gvisor/runsc
Nicolas Lacasse 1775a0e11e container.Destroy should clean up container metadata even if other cleanups fail
If the sandbox process is dead (because of a panic or some other problem),
container.Destroy will never remove the container metadata file, since it will
always fail when calling container.stop().

This CL changes container.Destroy() to always perform the three necessary
cleanup operations:
* Stop the sandbox and gofer processes.
* Remove the container fs on the host.
* Delete the container metadata directory.

Errors from these three operations will be concatenated and returned from
Destroy().

PiperOrigin-RevId: 225448164
Change-Id: I99c6311b2e4fe5f6e2ca991424edf1ebeae9df32
2018-12-13 15:38:10 -08:00
..
boot Add "trace signal" option 2018-12-11 16:12:41 -08:00
cgroup Order feature strings by block 2018-10-26 12:18:36 -07:00
cmd A sandbox process should wait until it has not been moved into cgroups 2018-12-06 15:28:29 -08:00
console Track paths and provide a rename hook. 2018-10-23 00:20:15 -07:00
container container.Destroy should clean up container metadata even if other cleanups fail 2018-12-13 15:38:10 -08:00
fsgofer Use RET_KILL_PROCESS if available in kernel 2018-11-20 22:56:51 -08:00
sandbox A sandbox process should wait until it has not been moved into cgroups 2018-12-06 15:28:29 -08:00
specutils Open source system call tests. 2018-12-10 14:42:34 -08:00
test Open source system call tests. 2018-12-10 14:42:34 -08:00
tools/dockercfg Internal change. 2018-11-28 14:01:48 -08:00
BUILD runsc: Pass log and config files to sandbox process by FD. 2018-09-04 20:10:01 -07:00
main.go Add "trace signal" option 2018-12-11 16:12:41 -08:00