[DO NOT MERGE] Sandbox status error test #18424
Conversation
The CNI directory arguments are already the defaults in cni.go and we shouldn't need to change them (unless upstream changes them, at which point we should probably match). The Hairpin mode has been ignored upstream for CNI plugins since the switch to CRI. It's up to CNI network plugins to do what they need to do WRT allowing pods to talk back to themselves via the service IP. (note that hairpin is still used for --network-plugin=kubenet and for the old rkt runtime).
Instead of defining things in a couple places, just use the Kubelet CNI driver conf dir define. If that's overridden via the kubelet --cni-conf-dir argument, use that instead. Also pass a known IPAM data dir to the host-local plugin, and use that in the metrics code, to protect against upstream changes to the defaults.
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: dcbw Assign the PR to them by writing The full list of commands accepted by this bot can be found here.
Needs approval from an approver in each of these OWNERS Files:
You can indicate your approval by writing |
openshift-ci-robot
requested review from
danwinship,
simo5 and
smarterclayton
openshift-ci-robot
added
the
size/M
openshift-merge-robot
added
the
vendor-update
|
/test gcp |
|
/test extended_conformance_gce |
|
So this is Kubelet being stupid. It will run StopPodSandbox() multiple times in parallel for the same sandbox. And dockershim's StopPodSanbox() itself calls GetPodSandboxStatus() which checks the IP and that's where the error is coming from. That's because there's a race where the first StopPodSandbox is in TearDownPod() (and thus the networkReady suppression code isn't triggered yet) and the second StopPodSandbox() then asks for the pod IP. Before that can run nsenter, TearDownPod() completes and the first StopPodSandbox() stops the sandbox container and destroys the netns. Then the second StopPodSanbox()'s GetPodSandboxStatus() finally gets around to running nsenter and fails. |
|
@dcbw: The following tests failed, say
Full PR test history. Your PR dashboard. Please help us cut down on flakes by linking to an open issue when you hit one in your PR. Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. I understand the commands that are listed here. |
Just an attempt to reproduce #18414 on GCE since I cannot locally.