Move the networking details out of the architecture doc and into a
separate file.
Partially fixes: #3246.
Signed-off-by: James O. D. Hunt <james.o.hunt@intel.com>
Move the background and example command details out of the architecture
doc and into separate files.
Partially fixes: #3246.
Signed-off-by: James O. D. Hunt <james.o.hunt@intel.com>
Move the historical details out of the architecture doc
and into a separate file.
Partially fixes: #3246.
Signed-off-by: James O. D. Hunt <james.o.hunt@intel.com>
Move the architecture document into a new `docs/design/architecture/` directory
in preparation for splitting it into more manageable pieces.
Signed-off-by: James O. D. Hunt <james.o.hunt@intel.com>
Allow using `expect()` for `Mutex.lock()` because it is almost
unrecoverable if failed in the lock acquisition
Fixes: #3285
Signed-off-by: Zack <zmlcc@linux.alibaba.com>
In `utils/kata-manager.sh`, we download the first asset listed for the
release, which used to be the static x86_64 tarball. If that happened to
not match the system architecture, we would abort. Besides that logic
being invalid for !x86_64 (despite not distributing other tarballs at
the moment), the first asset listed is also not the static tarball any
more, it is the vendored source tarball. Retrieve all _static_ tarballs
and select the appropriate one depending on architecture.
Fixes: #3254
Signed-off-by: Jakob Naucke <jakob.naucke@ibm.com>
Refresh the content and formatting of the architecture document.
Out of scope of these changes:
- Diagram updates.
- Updates to the Networking section.
Fixes: #3190.
Signed-off-by: James O. D. Hunt <james.o.hunt@intel.com>
not clear why this was commented out before -- ensure that we set
approprate annotation on the sandbox container's annotations to indicate
this is a sandbox.
Signed-off-by: Eric Ernst <eric_ernst@apple.com>
Today we assume that if the CRI/upper layer doesn't provide a container
type annotation, it should be treated as a sandbox. Up to this point, a
sandbox with a pause container in CRI context and a single container
(ala ctr run) are treated the same.
For VM sizing and container constraining, it'll be useful to know if
this is a sandbox or if this is a single container.
In updating this, we cleanup the type handling tests and we update the
containerd annotations vendoring.
Fixes: #2926
Signed-off-by: Eric Ernst <eric_ernst@apple.com>
Change io/ioutil to io/os packages because io/ioutil package
is deprecated from 1.16:
TempDir => os.MkdirTemp
Details: https://go.dev/doc/go1.16#ioutilFixes: #3265
Signed-off-by: bin <bin@hyper.sh>
since qemu 6.0, readonly flag for block devices must be enable or
disable with `on` or `off` respectively.
Signed-off-by: Julio Montes <julio.montes@intel.com>
According to https://endoflife.date/go golang 1.11.10 is not supported
anymore, 1.16.10 is the minimum supported version.
Fixes: #3265
Signed-off-by: bin <bin@hyper.sh>
This is a bug release from Cloud Hypervisor addressing the following
issues: 1) Networking performance regression with virtio-net; 2) Limit
file descriptors sent in vfio-user support; 3) Fully advertise PCI MMIO
config regions in ACPI tables; 4) Set the TSS and KVM identity maps so
they don't overlap with firmware RAM; 5) Correctly update the DeviceTree
on restore.
Details can be found: https://github.com/cloud-hypervisor/cloud-hypervisor/releases/tag/v20.1Fixes: #3262
Signed-off-by: Bo Chen <chen.bo@intel.com>
As containerd can properly run without having a existent
`/etc/containerd/config.toml` file (it'd run using the default
cobnfiguration), let's explicitly create the file in those cases.
This will avoid issues on ammending runtime classes to a non-existent
file.
Fixes: #3229
Signed-off-by: Fabiano Fidêncio <fabiano.fidencio@intel.com>
Tested-by: Jakob Naucke <jakob.naucke@ibm.com>
Now if no options/arguments specified, the shell scripts will return an error:
ERROR: Invalid rootfs directory: ''
This commit will show usage if no options/arguments specified.
Fixes: #3256
Signed-off-by: bin <bin@hyper.sh>
Use the same runtime used for podman run also for the podman build cmd
Additionally remove "docker" from the docker_run_args variable
Fixes: #3239
Signed-off-by: Snir Sheriber <ssheribe@redhat.com>
Unit-Test-Advice.md was moved to kata-containers repo but URLs pointing
to that document were not updated. This patch updates these URLs.
Depends-on: github.com/kata-containers/tests#4273
fixes#3240
Signed-off-by: Julio Montes <julio.montes@intel.com>
- Upgrade Alpine guest rootfs to 3.15
- Specify a minor version rather than patch level as the Alpine
repositories use that.
Signed-off-by: Jakob Naucke <jakob.naucke@ibm.com>
#2399 partially reverted #418, missing on returning to bootstrapping a
rootfs with `apk.static` instead of copying the entire root, which can
result in drastically larger (more than 10x) images. Revert this as well
(requires some updates to URL building).
Fixes: #3216
Signed-off-by: Jakob Naucke <jakob.naucke@ibm.com>
The help information of '-f' option is missing, and same issue
with 'BLOCK_SIZE' env variables, fix it in usage() function.
Fixes: #3231
Signed-off-by: zhanghj <zhanghj.lc@inspur.com>
Some new attributes are added to hypervisor config:
- VMStorePath
- RunStorePath
- SharedPath
These attributes should be handled in two places:
- reset when check the new hypervisor's config is suitable
to the base config.
- copy from new hypervisor's config when create new VM
Fixes: #3193
Signed-off-by: bin <bin@hyper.sh>
ppc64le & s390x have no (well supported) musl target for Rust,
therefore, the agent must use glibc and cannot use Alpine. Specify
Ubuntu as the distribution to be used for initrd.
Fixes: #3212
Signed-off-by: Jakob Naucke <jakob.naucke@ibm.com>
