From b2e118bf4fd93a31bdc173442b7e74b4ead9164f Mon Sep 17 00:00:00 2001
From: Logan Adams <loadams@microsoft.com>
Date: Thu, 13 Jul 2023 11:50:17 -0700
Subject: [PATCH] Update H100 workflow to open an issue if nightly CI fails

---
 .github/ISSUE_TEMPLATE/ci_failure_template.md | 10 ++++++++++
 .github/workflows/nv-h100.yml                 | 13 +++++++++++++
 2 files changed, 23 insertions(+)
 create mode 100644 .github/ISSUE_TEMPLATE/ci_failure_template.md

diff --git a/.github/ISSUE_TEMPLATE/ci_failure_template.md b/.github/ISSUE_TEMPLATE/ci_failure_template.md
new file mode 100644
index 00000000..27f07581
--- /dev/null
+++ b/.github/ISSUE_TEMPLATE/ci_failure_template.md
@@ -0,0 +1,10 @@
+---
+name: CI failure report
+about: Report a DeepSpeed CI failure
+title: "{{ env.TESTNAME }} CI test failure"
+labels: ci-failure
+assignees: ''
+
+---
+
+The Nightly CI for {{ env.TESTNAME }} failed.
diff --git a/.github/workflows/nv-h100.yml b/.github/workflows/nv-h100.yml
index 33f248c4..b19d8eca 100644
--- a/.github/workflows/nv-h100.yml
+++ b/.github/workflows/nv-h100.yml
@@ -8,6 +8,10 @@ concurrency:
   group: ${{ github.workflow }}-${{ github.ref }}
   cancel-in-progress: true
 
+permissions:
+  contents: read
+  issues: write
+
 jobs:
   unit-tests:
     runs-on: [self-hosted, nvidia, h100]
@@ -49,3 +53,12 @@ jobs:
           cd tests
           python -m pytest $PYTEST_OPTS -n 4  unit/ --torch_ver="2.0" --cuda_ver="12"
           python -m pytest $PYTEST_OPTS -m 'sequential' unit/ --torch_ver="2.0" --cuda_ver="12"
+
+      - name: Open GitHub issue if nightly CI fails
+        if: failure()
+        uses: JasonEtco/create-an-issue@v2
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          TESTNAME: nv-h100
+        with:
+          filename: .github/ISSUE_TEMPLATE/ci_failure_report.md
-- 
GitLab