From 79534c960724b728b8692cb8b5bd5e53703b5ff0 Mon Sep 17 00:00:00 2001
From: Micah Nagel <micah.nagel@defenseunicorns.com>
Date: Tue, 17 Dec 2024 08:59:40 -0700
Subject: [PATCH] chore: update infra ci to run weekly and on release pr
 (#1124)

## Description

Switches nightly CI to weekly, but also adds `milestoned` trigger to run
this CI pre-release. To date we have seen minimal issues with uds-core
in these environments that have actually been caused by core issues.
Reducing this CI to weekly + before release will still ensure that we
can catch issues, but also optimize our time/cost.

I also added a modification to the cluster wait script to wait on jobs.
Some of our recent CI has failed due to metallb not being installed
completely yet (it is installed via a rancher `helmchart` CR, which
creates a job). Example failure this should resolve:
https://github.com/defenseunicorns/uds-core/actions/runs/12327456734/job/34409486515

## Related Issue

Fixes https://github.com/defenseunicorns/uds-core/issues/1122

## Type of change

- [ ] Bug fix (non-breaking change which fixes an issue)
- [ ] New feature (non-breaking change which adds functionality)
- [x] Other (security config, docs update, etc)

## Checklist before merging

- [x] Test, docs, adr added or updated as needed
- [x] [Contributor
Guide](https://github.com/defenseunicorns/uds-template-capability/blob/main/CONTRIBUTING.md)
followed
---
 .github/workflows/test-aks.yaml  | 5 +++--
 .github/workflows/test-eks.yaml  | 5 +++--
 .github/workflows/test-rke2.yaml | 5 +++--
 tasks/iac.yaml                   | 2 +-
 4 files changed, 10 insertions(+), 7 deletions(-)

diff --git a/.github/workflows/test-aks.yaml b/.github/workflows/test-aks.yaml
index c7f6fa119..ff9a2dba7 100644
--- a/.github/workflows/test-aks.yaml
+++ b/.github/workflows/test-aks.yaml
@@ -1,12 +1,13 @@
 # Copyright 2024 Defense Unicorns
 # SPDX-License-Identifier: AGPL-3.0-or-later OR LicenseRef-Defense-Unicorns-Commercial
 
-name: AKS Nightly Test
+name: AKS Test
 
 on:
   schedule:
-    - cron: '0 6 * * *' # Runs at midnight Mountain every day
+    - cron: '0 0 * * 0' # Every Sunday Midnight (UTC) / Saturday 5pm MT
   pull_request:
+    types: [milestoned, opened, reopened, synchronize]
     paths:
       - tasks/iac.yaml
       - .github/bundles/aks/*
diff --git a/.github/workflows/test-eks.yaml b/.github/workflows/test-eks.yaml
index 338a1e299..c0b0cfe59 100644
--- a/.github/workflows/test-eks.yaml
+++ b/.github/workflows/test-eks.yaml
@@ -1,12 +1,13 @@
 # Copyright 2024 Defense Unicorns
 # SPDX-License-Identifier: AGPL-3.0-or-later OR LicenseRef-Defense-Unicorns-Commercial
 
-name: EKS Nightly Test
+name: EKS Test
 
 on:
   schedule:
-    - cron: '0 6 * * *' # Runs at midnight Mountain every day
+    - cron: '0 0 * * 0' # Every Sunday Midnight (UTC) / Saturday 5pm MT
   pull_request:
+    types: [milestoned, opened, reopened, synchronize]
     paths:
       - tasks/iac.yaml
       - .github/bundles/eks/*
diff --git a/.github/workflows/test-rke2.yaml b/.github/workflows/test-rke2.yaml
index db196031a..2187658f6 100644
--- a/.github/workflows/test-rke2.yaml
+++ b/.github/workflows/test-rke2.yaml
@@ -1,12 +1,13 @@
 # Copyright 2024 Defense Unicorns
 # SPDX-License-Identifier: AGPL-3.0-or-later OR LicenseRef-Defense-Unicorns-Commercial
 
-name: RKE2 Nightly Test
+name: RKE2 Test
 
 on:
   schedule:
-    - cron: '0 6 * * *' # Runs at midnight Mountain every day
+    - cron: '0 0 * * 0' # Every Sunday Midnight (UTC) / Saturday 5pm MT
   pull_request:
+    types: [milestoned, opened, reopened, synchronize]
     paths:
       - tasks/iac.yaml
       - .github/bundles/rke2/*
diff --git a/tasks/iac.yaml b/tasks/iac.yaml
index cadf35fc7..7eec9f854 100644
--- a/tasks/iac.yaml
+++ b/tasks/iac.yaml
@@ -115,7 +115,7 @@ tasks:
 
           # wait for cluster components
           while true; do
-            if [ $(uds zarf tools kubectl get po -A --no-headers=true | egrep -v 'Running|Completed' | wc -l) -gt 0 ]; then
+            if [ $(uds zarf tools kubectl get po,job -A --no-headers=true | egrep -v 'Running|Complete' | wc -l) -gt 0 ]; then
               echo "Waiting for cluster components to be ready...";
               sleep 5;
             else