> A Container can freeze a file system it mounted itself, since > ioctl_fsfreeze() and ioctl_fsthaw() only check CAP_SYS_ADMIN in > s_user_ns. That is a bad thing to allow: > > * Processes on top of a frozen file system hang in the kernel, so > CRIU can neither ptrace attach nor freeze them and the checkpoint > fails. > * CRIU does not dump the frozen state anyway, so after restore the > file system comes back thawed although it was frozen on the source. > > Nothing vital inside a Container is known to need it, and most file > systems a Container sees are not freezable at all. So keep the > existing check and return EPERM for FIFREEZE and FITHAW when called > from a Container. > > https://virtuozzo.atlassian.net/browse/VSTOR-155592 > > Feature: ve: allow various syscalls in CT root userns > Suggested-by: Pavel Tikhomirov <[email protected]> > Signed-off-by: Vasileios Almpanis <[email protected]>
Sashiko has reviewed this patch and found no issues. -- Sashiko AI review ยท https://sashiko.vzint.dev/#/patchset/[email protected]?part=1 _______________________________________________ Devel mailing list [email protected] https://lists.openvz.org/mailman/listinfo/devel
