Skip to content

ci: fix wireserver and imds test flakes #4828

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Behzad Mirkhanzadeh (behzad-mir) merged 4 commits into from
Sep 10, 2026
+14 −8
Merged

ci: fix wireserver and imds test flakes #4828

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
2323daf
ci: avoid conflicting pod name
QxBytes Aug 31, 2026
310f4e8
ci: add random for uniqueness across retries
QxBytes Sep 2, 2026
6df462c
increase timeout for pod startup
QxBytes Sep 3, 2026
4c01ab1
run wireserver tests only on linux nodes
QxBytes Sep 9, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
22 changes: 14 additions & 8 deletions test/network/wireserver_metadata_test.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,23 +1,29 @@
#!/bin/bash
# Optional per-pool parameterization: when POOL is set the throwaway pods are
# named "wget-${POOL}" and pinned to that agent pool via nodeSelector, so this
# test can run per-pool in parallel. When POOL is empty the behavior is
# unchanged (pod name "wget", no node selector).
# The two connectivity checks each run a throwaway busybox pod named
# "${POD_NAME}-wireserver-${SUFFIX}" / "${POD_NAME}-metadata-${SUFFIX}", where
# POD_NAME is "wget" (or "wget-${POOL}" when POOL is set) and SUFFIX is a random
# value for uniqueness across retries. When POOL is set the pods are also pinned
# to that agent pool. The pods always target Linux nodes (kubernetes.io/os=linux)
# so the Linux busybox image is never scheduled onto a Windows node, where it
# cannot be pulled/run and the check would time out.
POOL="${POOL:-}"
POD_NAME="wget"
OVERRIDES=()
NODE_SELECTOR='"kubernetes.io/os":"linux"'
if [ -n "$POOL" ]; then
POD_NAME="wget-${POOL}"
OVERRIDES=(--overrides="{\"spec\":{\"nodeSelector\":{\"agentpool\":\"${POOL}\"}}}")
NODE_SELECTOR="${NODE_SELECTOR},\"agentpool\":\"${POOL}\""
fi
OVERRIDES=(--overrides="{\"spec\":{\"nodeSelector\":{${NODE_SELECTOR}}}}")

kubectl run "$POD_NAME" -it --rm --image busybox --restart Never "${OVERRIDES[@]}" -- wget --timeout=3 --header=Metadata:true "http://168.63.129.16/machine/plugins?comp=nmagent&type=getinterfaceinfov1"
SUFFIX="${RANDOM}"
Comment thread
QxBytes marked this conversation as resolved.

kubectl run "$POD_NAME-wireserver-${SUFFIX}" -it --rm --image busybox --restart Never --pod-running-timeout=1m "${OVERRIDES[@]}" -- wget --timeout=3 --header=Metadata:true "http://168.63.129.16/machine/plugins?comp=nmagent&type=getinterfaceinfov1"
if [ $? -eq 0 ]; then
echo "wireserver connectivity expected to fail but succeeded"
exit 1
fi

kubectl run "$POD_NAME" -it --rm --image busybox --restart Never "${OVERRIDES[@]}" -- wget --timeout=3 --header=Metadata:true "http://169.254.169.254/metadata/instance?api-version=2021-02-01"
kubectl run "$POD_NAME-metadata-${SUFFIX}" -it --rm --image busybox --restart Never --pod-running-timeout=1m "${OVERRIDES[@]}" -- wget --timeout=3 --header=Metadata:true "http://169.254.169.254/metadata/instance?api-version=2021-02-01"
if [ $? -ne 0 ]; then
echo "metadata server connectivity expected to succeed but failed"
exit 1
Expand Down
Loading