Not sure if this is the right setup given the containerd change but I need
to have the built version of the nested exes (containerd, runc...) available
to me after the build is completed so I'm always testing using the latest
versions. This PR will copy them into the same bundles dir so people can
them use them if they wish w/o having to build each separately.
Signed-off-by: Doug Davis <dug@us.ibm.com>
Upstream-commit: 1bf5eb20e53b7e242792fcbe399cb997b6a2ba4b
Component: engine
This version supports async connections.
Signed-off-by: Pierre Carrier <pierre@meteor.com>
Upstream-commit: 9c2a0739fb9c5e3636425d669f0ab860ea1bfbae
Component: engine
This (the first tagged hcsshim release) fixes long-path bugs on
Windows TP5 that affect commit and save. These bugs were blocking
commit of Windows containers that had node.js installed.
Signed-off-by: John Starks <jostarks@microsoft.com>
Upstream-commit: b54058bafe07718a02e81a19b2dfc779dcdf11ba
Component: engine
It plumbs net/context.Context through entire API, see docker/engine-api#140
Signed-off-by: Vincent Demeester <vincent@sbr.pm>
Upstream-commit: 48339017dbb7a09827445d8c1a9db9ae6ecb8a4a
Component: engine
This revendor provides support for CloseWrite() in the npipe transport,
fixes a performance regression introduced in Go 1.6, and improves
npipe performance by allowing the pipe buffer size to be specified.
Signed-off-by: John Starks <jostarks@microsoft.com>
Upstream-commit: 87c2aad6f11c4993222dd29fb8c7c520b19ac8d9
Component: engine
Adjust "hack/make/.detect-daemon-osarch" to be the source of truth for "platform detection"
Upstream-commit: 133b3cccb50dadf2b8b10519e3c8d8b924a91b76
Component: engine
debhelper has changed the way it performs path validations and
building the deb package fails when it tries to compress the files.
Signed-off-by: David Calavera <david.calavera@gmail.com>
Upstream-commit: 027f4fdca678ae955db6f115d6af15d532f6cf4b
Component: engine
This test for libdevmapper was always silently failing because the
linker never got the `-ldevmapper` information. Putting the flag last
corrects the test.
Docker-DCO-1.1-Signed-off-by: Phil Estes <estesp@linux.vnet.ibm.com> (github: estesp)
Upstream-commit: e2076453071f9fc82478557919d327c958eec712
Component: engine
Instead of being split between three files, let's let `hack/make/.detect-daemon-osarch` be our single source of truth for multiarch detection/vars. Not only does it make it slightly easier to make sure we change everything properly when these bits have to change, but it also makes it so that all bits of `hack/make.sh` (especially `hack/make/.ensure-frozen-images`) work properly outside the context of the `Makefile` on all platforms.
Signed-off-by: Andrew "Tianon" Page <admwiggin@gmail.com>
Upstream-commit: a667cd88c3dc9262f4ef4e2a317c10f6cdcdf5ae
Component: engine
The version of sed on MacOS X is different then the one on linux. The mac version
requires a parameter for the inline (-i) flag, where this isn't required on linux.
On the mac it thinks the -e flag is the parameter, and it causes the vendoring script
to fail.
This fix adds an empty string '' as a parameter to sed, which works fine on both the
mac and linux versions.
Signed-off-by: Ken Cochrane <kencochrane@gmail.com>
Upstream-commit: 9488696814f481d2b0dce03d77657ec6860b9701
Component: engine
Pull in changes for refresh token in the registry client
Signed-off-by: Derek McGowan <derek@mcgstyle.net> (github: dmcgowan)
Upstream-commit: 5730259f324a59a6dddf63a950ab8eb9cde3ca0c
Component: engine
In order to solve the issue metioned in https://github.com/docker/docker/issues/18712
it is required to use rfc5424 log formatted message, made available in latest srslog library.
Signed-off-by: Solganik Alexander <solganik@gmail.com>
Upstream-commit: f988741823c1c510e97f2508d6a5bed68f65440a
Component: engine
This fixes docker save when the daemon was launched from a remote
PowerShell session.
Signed-off-by: John Starks <jostarks@microsoft.com>
Upstream-commit: 845dce846528f1ab7f6dc43b760b32aee666d9ac
Component: engine
