This is the follow-up PR of #1279 to fix remaining issues:
Use init() to avoid race condition in IsIntelRdtEnabled().
Add also rename some variables and functions.
Signed-off-by: Xiaochen Shen <xiaochen.shen@intel.com>
This applies cgroups earlier for container creation before the init
process starts running and forking off any additional processes.
Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
With the help of userfaultfd CRIU supports lazy migration. Lazy
migration means that memory pages are only transferred from the
migration source to the migration destination on page fault.
This enables to reduce the downtime during process or container
migration to a minimum as the memory does not need to be transferred
during migration.
Lazy migration currently depends on userfaultfd being available on the
current Linux kernel and if the used CRIU version supports lazy
migration. Both dependencies can be checked by querying CRIU via RPC if
the lazy migration feature is available. Using feature checking instead
of version comparison enables runC to use CRIU features from the
criu-dev branch. This way the user can decide if lazy migration should
be available by choosing the right kernel and CRIU branch.
To use lazy migration the CRIU process during dump needs to dump
everything besides the memory pages and then it opens a network port
waiting for remote page fault requests:
# runc checkpoint httpd --lazy-pages --page-server 0.0.0.0:27 \
--status-fd /tmp/postcopy-pipe
In this example CRIU will hang/wait once it has opened the network port
and wait for network connection. As runC waits for CRIU to finish it
will also hang until the lazy migration has finished. To know when the
restore on the destination side can start the '--status-fd' parameter is
used:
#️ runc checkpoint --help | grep status
--status-fd value criu writes \0 to this FD once lazy-pages is ready
The parameter '--status-fd' is directly from CRIU and this way the
process outside of runC which controls the migration knows exactly when
to transfer the checkpoint (without memory pages) to the destination and
that the restore can be started.
On the destination side it is necessary to start CRIU in 'lazy-pages'
mode like this:
# criu lazy-pages --page-server --address 192.168.122.3 --port 27 \
-D checkpoint
and tell runC to do a lazy restore:
# runc restore -d --image-path checkpoint --work-path checkpoint \
--lazy-pages httpd
If both processes on the restore side have the same working directory
'criu lazy-pages' creates a unix domain socket where it waits for
requests from the actual restore. runC starts CRIU restore in lazy
restore mode and talks to 'criu lazy-pages' that it wants to restore
memory pages on demand. CRIU continues to restore the process and once
the process is running and accesses the first non-existing memory page
the 'criu lazy-pages' server will request the page from the source
system. Thus all pages from the source system will be transferred to the
destination system. Once all pages have been transferred runC on the
source system will end and the container will have finished migration.
This can also be combined with CRIU's pre-copy support. The combination
of pre-copy and post-copy (lazy migration) provides the possibility to
migrate containers with minimal downtimes.
Some additional background about post-copy migration can be found in
these articles:
https://lisas.de/~adrian/?p=1253https://lisas.de/~adrian/?p=1183
Signed-off-by: Adrian Reber <areber@redhat.com>
Before adding the actual lazy migration support, this adds the feature
check for lazy-pages. Right now lazy migration, which is based on
userfaultd is only available in the criu-dev branch and not yet in a
release. As the check does not dependent on a certain version but on
a CRIU feature which can be queried it can be part of runC without a new
version check depending on a feature from criu-dev.
Signed-off-by: Adrian Reber <areber@redhat.com>
About Intel RDT/CAT feature:
Intel platforms with new Xeon CPU support Intel Resource Director Technology
(RDT). Cache Allocation Technology (CAT) is a sub-feature of RDT, which
currently supports L3 cache resource allocation.
This feature provides a way for the software to restrict cache allocation to a
defined 'subset' of L3 cache which may be overlapping with other 'subsets'.
The different subsets are identified by class of service (CLOS) and each CLOS
has a capacity bitmask (CBM).
For more information about Intel RDT/CAT can be found in the section 17.17
of Intel Software Developer Manual.
About Intel RDT/CAT kernel interface:
In Linux 4.10 kernel or newer, the interface is defined and exposed via
"resource control" filesystem, which is a "cgroup-like" interface.
Comparing with cgroups, it has similar process management lifecycle and
interfaces in a container. But unlike cgroups' hierarchy, it has single level
filesystem layout.
Intel RDT "resource control" filesystem hierarchy:
mount -t resctrl resctrl /sys/fs/resctrl
tree /sys/fs/resctrl
/sys/fs/resctrl/
|-- info
| |-- L3
| |-- cbm_mask
| |-- min_cbm_bits
| |-- num_closids
|-- cpus
|-- schemata
|-- tasks
|-- <container_id>
|-- cpus
|-- schemata
|-- tasks
For runc, we can make use of `tasks` and `schemata` configuration for L3 cache
resource constraints.
The file `tasks` has a list of tasks that belongs to this group (e.g.,
<container_id>" group). Tasks can be added to a group by writing the task ID
to the "tasks" file (which will automatically remove them from the previous
group to which they belonged). New tasks created by fork(2) and clone(2) are
added to the same group as their parent. If a pid is not in any sub group, it
Is in root group.
The file `schemata` has allocation bitmasks/values for L3 cache on each socket,
which contains L3 cache id and capacity bitmask (CBM).
Format: "L3:<cache_id0>=<cbm0>;<cache_id1>=<cbm1>;..."
For example, on a two-socket machine, L3's schema line could be `L3:0=ff;1=c0`
which means L3 cache id 0's CBM is 0xff, and L3 cache id 1's CBM is 0xc0.
The valid L3 cache CBM is a *contiguous bits set* and number of bits that can
be set is less than the max bit. The max bits in the CBM is varied among
supported Intel Xeon platforms. In Intel RDT "resource control" filesystem
layout, the CBM in a group should be a subset of the CBM in root. Kernel will
check if it is valid when writing. e.g., 0xfffff in root indicates the max bits
of CBM is 20 bits, which mapping to entire L3 cache capacity. Some valid CBM
values to set in a group: 0xf, 0xf0, 0x3ff, 0x1f00 and etc.
For more information about Intel RDT/CAT kernel interface:
https://www.kernel.org/doc/Documentation/x86/intel_rdt_ui.txt
An example for runc:
Consider a two-socket machine with two L3 caches where the default CBM is
0xfffff and the max CBM length is 20 bits. With this configuration, tasks
inside the container only have access to the "upper" 80% of L3 cache id 0 and
the "lower" 50% L3 cache id 1:
"linux": {
"intelRdt": {
"l3CacheSchema": "L3:0=ffff0;1=3ff"
}
}
Signed-off-by: Xiaochen Shen <xiaochen.shen@intel.com>
This mirrors the standard_init_linux.go seccomp code, which only applies
seccomp early if NoNewPrivileges is enabled. Otherwise it's done
immediately before execve to reduce the amount of syscalls necessary for
users to enable in their seccomp profiles.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
Fixes: #1557
I'm not quite sure about the root cause, looks like
systemd still want them to be uint64.
Signed-off-by: Qiang Huang <h.huangqiang@huawei.com>
While we have significant protections in place against CVE-2016-9962, we
still were holding onto a file descriptor that referenced the host
filesystem. This meant that in certain scenarios it was still possible
for a semi-privileged container to gain access to the host filesystem
(if they had CAP_SYS_PTRACE).
Instead, open the FIFO itself using a O_PATH. This allows us to
reference the FIFO directly without providing the ability for
directory-level access. When opening the FIFO inside the init process,
open it through procfs to re-open the actual FIFO (this is currently the
only supported way to open such a file descriptor).
Signed-off-by: Aleksa Sarai <asarai@suse.de>
The documentation here:
https://docs.docker.com/engine/security/userns-remap/#user-namespace-known-limitations
says that readonly containers can't be used with user namespaces do to some
kernel restriction. In fact, there is a special case in the kernel to be
able to do stuff like this, so let's use it.
This takes us from:
ubuntu@docker:~$ docker run -it --read-only ubuntu
docker: Error response from daemon: oci runtime error: container_linux.go:262: starting container process caused "process_linux.go:339: container init caused \"rootfs_linux.go:125: remounting \\\"/dev\\\" as readonly caused \\\"operation not permitted\\\"\"".
to:
ubuntu@docker:~$ docker-runc --version
runc version 1.0.0-rc4+dev
commit: ae2948042b08ad3d6d13cd09f40a50ffff4fc688-dirty
spec: 1.0.0
ubuntu@docker:~$ docker run -it --read-only ubuntu
root@181e2acb909a:/# touch foo
touch: cannot touch 'foo': Read-only file system
Signed-off-by: Tycho Andersen <tycho@docker.com>
congfig.Sysctl setting is duplicated.
when contianer is rootless and Linux is nil, runc will panic.
Signed-off-by: Ma Shimiao <mashimiao.fnst@cn.fujitsu.com>
state.json should be a reflection of the container's
realtime state, including resource configurations,
so we should update state.json after updating container
resources.
Signed-off-by: Qiang Huang <h.huangqiang@huawei.com>
Fixes: #1228
It can be reproduced by applying this patch:
```diff
@@ -45,6 +46,7 @@ func registerMemoryEvent(cgDir string, evName string, arg string) (<-chan struct
go func() {
defer func() {
close(ch)
+ <-time.After(1 * time.Second)
eventfd.Close()
evFile.Close()
}()
```
We can close channel after fds were closed.
Signed-off-by: Qiang Huang <h.huangqiang@huawei.com>
Linux is not always not nil.
If Linux is nil, panic will occur.
Signed-off-by: Ma Shimiao <mashimiao.fnst@cn.fujitsu.com>
Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
Before this change, some file type would be treated as char devices
(e.g. symlinks).
Signed-off-by: Kenfe-Mickael Laventure <mickael.laventure@gmail.com>
This allows the libcontainer to automatically clean up runc:[1:CHILD]
processes created as part of nsenter.
Signed-off-by: Alex Fang <littlelightlittlefire@gmail.com>
With this runC also uses RPC to ask CRIU for its version. CRIU supports
a VERSION RPC since CRIU 3.0 and using the RPC interface does not
require parsing the console output of CRIU (which could change anytime).
For older CRIU versions which do not yet have the VERSION RPC runC falls
back to its old CRIU output parsing mode.
Once CRIU 3.0 is the minimum version required for runC the old code can
be removed.
v2:
* adapt to changes in the previous patches based on the review
Signed-off-by: Adrian Reber <areber@redhat.com>
Update criurpc.proto for the upcoming VERSION RPC.
This includes lazy_pages for the upcoming lazy migration support.
Signed-off-by: Adrian Reber <areber@redhat.com>
To use the CRIU VERSION RPC the criuSwrk function is adapted to work
with CriuOpts set to 'nil' as CriuOpts is not required for the VERSION
RPC.
Also do not print c.criuVersion if it is '0' as the first RPC call will
always be the VERSION call and only after that the version will be
known.
Signed-off-by: Adrian Reber <areber@redhat.com>
If the version of criu has already been determined there is no need to
ask criu for the version again. Use the value from c.criuVersion.
v2:
* reduce unnecessary code movement in the patch series
* factor out the criu version parsing into a separate function
Signed-off-by: Adrian Reber <areber@redhat.com>
The checkCriuVersion function used a string to specify the minimum
version required. This is more comfortable for an external interface
but for an internal function this added unnecessary complexity. This
changes to version string like '1.5.2' to an integer like 10502. This is
already the format used internally in the function.
Signed-off-by: Adrian Reber <areber@redhat.com>
go's switch statement doesn't need an explicit break. Remove it where
that is the case and add a comment to indicate the purpose where the
removal would lead to an empty case.
Found with honnef.co/go/tools/cmd/staticcheck
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Handle err return value of fmt.Scanf, os.Pipe and unix.ParseUnixRights.
Found with honnef.co/go/tools/cmd/staticcheck
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
This moves all console code to use github.com/containerd/console library to
handle console I/O. Also move to use EpollConsole by default when user requests
a terminal so we can still cope when the other side temporarily goes away.
Signed-off-by: Daniel Dao <dqminh89@gmail.com>
Refactor DeviceFromPath in order to get rid of package syscall and
directly use the functions from x/sys/unix. This also allows to get rid
of the conversion from the OS-independent file mode values (from the os
package) to Linux specific values and instead let's us use the raw
file mode value directly.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Updated logrus to use v1 which includes a breaking name change Sirupsen -> sirupsen.
This includes a manual edit of the docker term package to also correct the name there too.
Signed-off-by: Steven Hartland <steven.hartland@multiplay.co.uk>
Use ParseSocketControlMessage and ParseUnixRights from
golang.org/x/sys/unix instead of their syscall equivalent.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
fix#1476
If containerA shares namespace, say ipc namespace, with containerB, then
its ipc namespace path would be the same as containerB and be stored in
`state.json`. Exec into containerA will just read the namespace paths
stored in this file and join these namespaces. So, if containerB has
already been stopped, `docker exec containerA` will fail.
To address this issue, we should always save own namespace paths no
matter if we share namespaces with other containers.
Signed-off-by: Yuanhong Peng <pengyuanhong@huawei.com>
It appears as though these semantics were not fully thought out when
implementing them for rootless containers. It is not necessary (and
could be potentially dangerous) to set the owner of /run/ctr/$id to be
the root inside the container (if user namespaces are being used).
Instead, just use the e{g,u}id of runc to determine the owner.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
Use IoctlGetInt and IoctlGetTermios/IoctlSetTermios instead of manually
reimplementing them.
Because of unlockpt, the ioctl wrapper is still needed as it needs to
pass a pointer to a value, which is not supported by any ioctl function
in x/sys/unix yet.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Use unix.Prctl() instead of manually reimplementing it using
unix.RawSyscall. Also use unix.SECCOMP_MODE_FILTER instead of locally
defining it.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Using MS_PRIVATE meant that there was a race between the mount(2) and
the umount2(2) calls where runc inadvertently has a live reference to a
mountpoint that existed on the host (which the host cannot kill
implicitly through an unmount and peer sharing).
In particular, this means that if we have a devicemapper mountpoint and
the host is trying to delete the underlying device, the delete will fail
because it is "in use" during the race. While the race is _very_ small
(and libdm actually retries to avoid these sorts of cases) this appears
to manifest in various cases.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
replace #1492#1494fix#1422
Since https://github.com/opencontainers/runtime-spec/pull/876 the memory
specifications are now `int64`, as that better matches the visible interface where
`-1` is a valid value. Otherwise finding the correct value was difficult as it
was kernel dependent.
Signed-off-by: Justin Cormack <justin.cormack@docker.com>
Use unix.Eventfd() instead of calling manually reimplementing it using
the raw syscall. Also use the correct corresponding unix.EFD_CLOEXEC
flag instead of unix.FD_CLOEXEC (which can have a different value on
some architectures and thus might lead to unexpected behavior).
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
And Stat_t.PID and Stat_t.Name while we're at it. Then use the new
.State property in runType to distinguish between running and
zombie/dead processes, since kill(2) does not [1]. With this change
we no longer claim Running status for zombie/dead processes.
I've also removed the kill(2) call from runType. It was originally
added in 13841ef3 (new-api: return the Running state only if the init
process is alive, 2014-12-23), but we've been accessing
/proc/[pid]/stat since 14e95b2a (Make state detection precise,
2016-07-05, #930), and with the /stat access the kill(2) check is
redundant.
I also don't see much point to the previously-separate
doesInitProcessExist, so I've inlined that logic in runType.
It would be nice to distinguish between "/proc/[pid]/stat doesn't
exist" and errors parsing its contents, but I've skipped that for the
moment.
The Running -> Stopped change in checkpoint_test.go is because the
post-checkpoint process is a zombie, and with this commit zombie
processes are Stopped (and no longer Running).
[1]: https://github.com/opencontainers/runc/pull/1483#issuecomment-307527789
Signed-off-by: W. Trevor King <wking@tremily.us>
And convert the various start-time properties from strings to uint64s.
This removes all internal consumers of the deprecated
GetProcessStartTime function.
Signed-off-by: W. Trevor King <wking@tremily.us>
Use KeyctlJoinSessionKeyring, KeyctlString and KeyctlSetperm from
golang.org/x/sys/unix instead of manually reimplementing them.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
And use it only in local tooling that is forwarding the pseudoterminal
master. That way runC no longer has an opinion on the onlcr setting
for folks who are creating a terminal and detaching. They'll use
--console-socket and can setup the pseudoterminal however they like
without runC having an opinion. With this commit, the only cases
where runC still has applies SaneTerminal is when *it* is the process
consuming the master descriptor.
Signed-off-by: W. Trevor King <wking@tremily.us>
Use the NLA_ALIGNTO and NLA_HDRLEN constants from x/sys/unix instead of
syscall, as the syscall package shouldn't be used anymore (except for a
few exceptions).
This also makes the syscall_NLA_HDRLEN workaround for gccgo unnecessary.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Use the symlink xattr syscall wrappers Lgetxattr, Llistxattr and
Lsetxattr from x/sys/unix (introduced in
golang/sys@b90f89a1e7) instead of
providing own wrappers. Leave the functionality of system.Lgetxattr
intact with respect to the retry with a larger buffer, but switch it to
use unix.Lgetxattr.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Follow commit 3d7cb4293c ("Move libcontainer to x/sys/unix") and also
move the examples in README.md from syscall to x/sys/unix.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Since syscall is outdated and broken for some architectures,
use x/sys/unix instead.
There are still some dependencies on the syscall package that will
remain in syscall for the forseeable future:
Errno
Signal
SysProcAttr
Additionally:
- os still uses syscall, so it needs to be kept for anything
returning *os.ProcessState, such as process.Wait.
Signed-off-by: Christy Perez <christy@linux.vnet.ibm.com>
* User Case:
User could use prestart hook to add block devices to container. so the
hook should have a way to set the permissions of the devices.
Just move cgroup config operation before prestart hook will work.
Signed-off-by: Wentao Zhang <zhangwentao234@huawei.com>
FreeBSD does not support cgroups or namespaces, which the code suggested, and is not supported
in runc anyway right now. So clean up the file naming to use `_linux` where appropriate.
Signed-off-by: Justin Cormack <justin.cormack@docker.com>
A freezer cgroup allows to dump processes faster.
If a user wants to checkpoint a container and its storage,
he has to pause a container, but in this case we need to pass
a path to its freezer cgroup to "criu dump".
Signed-off-by: Andrei Vagin <avagin@virtuozzo.com>
When C/R was implemented, it was enough to call manager.Set to apply
limits and to move a task. Now .Set() and .Apply() have to be called
separately.
Fixes: 8a740d5391 ("libcontainer: cgroups: don't Set in Apply")
Signed-off-by: Andrei Vagin <avagin@virtuozzo.com>
Instead of relying on version numbers it is possible to check if CRIU
actually supports certain features. This introduces an initial
implementation to check if CRIU and the underlying kernel actually
support dirty memory tracking for memory pre-dumping.
Upstream CRIU also supports the lazy-page migration feature check and
additional feature checks can be included in CRIU to reduce the version
number parsing. There are also certain CRIU features which depend on one
side on the CRIU version but also require certain kernel versions to
actually work. CRIU knows if it can do certain things on the kernel it
is running on and using the feature check RPC interface makes it easier
for runc to decide if the criu+kernel combination will support that
feature.
Feature checking was introduced with CRIU 1.8. Running with older CRIU
versions will ignore the feature check functionality and behave just
like it used to.
v2:
- Do not use reflection to compare requested and responded
features. Checking which feature is available is now hardcoded
and needs to be adapted for every new feature check. The code
is now much more readable and simpler.
v3:
- Move the variable criuFeat out of the linuxContainer struct,
as it is not container specific. Now it is a global variable.
Signed-off-by: Adrian Reber <areber@redhat.com>
The original implementation is in C, which increases cognitive load and
possibly might cause us problems in the future. Since sys/unix is better
maintained than the syscall standard library switching makes more sense.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
Since this is a runC-specific feature, this belongs here over in
opencontainers/ocitools (which is for generic OCI runtimes).
In addition, we don't create a new network namespace. This is because
currently if you want to set up a veth bridge you need CAP_NET_ADMIN in
both network namespaces' pinned user namespace to create the necessary
interfaces in each network namespace.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
If the stdio of the container is owned by a group which is not mapped in
the user namespace, attempting to fchown the file descriptor will result
in EINVAL. Counteract this by simply not doing an fchown if the group
owner of the file descriptor has no host mapping according to the
configured GIDMappings.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
Previously Host{U,G}ID only gave you the root mapping, which isn't very
useful if you are trying to do other things with the IDMaps.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
The rootless cgroup manager acts as a noop for all set and apply
operations. It is just used for rootless setups. Currently this is far
too simple (we need to add opportunistic cgroup management), but is good
enough as a first-pass at a noop cgroup manager.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
This enables the support for the rootless container mode. There are many
restrictions on what rootless containers can do, so many different runC
commands have been disabled:
* runc checkpoint
* runc events
* runc pause
* runc ps
* runc restore
* runc resume
* runc update
The following commands work:
* runc create
* runc delete
* runc exec
* runc kill
* runc list
* runc run
* runc spec
* runc state
In addition, any specification options that imply joining cgroups have
also been disabled. This is due to support for unprivileged subtree
management not being available from Linux upstream.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
Effectively, !dumpable makes implementing rootless containers quite
hard, due to a bunch of different operations on /proc/self no longer
being possible without reordering everything.
!dumpable only really makes sense when you are switching between
different security contexts, which is only the case when we are joining
namespaces. Unfortunately this means that !dumpable will still have
issues in this instance, and it should only be necessary to set
!dumpable if we are not joining USER namespaces (new kernels have
protections that make !dumpable no longer necessary). But that's a topic
for another time.
This also includes code to unset and then re-set dumpable when doing the
USER namespace mappings. This should also be safe because in principle
processes in a container can't see us until after we fork into the PID
namespace (which happens after the user mapping).
In rootless containers, it is not possible to set a non-dumpable
process's /proc/self/oom_score_adj (it's owned by root and thus not
writeable). Thus, it needs to be set inside nsexec before we set
ourselves as non-dumpable.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
If we try to pause a container on the system without freezer cgroups,
we can found that runc tries to open ./freezer.state. It is obviously wrong.
$ ./runc pause test
no such directory for freezer.state
$ echo FROZEN > freezer.state
$ ./runc pause test
container not running or created: paused
Signed-off-by: Andrei Vagin <avagin@virtuozzo.com>
When process config doesnt specify capabilities anywhere, we should not panic
because setting capabilities are optional.
Signed-off-by: Daniel Dao <dqminh89@gmail.com>
This reverts commit d4091ef151.
d4091ef151 ("fix minor issue") doesn't actually make any sense, and
actually makes the code more confusing.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
This maybe a nice extra but it adds complication to the usecase. The
contract is listen on the socket and you get an fd to the pty master and
that is that.
Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
Runc needs to copy certain files from the top of the cgroup cpuset hierarchy
into the container's cpuset cgroup directory. Currently, runc determines
which directory is the top of the hierarchy by using the parent dir of
the first entry in /proc/self/mountinfo of type cgroup.
This creates problems when cgroup subsystems are mounted arbitrarily in
different dirs on the host.
Now, we use the most deeply nested mountpoint that contains the
container's cpuset cgroup directory.
Signed-off-by: Konstantinos Karampogias <konstantinos.karampogias@swisscom.com>
Signed-off-by: Will Martin <wmartin@pivotal.io>
In container process's Init function, we use
fd + execFifoFilename to open exec fifo, so this
field in init config is never used.
Signed-off-by: Qiang Huang <h.huangqiang@huawei.com>
Fixes: #1347Fixes: #1083
The root cause of #1083 is because we're joining an
existed cgroup whose kmem accouting is not initialized,
and it has child cgroup or tasks in it.
Fix it by checking if the cgroup is first time created,
and we should enable kmem accouting if the cgroup is
craeted by libcontainer with or without kmem limit
configed. Otherwise we'll get issue like #1347
Signed-off-by: Qiang Huang <h.huangqiang@huawei.com>
It should not be binded to container creation, for
example, runc restore needs to create a
libcontainer.Container, but it won't need exec fifo.
So create exec fifo when container is started or run,
where we really need it.
Signed-off-by: Qiang Huang <h.huangqiang@huawei.com>
The error message added here provides no value as the caller already
knows all the added details. However it is covering up the underyling
system error (typically `ENOTSUP`). There is no way to handle this error before
this change.
Signed-off-by: Brian Goff <cpuguy83@gmail.com>
getDevices() has been updated to skip `/dev/.lxc` and `/dev/.lxd-mounts`, which was breaking privileged Docker containers running on runC, inside of LXD managed Linux Containers
Signed-off-by: Carlton-Semple <carlton.semple@ibm.com>
CRIU gets pre-dump to complete iterative migration.
pre-dump saves process memory info only. And it need parent-path
to specify the former memory files.
This patch add pre-dump and parent-path arguments to runc checkpoint
Signed-off-by: Deng Guangxing <dengguangxing@huawei.com>
Signed-off-by: Adrian Reber <areber@redhat.com>
As the runtime-spec allows it, we want to be able to specify overlayfs
mounts with:
{
"destination": "/etc/pki",
"type": "overlay",
"source": "overlay",
"options": [
"lowerdir=/etc/pki:/home/amurdaca/go/src/github.com/opencontainers/runc/rootfs_fedora/etc/pki"
]
},
This patch takes care of allowing overlayfs mounts. Both RO and RW
should be supported.
Signed-off-by: Antonio Murdaca <runcom@redhat.com>
`label.InitLabels` takes options as a string slice in the form of:
user:system_u
role:system_r
type:container_t
level:s0:c4,c5
However, `DupSecOpt` and `DisableSecOpt` were still adding a docker
specifc `label=` in front of every option. That leads to `InitLabels`
not being able to correctly init selinux labels in this scenario for
instance:
label.InitLabels(DupSecOpt([%OPTIONS%]))
if `%OPTIONS` has options prefixed with `label=`, that's going to fail.
Fix this by removing that docker specific `label=` prefix.
Signed-off-by: Antonio Murdaca <runcom@redhat.com>
Correct container.Destroy() docs to clarify that destroy can only operate on containers in specific states.
Signed-off-by: Steven Hartland <steven.hartland@multiplay.co.uk>
If a relative pathed exe is used for InitArgs init will fail to run if Cwd is not set the original path.
Prevent failure of init to run by ensuring that exe in InitArgs is an absolute path.
Signed-off-by: Steven Hartland <steven.hartland@multiplay.co.uk>
If we pass a file descriptor to the host filesystem while joining a
container, there is a race condition where a process inside the
container can ptrace(2) the joining process and stop it from closing its
file descriptor to the stateDirFd. Then the process can access the
*host* filesystem from that file descriptor. This was fixed in part by
5d93fed3d2 ("Set init processes as non-dumpable"), but that fix is
more of a hail-mary than an actual fix for the underlying issue.
To fix this, don't open or pass the stateDirFd to the init process
unless we're creating a new container. A proper fix for this would be to
remove the need for even passing around directory file descriptors
(which are quite dangerous in the context of mount namespaces).
There is still an issue with containers that have CAP_SYS_PTRACE and are
using the setns(2)-style of joining a container namespace. Currently I'm
not really sure how to fix it without rampant layer violation.
Fixes: CVE-2016-9962
Fixes: 5d93fed3d2 ("Set init processes as non-dumpable")
Signed-off-by: Aleksa Sarai <asarai@suse.de>
When signaling children and the signal is SIGKILL wait for children
otherwise conditionally wait for children which are ready to report.
This reaps all children which exited due to the signal sent without
blocking indefinitely.
Also:
* Ignore ignore ECHILD, which means the child has already gone.
Signed-off-by: Steven Hartland <steven.hartland@multiplay.co.uk>
Ensure that the pipe is always closed during the error processing of StartInitialization.
Also:
* Fix a comment typo.
* Use newContainerInit directly as there's no need for i to be an initer.
* Move the comment about the behaviour of Init() directly above it, clarifying what happens for all defers.
Signed-off-by: Steven Hartland <steven.hartland@multiplay.co.uk>
Add the import of nsenter to the example in libcontainer's README.md, as without it none of the example code works.
Signed-off-by: Steven Hartland <steven.hartland@multiplay.co.uk>
POSIX mandates that `cmsg_len` in `struct cmsghdr` is a `socklen_t`,
which is an `unsigned int`. Musl libc as used in Alpine implements
this; Glibc ignores the spec and makes it a `size_t` ie `unsigned long`.
To avoid the `-Wformat=` warning from the `%lu` on Alpine, cast this
to an `unsigned long` always.
Signed-off-by: Justin Cormack <justin.cormack@docker.com>
Add godoc links to README.md files for runc and libcontainer so its easy to access the golang documentation.
Signed-off-by: Steven Hartland <steven.hartland@multiplay.co.uk>
signalAllProcesses was making the assumption that the requested signal was SIGKILL, possibly due to the signal parameter being added at a later date, and hence it was safe to wait for all processes which is not the case.
BaseContainer.Signal(s os.Signal, all bool) exposes this functionality to consumers, so an arbitrary signal could be used which is not guaranteed to make the processes exit.
Correct the documentation for signalAllProcesses around the signal delivered and update it so that the wait is only performed on SIGKILL hence making it safe to process other signals without risk of blocking forever, while still maintaining compatibility to SIGKILL callers.
Signed-off-by: Steven Hartland <steven.hartland@multiplay.co.uk>
The parameters passed to `GetExecUser` is not correct.
Consider the following code:
```
package main
import (
"fmt"
"io"
"os"
)
func main() {
passwd, err := os.Open("/etc/passwd1")
if err != nil {
passwd = nil
} else {
defer passwd.Close()
}
err = GetUserPasswd(passwd)
if err != nil {
fmt.Printf("%#v\n", err)
}
}
func GetUserPasswd(r io.Reader) error {
if r == nil {
return fmt.Errorf("nil source for passwd-formatted
data")
} else {
fmt.Printf("r = %#v\n", r)
}
return nil
}
```
If the file `/etc/passwd1` is not exist, we expect to return
`nil source for passwd-formatted data` error, and in fact, the func
`GetUserPasswd` return nil.
The same logic exists in runc code. this patch fix it.
Signed-off-by: Wang Long <long.wanglong@huawei.com>
The `bufio.Scanner.Scan` method returns false either by reaching the
end of the input or an error. After Scan returns false, the Err method
will return any error that occurred during scanning, except that if it
was io.EOF, Err will return nil.
We should check the error when Scan return false(out of the for loop).
Signed-off-by: Wang Long <long.wanglong@huawei.com>
This sets the init processes that join and setup the container's
namespaces as non-dumpable before they setns to the container's pid (or
any other ) namespace.
This settings is automatically reset to the default after the Exec in
the container so that it does not change functionality for the
applications that are running inside, just our init processes.
This prevents parent processes, the pid 1 of the container, to ptrace
the init process before it drops caps and other sets LSMs.
This patch also ensures that the stateDirFD being used is still closed
prior to exec, even though it is set as O_CLOEXEC, because of the order
in the kernel.
https://github.com/torvalds/linux/blob/v4.9/fs/exec.c#L1290-L1318
The order during the exec syscall is that the process is set back to
dumpable before O_CLOEXEC are processed.
Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
On some systems, when we mount some cgroup subsystems into
a same mountpoint, the name sequence of mount options and
cgroup directory name can not be the same.
For example, the mount option is cpuacct,cpu, but
mountpoint name is /sys/fs/cgroup/cpu,cpuacct. In current
runc, we set mount destination name from combining
subsystems, which comes from mount option from
/proc/self/mountinfo, so in my case the name would be
/sys/fs/cgroup/cpuacct,cpu, which is differernt from
host, and will break some applications.
Fix it by using directory name from host mountpoint.
Signed-off-by: Qiang Huang <h.huangqiang@huawei.com>
`HookState` struct should follow definition of `State` in runtime-spec:
* modify json name of `version` to `ociVersion`.
* Remove redundant `Rootfs` field as rootfs can be retrived from
`bundlePath/config.json`.
Signed-off-by: Zhang Wei <zhangwei555@huawei.com>
A remount of a mount point must include all the current flags or
these will be cleared:
```
The mountflags and data arguments should match the values used in the
original mount() call, except for those parameters that are being
deliberately changed.
```
The current code does not do this; the bug manifests in the specified
flags for `/dev` being lost on remount read only at present. As we
need to specify flags, split the code path for this from remounting
paths which are not mount points, as these can only inherit the
existing flags of the path, and these cannot be changed.
In the bind case, remove extra flags from the bind remount. A bind
mount can only be remounted read only, no other flags can be set,
all other flags are inherited from the parent. From the man page:
```
Since Linux 2.6.26, this flag can also be used to make an existing
bind mount read-only by specifying mountflags as:
MS_REMOUNT | MS_BIND | MS_RDONLY
Note that only the MS_RDONLY setting of the bind mount can be changed
in this manner.
```
MS_REC can only be set on the original bind, so move this. See note
in man page on bind mounts:
```
The remaining bits in the mountflags argument are also ignored, with
the exception of MS_REC.
```
Signed-off-by: Justin Cormack <justin.cormack@docker.com>
When checking if the provided networking namespace is the host
one or not, we should first check if it's a symbolic link or not
as in some cases we can use persistent networking namespace under
e.g. /var/run/netns/.
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
Both suffered from different race conditions.
SelinuxEnabled assigned selinuxEnabledChecked before selinuxEnabled.
Thus racing callers could see the wrong selinuxEnabled.
getSelinuxMountPoint assigned selinuxfs to "" before it know the right
value. Thus racing could see "" improperly.
The gate selinuxfs, enabled, and mclist all on the same lock
This fixes all of the tests that were broken as part of the console
rewrite. This includes fixing the integration tests that used TTY
handling inside libcontainer, as well as the bats integration tests that
needed to be rewritten to use recvtty (as they rely on detached
containers that are running).
This patch is part of the console rewrite patchset.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
Switch to the actual source of the official Docker library of images, so
that we have a proper source for the test filesystem. In addition,
update to the latest released version (1.25.0 [2016-06-23]) so that we
can use more up-to-date applets in our tests (such as stat(3)).
This patch is part of the console rewrite patchset.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
This allows for higher-level orchestrators to be able to have access to
the master pty file descriptor without keeping the runC process running.
This is key to having (detach && createTTY) with a _real_ pty created
inside the container, which is then sent to a higher level orchestrator
over an AF_UNIX socket.
This patch is part of the console rewrite patchset.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
Since the gid=X and mode=Y flags can be set inside config.json as mount
options, don't override them with our own defaults. This avoids
/dev/pts/* not being owned by tty in a regular container, as well as all
of the issues with us implementing grantpt(3) manually. This is the
least opinionated approach to take.
This patch is part of the console rewrite patchset.
Reported-by: Mrunal Patel <mrunalp@gmail.com>
Signed-off-by: Aleksa Sarai <asarai@suse.de>
This implements {createTTY, detach} and all of the combinations and
negations of the two that were previously implemented. There are some
valid questions about out-of-OCI-scope topics like !createTTY and how
things should be handled (why do we dup the current stdio to the
process, and how is that not a security issue). However, these will be
dealt with in a separate patchset.
In order to allow for late console setup, split setupRootfs into the
"preparation" section where all of the mounts are created and the
"finalize" section where we pivot_root and set things as ro. In between
the two we can set up all of the console mountpoints and symlinks we
need.
We use two-stage synchronisation to ensures that when the syscalls are
reordered in a suboptimal way, an out-of-place read() on the parentPipe
will not gobble the ancilliary information.
This patch is part of the console rewrite patchset.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
To make the code cleaner, and more clear, refactor the syncT handling
used when creating the `runc init` process. In addition, document the
state changes so that people actually understand what is going on.
Rather than only using syncT for the standard initProcess, use it for
both initProcess and setnsProcess. This removes some special cases, as
well as allowing for the use of syncT with setnsProcess.
Also remove a bunch of the boilerplate around syncT handling.
This patch is part of the console rewrite patchset.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
This adds C wrappers for sendmsg and recvmsg, specifically used for
passing around file descriptors in Go. The wrappers (sendfd, recvfd)
expect to be called in a context where it makes sense (where the other
side is carrying out the corresponding action).
This patch is part of the console rewrite patchset.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
In the cases that we got failure on a subsystem's Apply,
we'll get some subsystems' cgroup directories leftover.
On Docker's point of view, start a container failed, use
`docker rm` to remove the container, but some cgroup files
are leftover.
Sometimes we don't want to clean everyting up when something
went wrong, because we need these inter situation
information to debug what's going on, but cgroup directories
are not useful information we want to keep.
Signed-off-by: Qiang Huang <h.huangqiang@huawei.com>
This PR fix issue in this scenario:
```
in terminal 1:
~# cd /sys/fs/cgroup/cpuset
~# mkdir test
~# cd test
~# cat cpuset.cpus
0-3
~# echo 1 > cpuset.cpu_exclusive (make sure you don't have other cgroups under root)
in terminal 2:
~# echo $$ > /sys/fs/cgroup/cpuset/test/tasks
// set resources.cpu.cpus="0-2" in config.json
~# runc run test1
back to terminal 1:
~# cd test1
~# cat cpuset.cpus
0-2
~# echo 1 > cpuset.cpu_exclusive
in terminal 3:
~# echo $$ > /sys/fs/cgroup/test/tasks
// set resources.cpu.cpus="3" in config.json
~# runc run test2
container_linux.go:247: starting container process caused "process_linux.go:258:
applying cgroup configuration for process caused \"failed to write 0-3\\n to
cpuset.cpus: write /sys/fs/cgroup/cpuset/test2/cpuset.cpus: invalid argument\""
```
Signed-off-by: Qiang Huang <h.huangqiang@huawei.com>
This allows a user to send a signal to all the processes in the
container within a single atomic action to avoid new processes being
forked off before the signal can be sent.
This is basically taking functionality that we already use being
`delete` and exposing it ok the `kill` command by adding a flag.
Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
This moves the ambient capability support behind an `ambient` build tag
so that it is only compiled upon request.
Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
The default terminal setting for a new pty on Linux (unix98) has +ONLCR,
resulting in '\n' writes by a container process to be converted to
'\r\n' reads by the managing process. This is quite unexpected, and
causes multiple issues with things like bats testing. To fix it, make
the terminal sane after opening it by setting -ONLCR.
This patch might need to be rewritten after the console rewrite patchset
is merged.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
I use the same tool (https://github.com/client9/misspell)
as Daniel used a few days ago, don't why he missed these
typos at that time.
Signed-off-by: Qiang Huang <h.huangqiang@huawei.com>
config.cloneflag is not mandatory, when using `runc exec`,
config.cloneflag can be empty, and even then it won't be
`-1` but `0`.
So this validation is totally wrong and unneeded.
Signed-off-by: Qiang Huang <h.huangqiang@huawei.com>
In user namespaces devices are bind-mounted from the host, so
we need to add them as external mounts for CRIU.
Reported-by: Ross Boucher <boucher@gmail.com>
Signed-off-by: Andrei Vagin <avagin@virtuozzo.com>
When spec file contains duplicated namespaces, e.g.
specs: specs.Spec{
Linux: &specs.Linux{
Namespaces: []specs.Namespace{
{
Type: "pid",
},
{
Type: "pid",
Path: "/proc/1/ns/pid",
},
},
},
}
runc should report malformed spec instead of using latest one by
default, because this spec could be quite confusing.
Signed-off-by: Zhang Wei <zhangwei555@huawei.com>
Previously we only tested failures, which causes us to miss issues where
setting sysctls would *always* fail.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
When changing this validation, the code actually allowing the validation
to pass was removed. This meant that any net.* sysctl would always fail
to validate.
Fixes: bc84f83344 ("fix docker/docker#27484")
Reported-by: Justin Cormack <justin.cormack@docker.com>
Signed-off-by: Aleksa Sarai <asarai@suse.de>
This reverts part of the commit eb0a144b5e
That commit introduced two issues.
- We need to make parent mount of rootfs private before bind mounting
rootfs. Otherwise bind mounting root can propagate in other mount
namespaces. (If parent mount is shared).
- It broke test TestRootfsPropagationSharedMount() on Fedora.
On fedora /tmp is a mount point with "shared" propagation. I think
you should be able to reproduce it on other distributions as well
as long as you mount tmpfs on /tmp and make it "shared" propagation.
Reason for failure is that pivot_root() fails. And it fails because
kernel does following check.
IS_MNT_SHARED(new_mnt->mnt_parent)
Say /tmp/foo is new rootfs, we have bind mounted rootfs, so new_mnt
is /tmp/foo, and new_mnt->mnt_parent is /tmp which is "shared" on
fedora and above check fails.
So this change broke few things, it is a good idea to revert part of it.
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Namely, use an undocumented feature of pivot_root(2) where
pivot_root(".", ".") is actually a feature and allows you to make the
old_root be tied to your /proc/self/cwd in a way that makes unmounting
easy. Thanks a lot to the LXC developers which came up with this idea
first.
This is the first step of many to allowing runC to work with a
completely read-only rootfs.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
Without this patch applied, RHEL's SELinux policies cause container
creation to not really work. Unfortunately this might be an issue for
rootless containers (opencontainers/runc#774) but we'll cross that
bridge when we come to it.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
Print the error message to stderr if we are unable to return it back via
the pipe to the parent process. Also, don't panic here as it is most
likely a system or user error and not a programmer error.
Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
We need support for read/only mounts in SELinux to allow a bunch of
containers to share the same read/only image. In order to do this
we need a new label which allows container processes to read/execute
all files but not write them.
Existing mount label is either shared write or private write. This
label is shared read/execute.
Signed-off-by: Dan Walsh <dwalsh@redhat.com>
At some point InitLabels was changed to look for SecuritOptions
separated by a ":" rather then an "=", but DupSecOpt was never
changed to match this default.
Signed-off-by: Dan Walsh <dwalsh@redhat.com>
If copyup is specified for a tmpfs mount, then the contents of the
underlying directory are copied into the tmpfs mounted over it.
Signed-off-by: Mrunal Patel <mrunalp@gmail.com>
Depending on your SELinux setup, the order in which you join namespaces
can be important. In general, user namespaces should *always* be joined
and unshared first because then the other namespaces are correctly
pinned and you have the right priviliges within them. This also is very
useful for rootless containers, as well as older kernels that had
essentially broken unshare(2) and clone(2) implementations.
This also includes huge refactorings in how we spawn processes for
complicated reasons that I don't want to get into because it will make
me spiral into a cloud of rage. The reasoning is in the giant comment in
clone_parent. Have fun.
In addition, because we now create multiple children with CLONE_PARENT,
we cannot wait for them to SIGCHLD us in the case of a death. Thus, we
have to resort to having a child kindly send us their exit code before
they die. Hopefully this all works okay, but at this point there's not
much more than we can do.
Signed-off-by: Aleksa Sarai <asarai@suse.de>
Mrunal Patel
Michael Crosby
Xiaochen Shen
s7v7nislands
Ma Shimiao
Adrian Reber
Aleksa Sarai
Qiang Huang
Tycho Andersen
Nikolas Sepos
Kenfe-Mickael Laventure
Alex Fang
Daniel, Dao Quang Minh
Tobias Klauser
yangshukui
Steven Hartland
Yuanhong Peng
Justin Cormack
W. Trevor King
Kang Liang
Vladimir Stefanovic
Christy Perez
Wentao Zhang
Valentin Rothberg
Harshal Patil
Andrei Vagin
chchliang
CuiHaozhi
Derek Carr
Mrunal Patel
Craig Furman
Carlton-Semple
xuxinkun
Brian Goff
Mohammad Arab
Carlton Semple
Deng Guangxing
Antonio Murdaca
rainrambler
Wang Long
Lei Jitang
Zhang Wei
Samuel Ortiz
Eric Paris
yupeng
Xianlu Bird
allencloud
rajasec
Crazykev
Vivek Goyal
Ce Gao
Alexander Morozov
Yong Tang
Dan Walsh
Xianglin Gao