feat: upload status db in testrun-files artifact

.github/workflows/regression_reusable.yaml

@@ -225,6 +225,7 @@ jobs:
             run_workdir/requirements_coverage.json
             run_workdir/monitor.log
             run_workdir/failure_analysis.md
+            run_workdir/cm-status-*.db*
       - name: ↟ Upload CLI coverage
         uses: actions/upload-artifact@v7
         if: success() || failure()

.github/workflows/upgrade_reusable.yaml

@@ -172,6 +172,7 @@ jobs:
             run_workdir/scheduling.log
             run_workdir/errors_all.log
             run_workdir/failure_analysis.md
+            run_workdir/cm-status-*.db*
       - name: ↟ Upload CLI coverage
         uses: actions/upload-artifact@v7
         if: success() || failure()

cardano_node_tests/cluster_management/status_db.py

@@ -22,7 +22,7 @@
 The current status can be inspected with the stock `sqlite3` CLI. The `overview` view
 combines all status records into one human-readable table:
 
-    db=/tmp/pytest-of-$USER/pytest-0/status.db
+    db=/tmp/pytest-of-$USER/pytest-0/cm-status.db
     sqlite3 -readonly -header "$db" 'SELECT * FROM overview ORDER BY instance_num, kind'
 
 The underlying tables (`test_running`, `resources`, `flags`) can be queried directly

runner/node_upgrade.sh

@@ -190,6 +190,9 @@ fi
 
 _last_cleanup
 
+# Copy cluster status databases to workdir
+./runner/status_dbs.sh "$ARTIFACTS_DIR" "$WORKDIR" || :
+
 # prepare artifacts for upload in GitHub Actions
 if [ -n "${GITHUB_ACTIONS:-}" ]; then
   # save testing artifacts

runner/regression.sh

@@ -434,6 +434,9 @@ fi
 # Create results archive
 ./runner/create_results.sh "$REPORTS_DIR" "$WORKDIR" || :
 
+# Copy cluster status databases to workdir
+./runner/status_dbs.sh "$ARTIFACTS_DIR" "$WORKDIR" || :
+
 # Save testing artifacts
 ./runner/save_artifacts.sh "$ARTIFACTS_DIR" "$WORKDIR" || :
 

runner/status_dbs.sh

@@ -0,0 +1,45 @@
+#!/usr/bin/env bash
+
+set -euo pipefail
+
+if [ "$#" -ne 2 ]; then
+  echo "Usage: $0 <artifacts_dir> <output_dir>" >&2
+  exit 1
+fi
+
+artifacts_dir="$1"
+output_dir="$2"
+
+mkdir -p "$output_dir" || { echo "Cannot create $output_dir" >&2; exit 1; }
+
+# The status database was copied from the pytest temp dir to the artifacts dir together
+# with the other testing artifacts. Copy the database of each pytest run to the output
+# dir. The output databases are numbered in the run order (e.g. one per node upgrade
+# step) - the `pytest-N` component of the artifacts subdir names cannot be used directly,
+# as the number comes from pytest's numbered basetemp counter that is shared by all
+# pytest invocations in a testrun, including those that don't collect artifacts.
+# Copy also the WAL sidecar, which is present when database connections were still open
+# during artifacts collection - without it the database copy would miss the
+# not-yet-checkpointed writes. The `-shm` file is skipped on purpose, it is transient
+# and not needed for an offline copy.
+found=0
+num=0
+while read -r db; do
+  # Skip the unexpanded pattern when the glob didn't match anything
+  [ -f "$db" ] || continue
+  num=$((num + 1))
+  cp "$db" "${output_dir}/cm-status-${num}.db" \
+    || { echo "Failed to copy $db" >&2; continue; }
+  found=1
+  if [ -e "${db}-wal" ]; then
+    cp "${db}-wal" "${output_dir}/cm-status-${num}.db-wal" \
+      || echo "Failed to copy ${db}-wal" >&2
+  fi
+done < <(printf '%s\n' "$artifacts_dir"/pytest-*/cm-status.db | sort -V)
+
+if [ "$found" -eq 0 ]; then
+  echo "No status database copied from $artifacts_dir" >&2
+  if [ -n "${GITHUB_ACTIONS:-}" ]; then
+    echo "::warning::No cluster status database was copied from the testing artifacts, none uploaded."
+  fi
+fi