To workaround virtiofs' lack of inotify support, we'll special case
particular mounts which are typically watched, and pass on information
to the agent so it can ensure that the mount presented to the container
is indeed watchable (see applicable agent commit).
This commit will:
- identify watchable mounts based on file count and mount source
- create a watchable-bind storage object for these mounts to
communicate intent to the agent
- update the OCI spec to take the updated watchable mount source into account
Unit tests added and updated for the newly introduced
functionality/functions.
Signed-off-by: Eric Ernst <eric_ernst@apple.com>
Add support for watchable-bind storage driver. When watchable-bind storage
is present, the agent will create a watchable path in a tmpfs, and poll the
watchable-bind source to keep this new mount-point up to date.
This poll will allow the agent to present the mount-point to the
container, allowing for inotify usage by the container workload.
If a mount becomes too large, either in file count or in overall size,
we want to stop treating it as watchable, and instead just treat as a
bindmount. This'll help avoid DoS by growing tmpfs too large, as well
as limiting time spent scanning files. If a watchable-bind grows beyond
8 files (arbitrary sane number for certs/secrets) or 1MB (limit on ConfigMap size),
we treat it as a normal bind.
Fixes: #1879
Signed-off-by: Eric Ernst <eric_ernst@apple.com>
Signed-off-by: Maksym Pavlenko <pavlenko.maksym@gmail.com>
Signed-off-by: Samuel Ortiz <samuel.e.ortiz@protonmail.com>
agent: watcher: SandboxStorages check loop cleanup
There's no reason to pass the paths; they can be
determined when they are actually used.
Let's make the return values more comparable to the other mount handling
functions (we'll add storage object in future commit), and pass the mount maps as
function parameters.
...No functional changes here...
Signed-off-by: Eric Ernst <eric_ernst@apple.com>
After create an container/exec successfully, containerd
would wait it immediately, and if start it failed, there
is no chance to send value to exitCh, thus the wait goroutine
would blocked for ever and had no chance to exit.
Fixes: #2087
Signed-off-by: fupan.lfp <fupan.lfp@antgroup.com>
Currently the virtio-mem device is hotplugged on the root bus.
This doesn't work for PCIe machines like q35.
Hotplug the virtio-mem device into the pci bridge instead.
Fixes#1953
Signed-off-by: Marcel Apfelbaum <marcel@redhat.com>
Keeping around two different x86 machines has no added value
and require more tests and maintenance. Prefer the q35 machine
since it has more features and drop the pc machine.
Fixes#1953
Depends-on: github.com/kata-containers/tests#3586
Signed-off-by: Marcel Apfelbaum <marcel@redhat.com>
To print the correct value of kernel parameters, the log field
value should not be a function name. And for that qemuArchBase
doesn't contain debug flag, so the log contains debug/non-debug
parameters.
Fixes: #2048
Signed-off-by: bin <bin@hyper.sh>
`memory_offset` is used to increase the maximum amount of memory
supported in a VM, this offset is equal to the NVDIMM/PMEM device that
is hot added, in real use case workloads such devices are bigger than
4G, which is the current limit (uint32).
fixes#2006
Signed-off-by: Julio Montes <julio.montes@intel.com>
It's hard to visually scan over the list currently.
Therefore, we should sort the list alphabetically to scan easily.
Fixes: #1999
Signed-off-by: Manabu Sugimoto <Manabu.Sugimoto@sony.com>
Since SEV support has been added, an implementation mistake was also
added to TestQemuAmd64AppendProtectionDevice.
appendProtectionDevice() will, as it name says, append the protection
device to whatever was there previously. So, when SEV was added, we
broke the comparison done for TDX as we didn't append the expected
output for TDX with what we already had for SEV.
This should be enough to get the tests passing.
Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com>
This PR removes old links that were used in kata 1.x but not
longer valid for kata 2.x
Fixes#2019
Signed-off-by: Gabriela Cervantes <gabriela.cervantes.tellez@intel.com>
The call to Trace() in runHooks() does not return a context; fix this so
that the subsequent calls to runHook() produces a properly ordered trace
span.
Fixes#2001
Signed-off-by: Chelsea Mafrica <chelsea.e.mafrica@intel.com>
Add the following mount options to catch up with the runtime spec
- silent
- loud
- (no)acl
- (no)iversion
- (no)lazytime
Fixes: #1999
Signed-off-by: Manabu Sugimoto <Manabu.Sugimoto@sony.com>
We need to fix some agent's code to conform to the latest nix crate
to be able to use new features of the nix.
Fixes: #1987
Signed-off-by: Manabu Sugimoto <Manabu.Sugimoto@sony.com>
Remove storeSandbox() at the end of createSandboxFromConfig(),
because this callchain createSandboxFromConfig -> createContainers
has already calls storeSandbox().
This can improve the startup speed of the container,
even just for a little.
Fixes: #1980
Signed-off-by: Liang Zhou <zhoul110@chinatelecom.cn>
Secure Execution is a confidential computing technology on s390x (IBM Z
& LinuxONE). Enable the correspondent virtualization technology in QEMU
(where it is referred to as "Protected Virtualization").
- Introduce enableProtection and appendProtectionDevice functions for
QEMU s390x.
- Introduce CheckCmdline to check for "prot_virt=1" being present on the
kernel command line.
- Introduce CPUFacilities and avilableGuestProtection for hypervisor
s390x to check for CPU support.
Fixes: #1771
Signed-off-by: Jakob Naucke <jakob.naucke@ibm.com>
Previously, all consts were in single lines in
virtcontainers/qemu_s390x.go. Put them into a const block.
Signed-off-by: Jakob Naucke <jakob.naucke@ibm.com>
