github/kubernetes
1
0
Fork 0
mirror of https://github.com/k3s-io/kubernetes.git synced 2025-07-23 11:50:44 +00:00
Code Issues Projects Releases Wiki Activity

Merge pull request #51533 from shyamjvs/master-creation-retry

Browse Source 
Automatic merge from submit-queue

Retry master instance creation in case of retriable error (with sleep)

To help with our 5k-node CI tests failing to startup the cluster.
And also towards the greater goal - https://github.com/kubernetes/kubernetes/issues/43140

cc @kubernetes/sig-scalability-misc @kubernetes/sig-cluster-lifecycle-misc
This commit is contained in:
Kubernetes Submit Queue 2017-08-30 05:14:49 -07:00 committed by GitHub
commit e3210c6ccf
2 changed files with 58 additions and 26 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
cluster/gce/container-linux/master-helper.sh
View File

@ -66,6 +66,7 @@ function replicate-master-instance() {
function create-master-instance-internal() {
local gcloud="gcloud"
local retries=5
if [[ "${ENABLE_IP_ALIASES:-}" == 'true' ]]; then
gcloud="gcloud beta"
fi
@ -93,7 +94,8 @@ function create-master-instance-internal() {
disk="${disk},boot=no"
disk="${disk},auto-delete=no"
${gcloud} compute instances create "${master_name}" \
for attempt in $(seq 1 ${retries}); do
if result=$(${gcloud} compute instances create "${master_name}" \
--project "${PROJECT}" \
--zone "${ZONE}" \
--machine-type "${MASTER_SIZE}" \
@ -103,9 +105,23 @@ function create-master-instance-internal() {
--scopes "storage-ro,compute-rw,monitoring,logging-write" \
--metadata-from-file "${metadata}" \
--disk "${disk}" \
--boot-disk-size "${MASTER_ROOT_DISK_SIZE:-30}" \
--boot-disk-size "${MASTER_ROOT_DISK_SIZE}" \
${preemptible_master} \
${network}
${network} 2>&1); then
echo "${result}" >&2
return 0
else
echo "${result}" >&2
if [[ ! "${result}" =~ "try again later" ]]; then
echo "Failed to create master instance due to non-retryable error" >&2
return 1
fi
sleep 10
fi
done
echo "Failed to create master instance despite ${retries} attempts" >&2
return 1
}
function get-metadata() {

20
cluster/gce/gci/master-helper.sh
View File

@ -75,6 +75,7 @@ function replicate-master-instance() {
function create-master-instance-internal() {
local gcloud="gcloud"
local retries=5
if [[ "${ENABLE_IP_ALIASES:-}" == 'true' ]]; then
gcloud="gcloud beta"
fi
@ -106,7 +107,8 @@ function create-master-instance-internal() {
disk="${disk},boot=no"
disk="${disk},auto-delete=no"
${gcloud} compute instances create "${master_name}" \
for attempt in $(seq 1 ${retries}); do
if result=$(${gcloud} compute instances create "${master_name}" \
--project "${PROJECT}" \
--zone "${ZONE}" \
--machine-type "${MASTER_SIZE}" \
@ -118,7 +120,21 @@ function create-master-instance-internal() {
--disk "${disk}" \
--boot-disk-size "${MASTER_ROOT_DISK_SIZE}" \
${preemptible_master} \
${network}
${network} 2>&1); then
echo "${result}" >&2
return 0
else
echo "${result}" >&2
if [[ ! "${result}" =~ "try again later" ]]; then
echo "Failed to create master instance due to non-retryable error" >&2
return 1
fi
sleep 10
fi
done
echo "Failed to create master instance despite ${retries} attempts" >&2
return 1
}
function get-metadata() {