Better failure reporting

.buildkite/report.sh

@@ -7,14 +7,15 @@ reports_dir="$tests_repo/.reports"
 
 get_coverage() {
   if [ ! -e "$tests_repo/.cli_coverage" ] || ! hash cardano-cli; then
-    return
+    return 1
   fi
   oldpwd="$PWD"
   cd "$tests_repo"
+  retval=0
   PYTHONPATH="$PWD:$PYTHONPATH" cardano_node_tests/cardano_cli_coverage.py \
-    -i .cli_coverage/cli_coverage_* \
-    -o "$1"
+    -i .cli_coverage/cli_coverage_* -o "$1" || retval=1
   cd "$oldpwd"
+  return "$retval"
 }
 
 allure_report_dir="$tests_repo/allure-report"
@@ -27,4 +28,4 @@ allure generate "$reports_dir" -o "$allure_report_dir" --clean
 echo "Creating report archive $tests_repo/allure-report.tar.bz2"
 tar -C "$tests_repo" -cjf "$tests_repo/allure-report.tar.bz2" allure-report
 echo "Generating CLI coverage report to $tests_repo/cli_coverage.json"
-get_coverage "$tests_repo/cli_coverage.json"
+get_coverage "$tests_repo/cli_coverage.json" || : > "$tests_repo/cli_coverage.json"

cardano_node_tests/cardano_cli_coverage.py

@@ -228,7 +228,7 @@ def main() -> int:
         "cardano-cli": get_available_commands(["cardano-cli"], ignore_skips=args.ignore_skips)
     }
     try:
-        coverage = get_coverage(args.input_files, available_commands)
+        coverage = get_coverage(input_jsons=args.input_files, available_commands=available_commands)
     except AttributeError as exc:
         LOGGER.error(str(exc))
         return 1

cardano_node_tests/utils/cluster_management.py

@@ -395,14 +395,14 @@ def _restart(self, start_cmd: str = "", stop_cmd: str = "") -> bool:  # noqa: C9
             excp = None
             if i > 0:
                 self.cm._locked_log(
-                    f"c{self.cm.cluster_instance}: failed to start cluster, retrying"
+                    f"c{self.cm.cluster_instance}: failed to start cluster:\n{excp}\nretrying"
                 )
                 time.sleep(0.2)
 
             try:
                 cluster_nodes.stop_cluster(cmd=str(startup_files.stop_script))
-            except Exception:
-                pass
+            except Exception as err:
+                self.cm._locked_log(f"c{self.cm.cluster_instance}: failed to stop cluster:\n{err}")
 
             self._restart_save_cluster_artifacts(clean=True)
             try: