scottk/kubernetes
1
0
Fork 0
mirror of https://github.com/optim-enterprises-bv/kubernetes.git synced 2025-11-02 03:08:15 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #42767 from Random-Liu/cleanup-infra-container-on-error

Browse Source 
Automatic merge from submit-queue (batch tested with PRs 42768, 42760, 42771, 42767)

Stop sandbox container when hit network error.

Fixes https://github.com/kubernetes/kubernetes/issues/42698.

This PR stops the sandbox container when hitting a network error.
This PR also adds a unit test for it.

I'm not sure whether we should try teardown pod network after `SetUpPod` failure. We don't do that in dockertools https://github.com/kubernetes/kubernetes/blob/master/pkg/kubelet/dockertools/docker_manager.go#L2276.

@yujuhong @freehan
This commit is contained in:
Kubernetes Submit Queue
2017-03-09 00:08:01 -08:00
committed by GitHub
parent cc28936bd1 2690461cbb
commit 4cf553f78e
2 changed files with 52 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
pkg/kubelet/dockershim/docker_sandbox.go
View File

@@ -105,7 +105,12 @@ func (ds *dockerService) RunPodSandbox(config *runtimeapi.PodSandboxConfig) (str
// recognized by the CNI standard yet.
cID := kubecontainer.BuildContainerID(runtimeName, createResp.ID)
err = ds.network.SetUpPod(config.GetMetadata().Namespace, config.GetMetadata().Name, cID, config.Annotations)
// TODO: Do we need to teardown on failure or can we rely on a StopPodSandbox call with the given ID?
if err != nil {
// TODO(random-liu): Do we need to teardown network here?
if err := ds.client.StopContainer(createResp.ID, defaultSandboxGracePeriod); err != nil {
glog.Warningf("Failed to stop sandbox container %q for pod %q: %v", createResp.ID, config.Metadata.Name, err)
}
}
return createResp.ID, err
}