Merge pull request #43462 from csbell/federation-up-timeout

Automatic merge from submit-queue [Federation][e2e] Ensure kubefed times out in federation-up.sh Although this should eventually be moved into kubefed itself, monitor kubefed from federation-up.sh and force it to timeout after being unable to initialize. The motivating factor here is to ensure that CI can timeout after a reasonable attempt at trying to initialize the FCP.

Merge pull request #43462 from csbell/federation-up-timeout
827591cc · Kubernetes Submit Queue · GitHub · 63d8e244 · 699c3ad2 · 827591cc
Commit 827591cc authored Mar 21, 2017 by Kubernetes Submit Queue Committed by GitHub Mar 21, 2017
Hide whitespace changes
Inline Side-by-side

Showing with 3 additions and 1 deletion

federation-up.sh federation/cluster/federation-up.sh +3 -1

No files found.
--- a/federation/cluster/federation-up.sh
+++ b/federation/cluster/federation-up.sh
@@ -82,7 +82,9 @@ function init() {
  kube::log::status "DNS_ZONE_NAME: \"${DNS_ZONE_NAME}\", DNS_PROVIDER: \"${DNS_PROVIDER}\""
  kube::log::status "Image: \"${kube_registry}/hyperkube-amd64:${kube_version}\""
-  "${KUBE_ROOT}/federation/develop/kubefed.sh" init \
+  # Send INT after 20m and KILL 1m after that if process is still alive.
+  timeout --signal=INT --kill-after=1m 20m \
+      "${KUBE_ROOT}/federation/develop/kubefed.sh" init \
      "${FEDERATION_NAME}" \
      --host-cluster-context="${HOST_CLUSTER_CONTEXT}" \
      --dns-zone-name="${DNS_ZONE_NAME}" \