From f8a9466b93de6ace15b65fdb651e0c151a56fe96 Mon Sep 17 00:00:00 2001
From: Swung0x48 <swung0x48@outlook.com>
Date: Wed, 7 Oct 2026 02:19:59 -0400
Subject: [PATCH] [Fix] (CI): retrace exit detection tolerates a single failed
 pidof, and artifact cleanup lists jobs in pages the API serves.

---
 .github/workflows/apk.yml         |  3 ++-
 .github/workflows/test.yml        |  3 ++-
 android-plugin/trace-replay-ci.sh | 12 ++++++++++--
 3 files changed, 14 insertions(+), 4 deletions(-)

diff --git a/.github/workflows/apk.yml b/.github/workflows/apk.yml
index 15289659..b6c32398 100644
--- a/.github/workflows/apk.yml
+++ b/.github/workflows/apk.yml
@@ -733,7 +733,8 @@ jobs:
           GH_TOKEN: ${{ github.token }}
         run: |
           # Fail before deletion if either paginated inventory cannot be read.
-          gh api --paginate "repos/${GITHUB_REPOSITORY}/actions/runs/${GITHUB_RUN_ID}/jobs?per_page=100" \
+          # 50 rows per page: the jobs endpoint answers HTTP 502 for 90+ rows on a re-run attempt of a run this size.
+          gh api --paginate "repos/${GITHUB_REPOSITORY}/actions/runs/${GITHUB_RUN_ID}/jobs?per_page=50" \
             --jq '.jobs[] | select(.name | startswith("retrace (")) | select(.conclusion == "failure" or .conclusion == "cancelled" or .conclusion == "timed_out" or .conclusion == "action_required") | .name' \
             > "${RUNNER_TEMP}/cleanup-jobs.txt"
           gh api --paginate "repos/${GITHUB_REPOSITORY}/actions/runs/${GITHUB_RUN_ID}/artifacts?per_page=100" \
diff --git a/.github/workflows/test.yml b/.github/workflows/test.yml
index a069e35b..7106e74e 100644
--- a/.github/workflows/test.yml
+++ b/.github/workflows/test.yml
@@ -3465,7 +3465,8 @@ jobs:
           # Read the complete inventories before deleting anything. A failed API
           # request must fail this step, not look like "no failed cases" through
           # an unchecked process substitution.
-          gh api --paginate "repos/${GITHUB_REPOSITORY}/actions/runs/${GITHUB_RUN_ID}/jobs?per_page=100" \
+          # 50 rows per page: the jobs endpoint answers HTTP 502 for 90+ rows on a re-run attempt of a run this size.
+          gh api --paginate "repos/${GITHUB_REPOSITORY}/actions/runs/${GITHUB_RUN_ID}/jobs?per_page=50" \
             --jq '.jobs[] | select((.name | startswith("retrace (")) or (.name | startswith("retrace verify (")) or (.name | startswith("retrace split ("))) | select(.conclusion == "failure" or .conclusion == "cancelled" or .conclusion == "timed_out" or .conclusion == "action_required") | .name' \
             > "${RUNNER_TEMP}/cleanup-jobs.txt"
           gh api --paginate "repos/${GITHUB_REPOSITORY}/actions/runs/${GITHUB_RUN_ID}/artifacts?per_page=100" \
diff --git a/android-plugin/trace-replay-ci.sh b/android-plugin/trace-replay-ci.sh
index 32b29875..f685da5b 100644
--- a/android-plugin/trace-replay-ci.sh
+++ b/android-plugin/trace-replay-ci.sh
@@ -511,6 +511,10 @@ run_retrace() {
 
   app_exited=0
   saw_app_process=0
+  # One failed pidof is not an exit: under memory pressure the adb shell call
+  # itself can fail while the app is alive (seen with the process still in
+  # pidof and the activity still initializing right after the "exit").
+  missed_app_polls=0
   poll_started="$(date +%s)"
   for _ in $(seq 1 "${timeout_seconds}"); do
     if adb_device_path shell run-as "${package_name}" ls "${app_dir}/output/result.json" >/dev/null 2>&1; then
@@ -518,6 +522,7 @@ run_retrace() {
     fi
     if adb_device_path shell pidof "${package_name}" >/dev/null 2>&1; then
       saw_app_process=1
+      missed_app_polls=0
     else
       poll_adb_state="$(adb_device_path get-state 2>/dev/null | tr -d '\r' || true)"
       if [ "${poll_adb_state}" != "device" ]; then
@@ -525,8 +530,11 @@ run_retrace() {
         break
       fi
       if [ "${saw_app_process}" -eq 1 ]; then
-        app_exited=1
-        break
+        missed_app_polls=$((missed_app_polls + 1))
+        if [ "${missed_app_polls}" -ge 3 ]; then
+          app_exited=1
+          break
+        fi
       fi
     fi
     sleep 1