Commit Graph

415 Commits

Author SHA1 Message Date
Steve Horsman
5da86f3e25 Merge pull request #6440 from GeorginaKin/CCv0
CCv0: Merge main into CCv0 branch
2023-03-15 10:40:55 +00:00
Georgina Kinge
ec6ce46e7e CCv0: Merge main into CCv0 branch
Merge remote-tracking branch 'upstream/main' into CCv0

Fixes: #6428
Signed-off-by: Georgina Kinge <georgina.kinge@ibm.com>
2023-03-09 16:14:53 +00:00
yanggang
b6880c60d3 logging: Correct the code notes
Fix wrong notes for func GetSandboxesStoragePathRust()

Fixes: #6394

Signed-off-by: yanggang <gang.yang@daocloud.io>
2023-03-01 19:20:25 +08:00
Jim Cadden
4510aeaa91 runtime: Adds annotations for SEV/kbs controls at the pod level
Note: only for online-kbs configuration

Fixes #5782

Signed-off-by: Jim Cadden <jcadden@ibm.com>
2023-02-28 10:51:46 -05:00
Jim Cadden
4eb88d6a74 runtime: Add online-kbs to kernel params
In online-kbs attestation the guest is given the location of the
keybroker server to connect after launch. This patch appends the
IP:Port of the online-kbs to the kernel params of the guest.

Patch also simplifies the kbs config into "mode" = offline/online,
and updates SEV config variable names and default values

Fixes: #5661 #5715

Signed-off-by: Jim Cadden <jcadden@ibm.com>
2023-02-28 10:51:46 -05:00
Megan Wright
e85a6f77c6 CCv0: Merge main into CCv0 branch
Merge remote-tracking branch 'upstream/main' into CCv0

Fixes: #6241
Signed-off-by: Megan Wright megan.wright@ibm.com
2023-02-21 11:53:30 +00:00
zhaojizhuang
ca02c9f512 runtime: add reconnect timeout for vhost user block
Fixes: #6075
Signed-off-by: zhaojizhuang <571130360@qq.com>
2023-02-13 14:33:46 +08:00
Bin Liu
95602c8c08 Merge pull request #5999 from yaoyinnan/5998/feat/cgroup-metrics
runtime: support cgroup v2 metrics marshal guest metrics
2023-02-11 19:26:24 +08:00
Bin Liu
ecbd94d80c Merge pull request #6064 from yaoyinnan/6063/feat/rootfs-erofs
rootfs: support EROFS filesystem
2023-02-11 11:10:23 +08:00
yaoyinnan
bdf20b5d26 rootfs: support EROFS filesystem
For kata containers, rootfs is used in the read-only way.
EROFS can noticably decrease metadata overhead.

On the basis of supporting the EROFS file system, it supports using the config parameter to switch the file system used by rootfs.

Fixes: #6063

Signed-off-by: Gao Xiang <hsiangkao@linux.alibaba.com>
Signed-off-by: yaoyinnan <yaoyinnan@foxmail.com>
2023-02-11 00:44:13 +08:00
GabyCT
86501d5f6f Merge pull request #6200 from gkurz/improve-appendFDs-doc
runtime: Improve documentation of appendFDs
2023-02-09 15:50:37 -06:00
Fabiano Fidêncio
01e56a7c6d Merge pull request #5796 from niteeshkd/CCv0
CCv0: Enable 'policy' for SNP container
2023-02-09 20:24:25 +01:00
yaoyinnan
01765e1734 runtime: support cgroup v2 metrics marshal guest metrics
Support to use cgroup v2 metrics marshal guest metrics.

Fixes: #5998

Signed-off-by: yaoyinnan <yaoyinnan@foxmail.com>
2023-02-09 19:14:09 +08:00
Bin Liu
56071c6e7b virtiofsd: change cache mod to const
Change cache mod from literal to const and place them in one place.

Also set default cache mode from `none` to `never` in
`pkg/katautils/config-settings.go.in`.

Fixes: #6151

Signed-off-by: Bin Liu <bin@hyper.sh>
2023-02-08 15:06:52 +08:00
Bin Liu
71a3b73cb0 Merge pull request #6223 from d3c3mber/rm-unused-shim-config
runtime: remove not used shim configurations
2023-02-08 10:00:52 +08:00
d3c3mber
390916b33c runtime: remove not used shim configurations
ShimPath and ShimDebug are not needed anymore.

Fixes: #6147

Signed-off-by: d3c3mber <tangbo_gl_2022@163.com>
2023-02-07 14:06:12 +08:00
GabyCT
7fc35f19eb Merge pull request #6056 from jongwu/perm_deny
arm64/CI: fix unit test failure on arm64
2023-02-03 10:53:38 -06:00
Jianyong Wu
59f104c022 runtime: skip unit test that fail regularly on aarch64
There are lots of unit test cases fails regularly on aarch64, including
TestIOCopy, create_tmpfs. Temporarily skip it for now and enable it
after them get fixed.

Fixes: #6194
Signed-off-by: Jianyong Wu <jianyong.wu@arm.com>
2023-02-03 11:34:39 +08:00
Greg Kurz
3c48f2202c runtime: Improve documentation of appendFDs
The cmd.ExtraFiles feature that is used to implement appendFDs takes an
array of arbitray file descriptors and internally renumbers them to be
consecutive starting from 3, using dup2().

This isn't especially obvious : document it for the sake of clarity.

Fixes #6199

Signed-off-by: Greg Kurz <groug@kaod.org>
2023-02-02 12:52:10 +01:00
Niteesh Dubey
9ef28e3e5e CCv0: Enable 'policy' for SNP container
includes conflict resolution in qemu_amd64.go
Signed-off-by: Niteesh Dubey <niteesh@linux.ibm.com>
2023-02-02 04:21:56 +00:00
Peng Tao
a34f36f8f4 Merge pull request #6149 from openanolis/fix_kata_runtime
runtime:fix stat uds path
2023-02-02 11:00:07 +08:00
Georgina Kinge
5b4075d18a CCv0: Merge main into CCv0 branch
Merge remote-tracking branch 'upstream/main' into CCv0

Fixes: #6181
Signed-off-by: Georgina Kinge <georgina.kinge@ibm.com>
2023-02-01 10:10:40 +00:00
Greg Kurz
334c4b8bdc runtime: Drop QEMU log file support
The QEMU log file is essentially about fine grain tracing of QEMU
internals and mostly useful for developpers, not production. Notably,
the log file isn't limited in size, nor rotated in any way. It means
that a container running in the VM could possibly flood the log file
with a guest triggerable trace. For example, on openshift, the log
file is supposed to reside on a per-VM 14 GiB tmpfs mount. This means
that each pod running with the kata runtime could potentially consume
this amount of host RAM which is not acceptable.

Error messages are best collected from QEMU's stderr as kata is doing
now since PR #5736 was merged. Drop support for the QEMU log file
because it doesn't bring any value but can certainly do harm.

Fixes #6173

Signed-off-by: Greg Kurz <groug@kaod.org>
2023-01-31 09:20:29 +01:00
Zhongtao Hu
1e531b44dc runtime:fix stat uds path
os.Stat("unix:///run/vc/sbs/sid/shim-monitor.sock") will fail,
should be os.Stat("/run/vc/sbs/sid/shim-monitor.sock")

Fixes:#6148
Signed-off-by: Zhongtao Hu <zhongtaohu.tim@linux.alibaba.com>
2023-01-29 15:08:13 +08:00
zhaojizhuang
9092c23a2e runtime: Add hmp for qemu
Fixes: #6092
Signed-off-by: zhaojizhuang <571130360@qq.com>
2023-01-29 14:22:04 +08:00
Niteesh Dubey
55b36212e6 CCv0: Remove qemu paramter 'policy' for SNP container
Removed the qemu paramter 'policy' (and also dh-cert-file, session-file, kernel-hashes=on)
for SNP container.

Fixes: #5795

Signed-off-by: Niteesh Dubey <niteesh@linux.ibm.com>
2023-01-27 17:18:58 +00:00
Greg Kurz
af125b1498 Merge pull request #5736 from gkurz/no-qemu-daemonize
runtime: Start QEMU undaemonized and get logs
2023-01-27 16:33:48 +01:00
Megan Wright
36cd741943 CCv0: Merge main into CCv0 branch
Merge remote-tracking branch 'upstream/main' into CCv0

Fixes: #6126
Signed-off-by: Megan Wright megan.wright@ibm.com
2023-01-25 11:23:18 +00:00
Greg Kurz
39fe4a4b6f runtime: Collect QEMU's stderr
LaunchQemu now connects a pipe to QEMU's stderr and makes it
usable by callers through a Go io.ReadCloser object. As
explained in [0], all messages should be read from the pipe
before calling cmd.Wait : introduce a LogAndWait helper to handle
that.

Fixes #5780

Signed-off-by: Greg Kurz <groug@kaod.org>
2023-01-24 23:09:17 +01:00
Greg Kurz
bf4e3a618f runtime: Launch QEMU with cmd.Start()
LaunchCustomQemu() currently starts QEMU with cmd.Run() which is
supposed to block until the child process terminates. This assumes
that QEMU daemonizes itself, otherwise LaunchCustomQemu() would
block forever. The virtcontainers package indeed enables the
Daemonize knob in the configuration but having such an implicit
dependency on a supposedly configurable setting is ugly and fragile.

cmd.Run() is :

func (c *Cmd) Run() error {
	if err := c.Start(); err != nil {
		return err
	}
	return c.Wait()
}

Let's open-code this : govmm calls cmd.Start() and returns the
cmd to virtcontainers which calls cmd.Wait().

If QEMU doesn't start, e.g. missing binary, there won't be any
errors to collect from QEMU output. Just drop these lines in govmm.
Similarily there won't be any log file to read from in virtcontainers.
Drop that as well.

Signed-off-by: Greg Kurz <groug@kaod.org>
2023-01-24 23:09:11 +01:00
Greg Kurz
8a4f08cb0f govmm: Optionally pass QMP listener to QEMU
QEMU's -qmp option can be passed the file descriptor of a socket that
is already in listening mode. This is done with by passing `fd=XXX`
to `-qmp` instead of a path. Note that these two options are mutually
exclusive : QEMU errors out if both are passed, so we check that as
well in the validation function.

While here add the `path=` stanza in the path based case for clarity.

Signed-off-by: Greg Kurz <groug@kaod.org>
2023-01-24 23:08:48 +01:00
Greg Kurz
219bb8e7d0 govmm: Optionally start QMP with a pre-configured connection
When QEMU is launched daemonized, we have the guarantee that the
QMP socket is available. In order to launch a non-daemonized QEMU,
the QMP connection should be created before QEMU is started in order
to avoid a race. Introduce a variant of QMPStart() that can use such
an existing connection.

Signed-off-by: Greg Kurz <groug@kaod.org>
2023-01-24 19:16:47 +01:00
Peng Tao
7d1a604bad Merge pull request #6060 from ls-ggg/6055/service.mu-deadlock
runtime:all APIs are hang in the service.mu
2023-01-18 10:50:00 +08:00
Bin Liu
790f45190b Merge pull request #6074 from zhaojizhuang/enablevhostuserstore
runtime: paas enablevhostuserstore annotation to hypervisor config
2023-01-17 11:43:43 +08:00
ls
69fc8de712 runtime:all APIs are hang in the service.mu
When the vmm process exits abnormally, a goroutine sets s.monitor
to null in the 'watchSandbox' function without getting service.mu,
This will cause another goroutine to block when sending a message
to s.monitor, and it holds service.mu, which leads to a deadlock.
For example, the wait function in the file
.../pkg/containerd-shim-v2/wait.go will send a message to s.monitor
after obtaining service.mu, but s.monitor may be null at this time

Fixes: #6059

Signed-off-by: ls <335814617@qq.com>
2023-01-16 14:45:37 +08:00
Eric Ernst
458fe865ea Merge pull request #6052 from egernst/add-darwin-skeletons
Add darwin skeletons
2023-01-13 13:14:16 -08:00
zhaojizhuang
cf1bae3521 runtime: paas enablevhostuserstore annotation to hypervisor config
Fixes: #6073
Signed-off-by: zhaojizhuang <571130360@qq.com>
2023-01-13 17:07:38 +08:00
Samuel Ortiz
a9626682af virtcontainers: resourcecontrol: Add skeleton for Darwin
Cgroups do not exist on Darwin, so use an empty implementation for
resourcecontrol for the time being. In the process, ensure that the
utilized cgroup handling (ie, isSystemdCgroup) is kept in general file,
since we use this to help assess/constrain the container spec we pass to
the guest.

Fixes: #6051

Signed-off-by: Samuel Ortiz <s.ortiz@apple.com>
Signed-off-by: Eric Ernst <eric_ernst@apple.com>
2023-01-12 15:53:28 -08:00
Eric Ernst
6ee550e9a5 runtime: vCPUs pinning is sandbox specific, not hypervisor
While at it, make sure we persist this and fix a misc typo.

Signed-off-by: Eric Ernst <eric_ernst@apple.com>
2023-01-12 15:44:25 -08:00
Eric Ernst
f137048be3 resource-control: add helper function for setting CPU affinity
Let's abstract the CPU affinity

Fixes: #6044

Signed-off-by: Eric Ernst <eric_ernst@apple.com>
2023-01-11 17:55:53 -08:00
Bin Liu
86a82cace9 runtime: change cache mode from none to never
New Rust virtiofsd's `cache` mode doesn't support `none` mode,
we should use `never` to replace it.

Fixes: #6018

Signed-off-by: Bin Liu <bin@hyper.sh>
2023-01-10 17:29:48 +08:00
Georgina Kinge
e8c63ccc4d CCv0: Merge main into CCv0 branch
Merge remote-tracking branch 'upstream/main' into CCv0

Fixes: #6011
Signed-off-by: Georgina Kinge <georgina.kinge@ibm.com>
2023-01-09 12:02:48 +00:00
Bin Liu
2c10b37172 Merge pull request #5991 from dcantah/darwin-sigs
runtime: Define Darwin handled signals list
2023-01-07 11:19:48 +08:00
Fabiano Fidêncio
175794458f Merge pull request #5972 from bergwolf/github/hook
fix moby prestart hook handling
2023-01-06 14:54:39 +01:00
Samuel Ortiz
3b4420eb8e runtime: Define Darwin handled signals list
Fixes: #5990

Some signals may not be defined on non Linux host OSes, like
SIGSTKFLT for example. It's also not defined on certain architectures,
but irrelevant for this.

Signed-off-by: Samuel Ortiz <s.ortiz@apple.com>
Signed-off-by: Danny Canter <danny@dcantah.dev>
2023-01-05 17:50:47 -08:00
Danny Canter
24b05a99b6 schedcore: Make buildable on !linux
Fixes: #5983

sched-core only makes sense on Linux hosts. Let's add stub/error for
other platforms.

Signed-off-by: Eric Ernst <eric_ernst@apple.com>
Signed-off-by: Danny Canter <danny@dcantah.dev>
2023-01-05 11:51:04 -08:00
Peng Tao
d085389127 vc: fix up UT for CreateSandbox API change
Need to adapt the UT as well.

Signed-off-by: Peng Tao <bergwolf@hyper.sh>
2023-01-03 22:30:42 +08:00
Peng Tao
578a9c25f0 vc: rescan network endpoints after running prestart hooks
Moby relies on the prestart hooks to configure network endpoints. We
should rescan the netns after running them so that the newly added
endpoints can be found and plugged to the guest.

Fixes: #5941
Signed-off-by: Peng Tao <bergwolf@hyper.sh>
2023-01-03 22:30:41 +08:00
Peng Tao
cb84b0fb02 katautils: run prestart hooks after starting VM
So that we can pass the hypervisor pid to the hook instead of the
runtime process's.

Signed-off-by: Peng Tao <bergwolf@hyper.sh>
2023-01-03 10:52:32 +00:00
Zhongtao Hu
dae6670628 kata-runtime: add rust runtime path for kata-runtime exec
add rust runtime path for kata-runtime exec

Fixes:#5963
Signed-off-by: Zhongtao Hu <zhongtaohu.tim@linux.alibaba.com>
2022-12-30 13:34:34 +08:00