This reverts commit 74b069d0fa.
Now that pkg/kubelet/userns has a windows stub, there is no need to
handle this here. Let's remove this code so if it's ever called from
windows, we find an error and the bug surfaces immediately.
I've added it yesterday as an abundance of caution, to make sure the
tests go back to green in the master branch. But now that everything is
green again, it should be safe to remove it.
Signed-off-by: Rodrigo Campos <rodrigoca@microsoft.com>
Instead of hard-coding two instances of the hostpathplugin which listen on
certain socket paths, the hostpathplugin now gets started through Pod exec as
needed. The advantage is that the helper code is in charge of socket naming,
just like it would be in real deployment.
One nuisance is that exec.StreamWithContext always complains in copyFromStdout
and copyFromStderr when the remote hostpathplugin gets killed via context
cancellation:
E0312 11:56:31.637669 289446 v2.go:167] "Unhandled Error" err="next reader: read tcp [::1]:59006->[::1]:6444: use of closed network connection" logger="UnhandledError"
E0312 11:56:31.637749 289446 v2.go:150] "Unhandled Error" err="next reader: read tcp [::1]:59006->[::1]:6444: use of closed network connection" logger="UnhandledError"
These can be ignored.
When supporting rolling updates, we cannot use the same fixed socket paths for
old and new pod. With the revised API, the caller no longer specifies the full
socket paths, only directories. The logic about how to name sockets then can be
in the helper.
While at it, avoid passing a context to the gRPC helper code when
all that the helper code needs is a logger. That leads to confusion
about whether cancellation has an effect.
The goal is to simplify writing DRA drivers. This is also a first step towards
supporting seamless upgrades.
DRA drivers no longer need to implement the kubelet plugin API
directly. Instead, the helper wraps an implementation of an interface. The
helper then provides common functionality:
- retrieve and validate ResourceClaims
- serialize gRPC calls (enabled by default, can be opted out)
- gRPC logging
The definition of that interface is meant to be comprehensive enough that a
correct DRA driver can be implemented by following the documentation of the
package, without having to cross-reference KEPs.
The DRAPlugin interface used to be the abstract API of the helper. Now it's
what the DRA driver kubelet plugin needs to implement. The helper is a concrete
Server struct with no exported fields. It only exports the methods that
drivers need when using the helper.
While at it, support for the v1alpha4 API gets removed from the helper, which
implies removing the corresponding E2E tests. The kubelet implementation will
be dropped separately.
Noticed that cache might not nesseserly observe the write causing test
to flake. Fixed that changing the logic to require LessOrEqual of
writeRV instead of equal to writeRV. Also added comments explaining
edge cases.
The image puller's PullImage() method should be just a dumb pull
without any further logic. Make it accept everything it needs
to pull an image and defer any other magic to the image manager.
