Merge pull request #27016 from andyzheng0831/gci-docker0

Automatic merge from submit-queue GCI: fix the issue #26379 This PR deletes docker0 explicitly to fix the issue. In some cases, coexistence of docker0 and cbr0 make troubles in GCI-based cluster instances. I verified it in GKE. With the fix, fluentd-gcp pod shows no error. "curl google.com" can work inside a pod. Mark it as P0 to match the issue priority. @a-robinson @roberthbailey @freehan @kubernetes/goog-image

Merge pull request #27016 from andyzheng0831/gci-docker0
d53ddbe6 · k8s-merge-robot · 46f16718 · 64ffe905 · d53ddbe6
Commit d53ddbe6 authored Jun 08, 2016 by k8s-merge-robot
Hide whitespace changes
Inline Side-by-side

Showing with 5 additions and 0 deletions

configure-helper.sh cluster/gce/gci/configure-helper.sh +5 -0

No files found.
--- a/cluster/gce/gci/configure-helper.sh
+++ b/cluster/gce/gci/configure-helper.sh
@@ -359,6 +359,11 @@ function start-kubelet {
  fi
  echo "KUBELET_OPTS=\"${flags}\"" > /etc/default/kubelet

+  # Delete docker0 to avoid interference
+  iptables -t nat -F || true
+  ip link set docker0 down || true
+  brctl delbr docker0 || true
+
  systemctl start kubelet.service
 }