All the code related to HasCRIContainerType is useless and no longer needed
since the CRIContainerType annotation is not considered for constraining or
not the sandbox
Signed-off-by: Julio Montes <julio.montes@intel.com>
Signed-off-by: Peng Tao <bergwolf@hyper.sh>
Kata relies on the cgroup parent created and configured by the container
engine, but sometimes the sandbox cgroup is not configured and the container
may have access to all the resources, hence the runtime must constrain the
sandbox and update the list of devices with the devices hotplugged in the
hypervisor.
Fixes: kata-containers/runtime#2605
Signed-off-by: Julio Montes <julio.montes@intel.com>
Signed-off-by: Peng Tao <bergwolf@hyper.sh>
The hypervisor needs access to `/dev/vfio/vfio` to use VFIO devices.
Remove all devicemapper devices from the allowed list, the device cgroup
must be updated when before hotpluggin any device.
Signed-off-by: Julio Montes <julio.montes@intel.com>
Signed-off-by: Peng Tao <bergwolf@hyper.sh>
add `AddDevice` and `RemoveDevice` to cgroup manager to allow adding
and removing devices from the device cgroup
Signed-off-by: Julio Montes <julio.montes@intel.com>
Signed-off-by: Peng Tao <bergwolf@hyper.sh>
Add functions to convert a host device to a cgroup device or linux device,
the first one is used to update the device cgroup and the second one to
update the resources in the OCI spec.
Signed-off-by: Julio Montes <julio.montes@intel.com>
Signed-off-by: Peng Tao <bergwolf@hyper.sh>
`GetHostPath()` method returns the device path in the host, this way the
runtime can get the device information for updating the sandbox's device
cgroup.
Signed-off-by: Julio Montes <julio.montes@intel.com>
Signed-off-by: Peng Tao <bergwolf@hyper.sh>
Hardcode the Cargo.lock file to prevent dependencies
change which would cause some compatible issues.
Fixes: #230
Signed-off-by: fupan.lfp <fupan.lfp@antfin.com>
Since the ttrpc upgraded with async supported, which isn't
compatible with current agent, thus it's better to change
the dependency to a stable branch.
Fixes: #229
Signed-off-by: fupan.lfp <fupan.lfp@antfin.com>
Only load runtime config when it is not set.
We do not expect a service's runtime config to change while
it is running.
Signed-off-by: Peng Tao <bergwolf@hyper.sh>
To use the kata-containers repo path.
Most of the change is generated by script:
find . -type f -name "*.go" |xargs sed -i -e \
's|github.com/kata-containers/runtime|github.com/kata-containers/kata-containers/src/runtime|g'
Fixes: #201
Signed-off-by: Peng Tao <bergwolf@hyper.sh>
It is simply wrong to test kata-check within before sub commands
as it is NOT before at all. Besides it causes errors if kata is
not installed.
Signed-off-by: Peng Tao <bergwolf@hyper.sh>
We should not use a plain unix socket reader to act as grpc
server. Place a really mock grpc server instead.
Signed-off-by: Peng Tao <bergwolf@hyper.sh>
It's better to check whether the sandbox's get_container
result instead of unwrap it directly, otherwise it would
crash the agent if the conainer id is invalid.
Fixes: #178
Signed-off-by: fupan.lfp <fupan.lfp@antfin.com>
When kill a process, if the exec id is empty, then
it means to kill all processes in the container, if
the exec id isn't empty, then it will only kill the
specific exec process.
Signed-off-by: fupan.lfp <fupan.lfp@antfin.com>
Add the container id and exec id to start container's log
which would make it clearly to check the log.
Fixes: #173
Signed-off-by: fupan.lfp <fupan.lfp@antfin.com>
In the previous implementation, create a container process
by forking the parent process as the container process,
and then at the forked child process do much more setting,
such as rootfs mounting, drop capabilities and so on, at
last exec the container entry cmd to switch into container
process.
But since the parent is a muti thread process, which would
cause a dead lock in the forked child. For example, if one
of the parent process's thread do some malloc operation, which
would take a mutex lock, and at the same time, the parent forked
a child process, since the mutex lock status would be inherited
by the child process but there's no chance to release the lock
in the child since the child process only has a single thread
which would meet a dead lock if it would do some malloc operation.
Thus, the new implementation would do exec directly after forked
and then do the setting in the exec process. Of course, this requred
a data communication between parent and child since the child cannot
depends on the shared memory by fork way.
Fixes: #166Fixes: #133
Signed-off-by: fupan.lfp <fupan.lfp@antfin.com>
Adds a cmdline option to configure the stdout/stderr pipe sizes.
Uses `F_SETPIPE_SZ` to resize the write side of the pipe after
creation.
Example Cmdline option: `agent.container_pipe_size=2097152`
fixes#152
Signed-off-by: Alex Price <aprice@atlassian.com>
I add another sub-command `build-service` in Makefile to
generate rust-agent-related systemd service files, which
are necessary for building guest rootfs image.
The whole design is following the one in go-agent.
Fixes: #144
Signed-off-by: Penny Zheng <penny.zheng@arm.com>
It's better to catch the error of couldn't find the process
in tty_win_resize service, other wise, an invalid process id
could crash the agent.
Fixes: #137
Signed-off-by: fupan.lfp <fupan.lfp@antfin.com>
This reverts commit a3e46a369f.
There is still problem with static link, built binary will
segmentfault on clearlinux. So revert this patch for now.
Depends-on: github.com/kata-containers/tests#2293
Fixes: #69
Signed-off-by: Yang Bo <bo@hyper.sh>
Since the codes in logging.rs is weakly related to the project,
separating it from the project will reduce coupling and make it reusable.
Fixes: #131
Signed-off-by: Tim Zhang <tim@hyper.sh>
It should restore to it's previous cwd after it
create container in which it would change it's
cwd to container's bundle path.
Fixes: #126
Signed-off-by: fupan.lfp <fupan.lfp@antfin.com>
To check is the oci spec passed in, other wise,
it would crash the agent unwrap it directly.
Fixes: #124
Signed-off-by: fupan.lfp <fupan.lfp@antfin.com>
add oci compatibility test case for src/agent/oci/src/lib.rs
follow by Open Container Initiative Runtime Specification
Fixes: #118
Signed-off-by: quanweiZhou <quanweiZhou@linux.alibaba.com>
When using default cc linker, we will have segfault.
Debugging with `rust-gdb`, the specific error is as follows:
src/string/memcpy.c: No such file or directory.
Only changing linker with `aarch64-linux-musl-gcc`, the
`rust-agent` could be totally statically linked and run successfully.
Fixes: #107
Signed-off-by: Penny Zheng <penny.zheng@arm.com>
The __addtf3, __subtf3 and __multf3 symbols are used by aarch64-musl,
but are not provided by rust compiler-builtins.
For now, the only temporary but functional workaround accepted by rust
communities is to get them from libgcc.
Fixes: #107
Signed-off-by: Penny Zheng <penny.zheng@arm.com>
These are the unit tests for the sandbox struct. This is the summary
of the most important changes:
- To test containers it was needed to create a `LinuxContainer` type
and this requires root privileges. So, some tests now requires root
user to be run.
- There was a bug in the `unset_sandbox_storage` method. The return
type was wrapped in a `Result` to avoid this problem.
Fixes: #50
Signed-off-by: Erich Cordoba <erich.cordoba.malibran@intel.com>
1) pass reference instead of value when possible.
2) simplify code.
3) rename get_device_pci_address() as get_pci_device_address() to keep
consistency get_pci_device_name().
4) refine get_device_name() for maintenance.
Signed-off-by: Liu Jiang <gerry@linux.alibaba.com>
Function set_sandbox_storage() is designed to return true when the
reference count drops from 1 to 0. But current implementation always
return true no matter the reference count is, which may cause removing
an in use mountpoint.
Fixes: #88
Signed-off-by: Liu Jiang <gerry@linux.alibaba.com>
Refine uevent.rs for better maintenance:
1) use dedicated function to handle uevents.
2) use dedicated function to handle blk add events.
Signed-off-by: Liu Jiang <gerry@linux.alibaba.com>
Improve loger implementation by:
1) avoid unnecessary clone() operations.
2) change Arc<Mutex<slog::Level>> to Mutex<slog::Level>. We should use
atomic<usize> instead of Mutex<slog::Level> for better performance here.
But with slog_async::Async drainer in the pipeline, RuntimeLevelFilter
drainer will only get from a single-thread context, so keep it as is.
3) minor syntax cleanups.
Signed-off-by: Liu Jiang <gerry@linux.alibaba.com>
It's should use string.eq() to match option's
key words exactly instead of using string.starts_with()
for single key options and for "key=value" options it's
bettet to match "key=" instead of "key" with string.starts_with()
which would match wrongly such as passed options "agent.log_vsock"
which would match "agent.log" with string.starts_with()
and trigger parsing issues.
Fixes: #96
Signed-off-by: lifupan <lifupan@gmail.com>
When call "C" func directly, it's needed to change the string to
CString. To avoid using the unsafe calling, replace it with the
rust safe function to set mode for a file/dir.
Signed-off-by: lifupan <lifupan@gmail.com>
Once parsed cmdline and set the config on AGENT_CONFIG,
release the write lock as soon as possible. In case other
thread would get read lock on it.
Fixes:#87
Signed-off-by: lifupan <lifupan@gmail.com>
This patch adds a unit test for `namespace.rs`. The baseline for this
test comes from the Go agent tests in which the namespace is mounted in
a temporary folder.
In order to enable testing in the temporary folder the code was refactored
allow configuration for the specified namespace, thus the changes done
are described below:
- The `setup_persistent_ns` method was moved inside the `Namespace` type.
- A builder pattern was implemented for the `Namespace` type. This allows
the caller to set the type of the desired namespace as well as the root
folder which will help the testing.
- A new `NamespaceType` enum was introduced to represent the namespace type.
- The user of the `Namespace` type (sandbox.rs) was updated accordingly.
Fixes: #50
Signed-off-by: Erich Cordoba <erich.cordoba.malibran@intel.com>
When kata-agent run as init process in initrd, do the
init in which will do some base mount such as mount
/proc; thus the following config.parse_cmdline can access
/proc/cmdline to parse the parameters such as agent.log etc.
Fixes: #85
Signed-off-by: lifupan <lifupan@gmail.com>
The CI was failing due to two problems.
1. The `ci/static-checks.sh` was run from an incorrect place.
2. `rustfmt` was failing as some code wasn't correctly formatted.
This patch address all the changed requested by rustfmt and the
`static-checks.sh` script was updated
Fixes: #83
Signed-off-by: Erich Cordoba <erich.cordoba.malibran@intel.com>
This adds an option to the agent to control the hotplug timeout of block devices.
Retains the previous behaviour of defaulting to 3 seconds if not specified.
Can be increased when block device hot plugging is taking longer than expected.
fixes#62
Signed-off-by: Alex Price <aprice@atlassian.com>
The `skip_*` macros will be useful across the different tests so having
them in a separate module can help with code duplication. This change
creates a new module and exports the macros at crate level.
Signed-off-by: Erich Cordoba <erich.cordoba.malibran@intel.com>
Fixes to allow the rust agent to be built using a gnu target.
Specifically, remove assumptions about musl-specific types.
Fixes: #70.
Signed-off-by: James O. D. Hunt <james.o.hunt@intel.com>
Since build with --release produces corrupted binary in ci, we removed
--release. However, the make install target cannot find the binary,
set BUILDTYPE to debug
Fixes: #67
Signed-off-by: Yang Bo <bo@hyper.sh>
The get_key_value method is currently only avaiable in nightly rust.
As only this feature is required it worth to refactor and enable building
in the stable channel.
The method was removed by first getting the value from the CGROUPS hashmap,
then key is get by iterating over all the keys. The checks for an empty key and
key == "devices" were moved out of the hashmap block.
The README.md was updated as well to detail the instructions for stable rust.
Signed-off-by: Erich Cordoba <erich.cordoba.malibran@intel.com>
Create a config module and add more tests. Also enable setting the log
level from the kernel command-line.
Fixes: #59, #63.
Signed-off-by: James O. D. Hunt <james.o.hunt@intel.com>
