There is a new "ReadOnly" option added to nvdimm device in qemu
and now added to kata. However, qemu used for arm64 is a little
old and has no this feature. Here we remove this feature for arm.
Depends-on: github.com/kata-containers/tests#3831
Fixes: #2320
Signed-off-by: Jianyong Wu <jianyong.wu@arm.com>
It looks like the version check for cloud hypervisor (clh) was added
initially when clh was actively evolving its API. We no longer need the
version check as clh API has been fairly stable for its recent releases.
Fixes: #1991
Signed-off-by: Bo Chen <chen.bo@intel.com>
(cherry picked from commit d08603bebb)
When running a firecracker instance jailed, block devices
are not removed correctly, as the jailerRoot path is not
stripped from the PATCH command sent to the FC API.
This patch differentiates the jailed case from the non-jailed
one and allows the firecracker instance to be properly
terminated.
Fixes#2387
Signed-off-by: Anastassios Nanos <ananos@nubificus.co.uk>
(cherry picked from commit 64dd35ba4f)
Bring read-only nvdimm support
Shortlog:
335fa81 qemu: fix golangci-lint errors
61b6378 .github/workflows: reimplement github actions CI
9d6e797 go: support go modules
0d21263 qemu: support read-only nvdimm
ff34d28 qemu: Consistent parameter building
Signed-off-by: Julio Montes <julio.montes@intel.com>
(Based-on: 070590fb53)
Improve security by making rootfs image read-only, nobody
will be able to modify it from the guest.
fixes#1916
Signed-off-by: Julio Montes <julio.montes@intel.com>
We'd better stop the virtiofsd specifically after stop qemu,
instead of depending on the qemu's termination to notify virtiofsd
to exit.
Fixes: #2211
Signed-off-by: fupan.lfp <fupan.lfp@antgroup.com>
Since we only send an shutdown qmp command to qemu when do
stopSandbox, and didn't wait until qemu process's exit, thus
we'd better to make sure it had exited when shimv2 terminated.
Thus here to do the last cleanup of the hypervisor.
Fixes: #2198
Signed-off-by: fupan.lfp <fupan.lfp@antgroup.com>
get_mounts() parses /proc/self/mountinfo in order to get the mountpoints
for various cgroup filesystems. One of the entries in mountinfo is the
"device" for each filesystem, but for virtual filesystems like /proc, /sys
and cgroups, the device entry is arbitrary. Depending on the exact rootfs
setup, it can end up being "-".
This breaks get_mounts() because it uses " - " as a separator. There
really is a " - " separator in mountinfo, but in this case the device entry
shows up as a second one. Fix this, by changing a split to a splitn, which
will effectively only consider the first " - " in the line.
While we're there, make the warning message more useful, by having it
actually show which line it wasn't able to parse.
fixes#2182
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Use which to find the full path of exe before run execute_hook
to avoid error: 'No such file or directory'
Fixes: #2172
Signed-off-by: Tim Zhang <tim@hyper.sh>
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>
Architectures that do not support memory hotplugging will fail when
memory limits are set because that amount is hotplugged. Issue a warning
instead. The long-term solution is virtio-mem.
Fixes: #1412
Signed-off-by: Jakob Naucke <jakob.naucke@ibm.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>
covertool has no active since 2018 and is not compatible with go1.16
../vendor/github.com/dlespiau/covertool/pkg/cover/cover.go:76:29: cannot use f (type dummyTestDeps) as type testing.testDeps in argument to testing.MainStart:
dummyTestDeps does not implement testing.testDeps (missing SetPanicOnExit0 method)
Fixes: #1862
Signed-off-by: Shengjing Zhu <zhsj@debian.org>
(cherry picked from commit 1b60705646)
Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com>
When update memory limit, we should adapt the write sequence
for memory and swap memory, so it won't fail because
the new value and the old value don't fit kernel's
validation.
Fixes: #1917
Signed-off-by: fupan.lfp <fupan.lfp@antgroup.com>
(cherry picked from commit 30f4834c5b)
Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com>
The Rust standard library had suppressed the default SIGPIPE
behavior, see https://github.com/rust-lang/rust/pull/13158.
Since the parent's signal handler would be inherited by it's child
process, thus we should re-enable the standard SIGPIPE behavior as a
workaround.
Fixes: #1887
Signed-off-by: fupan.lfp <fupan.lfp@antgroup.com>
(cherry picked from commit 0ae364c8eb)
Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com>
This commit will fix two problems:
- Virtiofsd process ID returned to the caller will always be 0,
the pid var is never being assigned a value.
- Socket listen fd may leak in case of failure of starting virtiofsd process.
This is a port of be9ca0d58bFixes: #1931
Signed-off-by: bin <bin@hyper.sh>
(cherry picked from commit 773deca2f6)
Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com>
In kata-runtime check sub-command, checks cgroups and SandboxCgroupOnly
to show message if the SandboxCgroupOnly is not set to true
and cgroup v2 is used.
Fixes: #1927
Signed-off-by: bin <bin@hyper.sh>
When starting a container and dropping all capabilities,
the init child process has no permission to read the exec.fifo
file because the parent set the file mode 0o622. So change the exec.fifo file mode to 0o644.
fixes#1913
Signed-off-by: quanweiZhou <quanweiZhou@linux.alibaba.com>
(cherry picked from commit 3e4ebe10ac)
Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com>
Without this, if the shim dies, we will not have a reliable way to
identify what mounts should be cleaned up if `containerd-shim-kata-v2
cleanup` is called for the sandbox.
Before this, if you `ctr run` with a sandbox bindmount defined and SIGKILL the
containerd-shim-kata-v2, you'll notice the sandbox bindmount left on
host.
With this change, the shim is able to get the sandbox bindmount
information from disk and do the appropriate cleanup.
Fixes#1896
Signed-off-by: Eric Ernst <eric_ernst@apple.com>
(cherry picked from commit 7f1030d303)
Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com>
If for any reason there's an error when trying to setup the sandbox
bindmounts, make sure we roll back any mounts already created when
setting up the sandbox.
Without this, we'd leave shared directory mount and potentially
sandbox-bindmounts on the host.
Fixes: #1895
Signed-off-by: Eric Ernst <eric_ernst@apple.com>
(cherry picked from commit 089a7484e1)
Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com>
Since the propagation flags couldn't be combinted with the
standard mount flags, and they should be used with the remount,
thus it's better to split them from the standard mount flags.
Fixes: #1699
Signed-off-by: fupan.lfp <fupan.lfp@antgroup.com>
(cherry picked from commit e5fe572f51)
allow to set dialing timeout in configuration.toml
default is 30s
Fixes: #1789
(cherry-picked 01b56d6cbf)
Signed-off-by: Snir Sheriber <ssheribe@redhat.com>
Gathering stats for a given sandbox is pretty useful; let's export a
function from katamonitor pkg to do this.
Signed-off-by: Eric Ernst <eric_ernst@apple.com>
(cherry picked from commit 3787306107)
For easier debug, let's add subcommand to kata-runtime for gathering
metrics associated with a given sandbox.
kata-runtime metrics --sandbox-id foobar
Fixes: #1815
Signed-off-by: Eric Ernst <eric_ernst@apple.com>
(cherry picked from commit 8068a4692f)
(1) Add an accessor function, SocketAddress, to the shim-v2 code for
determining the shim's abstract domain socket address, given the sandbox
ID.
(2) In kata monitor, create a function, BuildShimClient, for obtaining the appropriate
http.Client for communicating with the shim's monitoring endpoint.
(3) Update the kata CLI and kata-monitor code to make use of these.
(4) Migrate some kata monitor methods to be functions, in order to ease
future reuse.
(5) drop unused namespace from functions where it is no longer needed.
Signed-off-by: Eric Ernst <eric_ernst@apple.com>
(cherry picked from commit 3caed6f88d)
Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com>
Instead of having something like
"/containerd-shim/$namespace/$sandboxID/shim-monitor.sock", let's change
the approach to:
* create the file in a more neutral location "/run/vc", instead of
"/containerd-shim";
* drop the namespace, as the sandboxID should be unique;
* remove ".sock" from the socket name.
This will result on a name that looks like:
"/run/vc/$sandboxID/shim-monitor"
Fixes: #497
Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com>
(cherry picked from commit 4bc006c8a4)
Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com>
There are no pod/container operations in kata-runtime binary,
tracing in this package is meaningless.
Fixes: #1748
Signed-off-by: bin <bin@hyper.sh>
(cherry picked from commit 13c23fec11)
Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com>
Some applications may fail if NOFILE limit is set to unlimited.
Although in some environments this value is explicitly overridden,
lets set it to a more sane value in case it doesn't.
Fixes#1715
Signed-off-by: Snir Sheriber <ssheribe@redhat.com>
(cherry picked from commit a188577ebf)
Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com>
Now enabling enable_pprof for individual pods is supported,
but not documented.
This commit will add per-Pod Kata configurations for `enable_pprof`
in file `docs/how-to/how-to-set-sandbox-config-kata.md`
Fixes: #1744
Signed-off-by: bin <bin@hyper.sh>
(cherry picked from commit 95e54e3f48)
Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com>
Got:
FATA[0000] run pod sandbox: rpc error: code = Unknown desc = failed to
create containerd task: Add 189759MB virtio-mem-pci fail QMP command
failed: backend memory size must be multiple of 0x200000: unknown
This commit let sizeMB be multiple of 2Mib to fix the issue.
Fixes: #1796
Signed-off-by: Hui Zhu <teawater@antfin.com>
