This adds null IO option for efficient handling of IO.
It provides a container directly with `/dev/null` and does not require
any io.Copy within the shim whenever a user does not want the IO of the
container.
Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
Adding image.Config,image.Size,image.RootFS to retrieve the internal image information
which will be needed by consumers of containerd
Signed-off-by: Abhinandan Prativadi <abhi@docker.com>
This could render tasks for a container unresolvalbe. If there is a use
case for changing the runtime of a container, we should think it through
carefully.
Signed-off-by: Stephen J Day <stephen.day@docker.com>
Differ is updated to set a label for the uncompressed hash
of compressed content. This allows compressed blobs to
be used and looked up for their uncompressed hashes.
Uses commit options to set labels.
Updates rootfs default to create compressed archives.
Signed-off-by: Derek McGowan <derek@mcgstyle.net>
Add commit options which allow for setting labels on commit.
Prevents potential race between garbage collector reading labels
after commit and labels getting set.
Signed-off-by: Derek McGowan <derek@mcgstyle.net>
This is needed to support daemon restart as the unix socket is recreated at
every start.
Signed-off-by: Kenfe-Mickael Laventure <mickael.laventure@gmail.com>
Define the release process for containerd and outline the components
that are and are not covered by versioning guarantees. Please read the
document for details.
Signed-off-by: Stephen J Day <stephen.day@docker.com>
Because runc will delete a container after a successful checkpoint we
need to handle a NotFound error from runc on delete.
There is also a race between SIGKILL'ing the shim and it actually
exiting to unmount the tasks rootfs, we need to loop and wait for the
task to actually be reaped before trying to delete the rootfs+bundle.
Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
This adds a `stress` binary to help stress test containerd. It is
different from a benchmarking tool as it only gives a simple summary at
the end.
It is built to run long, multi hour/day stress tests across builds of
containerd.
Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
By moving the binaries to be rooted at `bin/` in the tar file,
installation can be done from a tar file in one fell swoop:
```console
$ make release
$ sudo tar -C /usr/local -xvf releases/containerd-1.0.0-alpha3-72-ge6ba48f2.m.linux-amd64.tar.gz
```
Signed-off-by: Stephen J Day <stephen.day@docker.com>
