Commit Graph

23 Commits

Author SHA1 Message Date
Michael Crosby ba2ce3b25a Cgroup set order for systemd
Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
2015-10-19 13:32:45 -07:00
Michael Crosby 2554f49d5e Use array instead of map for cgroup subsystems
Also add cpuset as the first in the list to address issues setting the
pid in any cgroup before the cpuset is populated.

Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
2015-10-15 15:24:53 -07:00
Michael Crosby 02fdc70837 Add Name() to cgroup subsystems
Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
2015-10-15 15:19:23 -07:00
Mrunal Patel 3be7f87b1b Merge pull request #334 from hqhq/hq_set_cpus_mems_first
Set cpuset.cpus and cpuset.mems before join the cgroup
2015-10-15 14:33:28 -07:00
Qiang Huang be6764508e Set cpuset.cpus and cpuset.mems before join the cgroup
It can avoid unnecessary task migrataion, see this scenario:
 - container init task is on cpu 1, and we assigned it to cpu 1,
   but parent cgroup's cpuset.cpus=2
 - we created the cgroup dir and inherited cpuset.cpus from parent as 2
 - write container init task's pid to cgroup.procs
 - [it's possibile the container init task migrated to cpu 2 here]
 - set cpuset.cpus as assigned to cpu 1
 - [the container init task has to be migrated back to cpu 1]

So we should set cpuset.cpus and cpuset.mems before writing pids
to cgroup.procs to aviod such problem.

Signed-off-by: Qiang Huang <h.huangqiang@huawei.com>
2015-10-15 11:16:56 +08:00
Alexander Morozov 6dad176d01 Get PIDs from cgroups recursively
Also lookup cgroup for systemd is changed to "device" to be consistent
with fs implementation.

Signed-off-by: Alexander Morozov <lk4d4@docker.com>
2015-10-13 10:19:01 -07:00
Mrunal Patel 223975564a Merge pull request #276 from runcom/adapt-spec-96bcd043aa8a28f6f64c95ad61329765f01de1ba
Adapt spec 96bcd043aa
2015-10-05 16:36:09 -07:00
Mrunal Patel 1940c73777 cgroups: Add a name cgroup
This is meant to be used in retrieving the paths so an exec
process enters all the cgroup paths correctly.

Signed-off-by: Mrunal Patel <mrunalp@gmail.com>
2015-10-05 14:23:05 -04:00
Antonio Murdaca c6e406af24 Adjust runc to new opencontainers/specs version
Godeps: Vendor opencontainers/specs 96bcd043aa

Fix a bug where it's impossible to pass multiple devices to blkio
cgroup controller files. See https://github.com/opencontainers/runc/issues/274

Signed-off-by: Antonio Murdaca <runcom@linux.com>
2015-10-03 12:25:33 +02:00
Alexander Morozov e32b3442ec Run tests for all HugetlbSizes
Signed-off-by: Alexander Morozov <lk4d4@docker.com>
2015-09-29 17:08:41 -07:00
Qiang Huang f2ec7eff7e Rename FindCgroupMountpointAndSource
Rename it to FindCgroupMountpointAndRoot.

Signed-off-by: Qiang Huang <h.huangqiang@huawei.com>
2015-09-09 09:29:11 +08:00
Mrunal Patel c20bda3f71 Merge pull request #206 from mountkin/ensure-cleanup
Ensure the cleanup jobs in the deferrer are executed on error
2015-08-18 14:16:31 -07:00
Shijiang Wei f0679089b9 Ensure the cleanup jobs in the deferrer are executed on error
Signed-off-by: Shijiang Wei <mountkin@gmail.com>
2015-08-16 12:29:04 +08:00
Alexander Morozov 2b28b3c276 Always use cgroup root of current process
Because for host PID namespace /proc/1/cgroup can point to whole other
world of cgroups.

Signed-off-by: Alexander Morozov <lk4d4@docker.com>
2015-08-11 18:04:59 -07:00
Alexander Morozov 5aa6005498 Revert "Fix cgroup parent searching"
This reverts commit 2f9052ca29.

Signed-off-by: Alexander Morozov <lk4d4@docker.com>
2015-08-11 18:04:55 -07:00
Alexander Morozov 2f9052ca29 Fix cgroup parent searching
I had pretty convenient input data to miss this bug.

Signed-off-by: Alexander Morozov <lk4d4@docker.com>
2015-08-10 14:30:05 -07:00
Kir Kolyshkin 6f82d4b544 Simplify and fix os.MkdirAll() usage
TL;DR: check for IsExist(err) after a failed MkdirAll() is both
redundant and wrong -- so two reasons to remove it.

Quoting MkdirAll documentation:

> MkdirAll creates a directory named path, along with any necessary
> parents, and returns nil, or else returns an error. If path
> is already a directory, MkdirAll does nothing and returns nil.

This means two things:

1. If a directory to be created already exists, no error is
returned.

2. If the error returned is IsExist (EEXIST), it means there exists
a non-directory with the same name as MkdirAll need to use for
directory. Example: we want to MkdirAll("a/b"), but file "a"
(or "a/b") already exists, so MkdirAll fails.

The above is a theory, based on quoted documentation and my UNIX
knowledge.

3. In practice, though, current MkdirAll implementation [1] returns
ENOTDIR in most of cases described in #2, with the exception when
there is a race between MkdirAll and someone else creating the
last component of MkdirAll argument as a file. In this very case
MkdirAll() will indeed return EEXIST.

Because of #1, IsExist check after MkdirAll is not needed.

Because of #2 and #3, ignoring IsExist error is just plain wrong,
as directory we require is not created. It's cleaner to report
the error now.

Note this error is all over the tree, I guess due to copy-paste,
or trying to follow the same usage pattern as for Mkdir(),
or some not quite correct examples on the Internet.

[1] https://github.com/golang/go/blob/f9ed2f75/src/os/path.go

Signed-off-by: Kir Kolyshkin <kir@openvz.org>
2015-07-29 18:03:27 -07:00
Alexander Morozov c0e18b96fb Fix subsystem path with abs parent
Sometimes subsystem can be mounted to path like "subsystem1,subsystem2",
so we need to handle this.

Signed-off-by: Alexander Morozov <lk4d4@docker.com>
2015-07-20 11:48:58 -07:00
Alexander Morozov fc31076c23 Substract source mount from cgroup dir
This is needed because for nested containers cgroups. Without this patch
they creating unnecessary intermediate cgroup like:
/sys/fs/cgroup/memory/system.slice/docker-9409d9f0b68fb9e9d7d532d5b3f35e7c7f9cca1312af392ae3b28436f1f2998f.scope/system.slice/docker-9409d9f0b68fb9e9d7d532d5b3f35e7c7f9cca1312af392ae3b28436f1f2998f.scope/docker/908ebcc9c13584a14322ec070bd971e0de62f126c0cd95c079acdb99990ad3a3

It is because in /proc/self/cgroup we see paths from host, and they don't
exist in container.

Signed-off-by: Alexander Morozov <lk4d4@docker.com>
2015-07-17 11:41:58 -07:00
Qiang Huang 4e244108ef Fix error when memory cgroup not mounted
Fixes: #57

Normally all cgroup subsystems are optional except device cgroup,
but memory cgroup optional was broken by:
https://github.com/docker/libcontainer/pull/637

This patch fixes this.

Signed-off-by: Qiang Huang <h.huangqiang@huawei.com>
2015-07-13 18:22:35 +08:00
Raghavendra K T 88104a4444 Treat -1 as default value for memory swappiness.
In some older kernels setting swappiness fails. This happens even
when nobody tries to configure swappiness from docker UI because
we would still get some default value from host config.
With this we treat -1 value as default value (set implicitly) and skip
the enforcement of swappiness.

However from the docker UI setting an invalid value anything other than
0-100 including -1 should fail. This patch enables that fix in docker UI.

without this fix container creation with invalid value succeeds with a
default value (60) which in incorrect.

Signed-off-by: Raghavendra K T <raghavendra.kt@linux.vnet.ibm.com>
2015-07-03 18:19:45 +05:30
Michael Crosby 080df7ab88 Update import paths for new repository
Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
2015-06-21 19:29:59 -07:00
Michael Crosby 8f97d39dd2 Move libcontainer into subdirectory
Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
2015-06-21 19:29:15 -07:00