Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add linux.resources.devices and drop linux.devices #98

Closed
wants to merge 2 commits into from
Closed
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
80 changes: 64 additions & 16 deletions config-linux.md
Original file line number Diff line number Diff line change
Expand Up @@ -55,28 +55,71 @@ within the container.

### Access to devices

Devices is an array specifying the list of devices from the host to make available in the container.
By providing a device name within the list the runtime should look up the same device on the host's `/dev`
and collect information about the device node so that it can be recreated for the container. The runtime
should not only create the device inside the container but ensure that the root user inside
the container has access rights for the device.
Devices required by the application should be supplied via the bundle filesystems and mounted via [mounts][].
Bundle authors can create these files using [`mknod`][] or by copying nodes from their local host.
For example:

```shell
$ mknod --mode a=rw rootfs/dev/random c 1 8
$ cp --archive /dev/tty rootfs/dev/tty
```

## Linux control groups

Also known as cgroups, they are used to restrict resource usage for a container and handle device access.
For more information, see the [kernel cgroups documentation][cgroups].
You can configure a container's cgroups via the "resources" field of the Linux configuration.

### Disable out-of-memory killer

FIXME

### Memory

FIXME

### CPU

FIXME

### Block I/O

FIXME

### Devices

Container-side devices are [mounted from the bundle filesystems][mount-devices].
Bundle authors can set major and minor nodes, owner IDs, filesystem permissions, etc. by altering those filesystems.
However, you cannot pass cgroup information via the bundle filesystem, so bundle authors that need special device cgroups should use the "devices" field of the resource configuration.
The fields are discussed [in the kernel documentation][cgroups-devices].
The entries are applied to the container in the order that they are listed in the configuration.

```json
"devices": [
"null",
"random",
"full",
"tty",
"zero",
"urandom"
]
{
"allow": false,
"type": "a",
"major": "*",
"minor": "*",
"access": "rwm",
},
{
"allow": true,
"type": "c",
"major": "1",
"minor": "3",
"access": "mr",
}
]
```

## Linux control groups
### Huge page limits

FIXME

### Network

Also known as cgroups, they are used to restrict resource usage for a container and handle
device access. cgroups provide controls to restrict cpu, memory, IO, and network for
the container. For more information, see the [kernel cgroups documentation](https://www.kernel.org/doc/Documentation/cgroups/cgroups.txt)
FIXME

## Linux capabilities

Expand Down Expand Up @@ -150,3 +193,8 @@ rootfsPropagation sets the rootfs's mount propagation. Its value is either slave

**TODO:** security profiles

[mounts]: config.md#mount-configuration
[mknod]: http://linux.die.net/man/1/mknod
[cgroups]: https://www.kernel.org/doc/Documentation/cgroups/cgroups.txt
[cgroups-devices]: https://www.kernel.org/doc/Documentation/cgroups/devices.txt
[mount-devices]: #access-to-devices
19 changes: 17 additions & 2 deletions spec_linux.go
Original file line number Diff line number Diff line change
Expand Up @@ -26,8 +26,6 @@ type Linux struct {
Namespaces []Namespace `json:"namespaces"`
// Capabilities are Linux capabilities that are kept for the container
Capabilities []string `json:"capabilities"`
// Devices are a list of device nodes that are created and enabled for the container
Devices []string `json:"devices"`
// RootfsPropagation is the rootfs mount propagation mode for the container
RootfsPropagation string `json:"rootfsPropagation"`
}
Expand Down Expand Up @@ -102,6 +100,21 @@ type BlockIO struct {
ThrottleWriteIOpsDevice string `json:"blkioThrottleWriteIopsDevice"`
}

// Device rule for Linux cgroup management
type Device struct {
// Whether the device is allowed (true) or denied (false)
Allow bool `json:"allow"`
// a (all), c (char), or b (block). 'all' means it applies to all
// types and all major and minor numbers
Type string `json:type`
// Major number. Either an integer or '*' for all.
Major string `json:major`
// Minor number. Either an integer or '*' for all.
Minor string `json:minor`
// a composition of r (read), w (write), and m (mknod).
Access string `json:access`
}

// Memory for Linux cgroup 'memory' resource management
type Memory struct {
// Memory limit (in bytes)
Expand Down Expand Up @@ -152,6 +165,8 @@ type Resources struct {
CPU CPU `json:"cpu"`
// BlockIO restriction configuration
BlockIO BlockIO `json:"blockIO"`
// Device configuration
Devices []Device `json:"devices"`
// Hugetlb limit (in bytes)
HugepageLimits []HugepageLimit `json:"hugepageLimits"`
// Network restriction configuration
Expand Down