docker might bind mount some files/dirs under container rootfs
without notifying runtime. We need to unmount them otherwise
docker will fail to clean up containers.
man umount(2):
MNT_DETACH (since Linux 2.4.11)
Perform a lazy unmount: make the mount point unavailable for new accesses, immediately
disconnect the filesystem and all filesystems mounted below it from each other and
from the mount table, and actually perform the unmount when the mount point ceases to be busy.
Signed-off-by: Peng Tao <bergwolf@gmail.com>
The agent code creates a directory at
`/run/kata-containers/shared/sandboxes/sbid/` to hold shared data
between host and guest. We need to clean it up when removing a sandbox.
Fixes: #1138
Signed-off-by: Peng Tao <bergwolf@gmail.com>
VM templates creates a symlink from `/run/vc/vm/sbid` to
`/run/vc/vm/vmid`. We need to clean up both of them.
Signed-off-by: Peng Tao <bergwolf@gmail.com>
Because the runtime v2 runs as a RPC server, the caller will at some
point use the Shutdown() API to shut down the server. Because this
will cause the server to exit, the caller cannot expect any valid
answer when calling this. That's why we cannot afford stopping and
deleting the sandbox from this function.
Instead, we move sandbox.Stop() and sandbox.Delete() to a more
appropriate API, the Delete() one.
Fixes#1150
Signed-off-by: Sebastien Boeuf <sebastien.boeuf@intel.com>
Remove `nvdimm` from qemu command line
as the upstream qemu on ppc64le does not have
nvdimm capabilities yet.
Fixes: #1136
Signed-off-by: Nitesh Konkar niteshkonkar@in.ibm.com
Before, we would only allow for a container-update command
to proceed if the container was in the running state. So
long as the container is created, this should be allowed.
This was found using the `static` policy for Kubernetes CPU
manager[1]. Where the `update` command is called after the
`create` runtime command (when the container state is `ready`).
[1] https://github.com/kubernetes/community/blob/95a4a1/contributors/design-proposals/node/cpu-manager.md#example-scenarios-and-interactionsFixes: #1083
Signed-off-by: Jose Carlos Venegas Munoz <jose.carlos.venegas.munoz@intel.com>
Bridge is representing a PCI/E bridge, so we're moving the bridge*.go
to types/pci*.go.
Fixes: #1119
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
In order to move the hypervisor implementations into their own package,
we need to put the capabilities type into the types package.
Fixes: #1119
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
In order to move the hypervisor implementations into their own package,
we need to put the asset type into the types package and break the
hypervisor->asset->virtcontainers->hypervisor cyclic dependency.
Fixes: #1119
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
Support building with `make HYPERVISOR=firecracker`.
Notes:
- Firecracker is currently only supported on x86_64, hence the
architecture check.
- The template file `cli/config/configuration-fc.toml.in` is based
on `cli/config/configuration.toml.in`.
Fixes#1105.
Signed-off-by: James O. D. Hunt <james.o.hunt@intel.com>
Improve the build to:
- Encapsulate the qemu configuration.
- Validate the chosen hypervisor.
- Display hypervisor details for `make help`.
This lays the groundwork to build configurations for multiple
hypervisors.
Notes:
- Variables that are hypervisor specific have now been suffixed with `_QEMU`.
- `make install` now installs `configuration-qemu.toml`.
To maintain compatibility with existing installations, a symbolic link
to this file called `configuration.toml` is created.
Signed-off-by: James O. D. Hunt <james.o.hunt@intel.com>
Move the quiet prefix from the `INSTALL_EXEC` recipe and use it at the
call sites. This makes the code robust when calling the recipe in a
`$(foreach ...)` loop.
Signed-off-by: James O. D. Hunt <james.o.hunt@intel.com>
The error message does not provide the max memory that is exceeded.
Fix it for better error information.
Fixes: #1120
Signed-off-by: Jose Carlos Venegas Munoz <jose.carlos.venegas.munoz@intel.com>
Function SetKernelParams is just to update the runtimeConfig according to itself.
It just around the configuration.
So this patch moves it to updateRuntimeConfig.
Fixes: #1106
Signed-off-by: Hui Zhu <teawater@hyper.sh>
Add firecracker version to versions.yaml. This allows kata
packaging to build and package firecracker.
Fixes: #1103
Signed-off-by: Manohar Castelino <manohar.r.castelino@intel.com>
The contents of /proc/cpuinfo were
trimmed and hence the "model" field could
not be parsed despite being a field in
/proc/cpuinfo. Fix this issue.
Fixes: #1089
Signed-off-by: Nitesh Konkar niteshkonkar@in.ibm.com
This reverts commit 802bfa26c9.
Seems that we have some performance issues when
using this new kernel. Reverting to have a clean CI and
have a proper investigation about the degradation.
Fixes: #1100.
Signed-off-by: Salvador Fuentes salvador.fuentes@intel.com
When overlay is used as storage driver, kata runtime creates a new bind mount
point to the merged directory, that way this directory can be shared with the
VM through 9p. By default the mount propagation is shared, that means mount
events are propagated, but umount events not, to deal with this problem and to
avoid left mount points in the host once container finishes, the mount
propagation of bind mounts should be set to private.
Depends-on: github.com/kata-containers/tests#971
fixes#794
Signed-off-by: Julio Montes <julio.montes@intel.com>
And add some additional log output for displaying the directories and
files created when kicking the virtcontainers tests.
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
We always call waitSandbox after we start the VM (startSandbox), so
let's simplify the hypervisor interface and integrate waiting for the VM
into startSandbox.
This makes startSandbox a blocking call, but that is practically the
case today.
Fixes: #1009
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
We always combine the hypervisor init and createSandbox, because what
we're trying to do is simply that: Set the hypervisor and have it create
a sandbox.
Instead of keeping a method with vague semantics, remove init and
integrate the actual hypervisor setup phase into the createSandbox one.
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
We can now remove all the sandbox shared types and convert the rest of
the code to using the new internal types package.
This commit includes virtcontainers, cli and containerd-shim changes in
one atomic change in order to not break bisect'ibility.
Fixes: #1095
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
