experiments to see if they might narrow it down (I can't say exactly why any of these might help): - remove -Wy from the lvcreate, in case the zeroing i/o has some effect. - remove -f from the lvremove, in case there is some cleanup skipped. (Removing those could potentially have some other effects in the test, so you may need some other adjustments to the test.) > minikube ssh -- sudo truncate --size=10G backing_store > minikube ssh -- sudo losetup -f backing_store I'd try running this on a real device in case loop is has some memory impact. > My environment is as follows: I don't see an lvm version here. Dave
