Now that we have most of the services required for use with containerd,
it was found that common patterns were used throughout services. By
defining a central `errdefs` package, we ensure that services will map
errors to and from grpc consistently and cleanly. One can decorate an
error with as much context as necessary, using `pkg/errors` and still
have the error mapped correctly via grpc.
We make a few sacrifices. At this point, the common errors we use across
the repository all map directly to grpc error codes. While this seems
positively crazy, it actually works out quite well. The error conditions
that were specific weren't super necessary and the ones that were
necessary now simply have better context information. We lose the
ability to add new codes, but this constraint may not be a bad thing.
Effectively, as long as one uses the errors defined in `errdefs`, the
error class will be mapped correctly across the grpc boundary and
everything will be good. If you don't use those definitions, the error
maps to "unknown" and the error message is preserved.
Signed-off-by: Stephen J Day <stephen.day@docker.com>
Based on feedback, a few adjustments have been made to the identifier
requirements. Identifiers that have components that start with a number
are now allowed. This violates RFC 1035 but does not do so for dns names
practically. A total length, of 76 characters is now also enforced. This
decision was completely arbitrary but satifies the requirement for a
maximum length. Often, this is used as the maximum length of a line in
editors, so it should be a good choice.
Signed-off-by: Stephen J Day <stephen.day@docker.com>
# github.com/containerd/containerd
./container_test.go:769: cannot use &limit (type *uint64) as type *int64 in field value
./container_test.go:812: cannot use &limit (type *uint64) as type *int64 in field value
Signed-off-by: Akihiro Suda <suda.akihiro@lab.ntt.co.jp>
Until we have a way to preserve the initial command used to start the
container, we have to default to the default `runc` found on the $PATH.
This code after the last refactor of shim/API is incorrectly using the
gRPC object reference of the v1 runtime as a binary name which causes
os.Exec() errors.
Signed-off-by: Phil Estes <estesp@gmail.com>
Signed-off-by: Phil Estes <estesp@linux.vnet.ibm.com>
Fix the behavior of removing snapshot on container delete.
Adds a flag to keep the snapshot if desired.
Signed-off-by: Derek McGowan <derek@mcgstyle.net>
Runtime is not printed while container listing due to typo introduced
in #935.
This fixes the Typo.
Signed-off-by: Kunal Kushwaha <kushwaha_kunal_v7@lab.ntt.co.jp>
A few days ago, we added validation for namespaces. We've decided to
expand these naming rules to include containers. To facilitate this, a
common package `identifiers` now provides a common validation area.
These rules will be extended to apply to task identifiers, snapshot keys
and other areas where user-provided identifiers may be used.
Signed-off-by: Stephen J Day <stephen.day@docker.com>
Other: How do you know these event conversions work?
Steve: Well, they've been tested and benchmarked, of course!
Other: Where are the tests?
Steve: It's easy, just check out the branch you want to test on my
machine and the tests are right there. So easy!
Other: <facepalm>
Signed-off-by: Stephen J Day <stephen.day@docker.com>
This moves the shim's API and protos out of the containerd services
package and into the linux runtime package. This is because the shim is
an implementation detail of the linux runtime that we have and it is not
a containerd user facing api.
Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
To simplify use of types, we have consolidate the packages for the mount
and descriptor protobuf types into a single Go package. We also drop the
versioning from the type packages, as these types will remain the same
between versions.
Signed-off-by: Stephen J Day <stephen.day@docker.com>
