Due to the iptables package being `init`ed at start of the docker
runtime, this means the iptables --wait command listing all rules
is run, no matter if the command is simply "docker -h". It makes
more sense to both locate the iptables command and check for the
wait flag support at the time iptables is actually used, as it
may not be used at all if certain network support is off/configured
differently.
Docker-DCO-1.1-Signed-off-by: Phil Estes <estesp@linux.vnet.ibm.com> (github: estesp)
Some calls like json.Encoder.Encode mask the number of bytes written to
an io.Writer. The solution provides a wrapper io.Writer around the
actual io.Writer that allows multiple calls to Write to be considered as
one and allow access to this count.
Signed-off-by: Pierre Wacrenier <pierre.wacrenier@gmail.com>
Also, this decouples the emptyfs script from the busybox one -- they're now functionally separate thanks to the scratch no-op change. 👍
Signed-off-by: Andrew "Tianon" Page <admwiggin@gmail.com>
Setting a header for all responses can help clients better determine
if the server speaks the legacy v1 API or the v2 API. It is important
that the header be set *BEFORE* routing the request.
Docker-DCO-1.1-Signed-off-by: Josh Hawn <josh.hawn@docker.com> (github: jlhawn)
unshare the mount namespace of the docker daemon to avoid other pids
outside the daemon holding mount references of docker containers.
Signed-off-by: Vincent Batts <vbatts@redhat.com>
Fixes#9629#9768
A couple of issues:
1) Volume config is not restored if we couldn't find it with the graph
driver, but bind-mounts would never be found by the graph driver since
they aren't in that dir
2) container volumes were only being restored if they were found in the
volumes repo, but volumes created by old daemons wouldn't be in the
repo until the container is at least started.
Signed-off-by: Brian Goff <cpuguy83@gmail.com>
This systemd.exec setting will construct a new mount namespace for the
docker daemon, and use slave shared-subtree mounts so that volume mounts
propogate correctly into containers.
By having an unshared mount namespace for the daemon it ensures that
mount references are not held by other pids outside of the docker
daemon. Frequently this can be seen in EBUSY or "device or resource
busy" errors.
Signed-off-by: Vincent Batts <vbatts@redhat.com>
This is literally the only failing test on Go 1.3.3: 🎉
```
--- FAIL: TestBuildWithTabs (0.43 seconds)
docker_cli_build_test.go:4307: Missing tabs.
Got:["/bin/sh","-c","echo\u0009one\u0009\u0009two"]
Exp:["/bin/sh","-c","echo\tone\t\ttwo"]
```
Signed-off-by: Andrew "Tianon" Page <admwiggin@gmail.com>
The assumption is not true if user specifies an IP address other than
the first IP, in that case the first IP address is never allocated to
any container.
Signed-off-by: Hu Tao <hutao@cn.fujitsu.com>
