testing: adjust logged messages by framework

Respect semantics of "verbose" vs "debug", to make it possible to run tests in
debug mode but without verbose messages.

Also adjust "Starting", this message was not clear

slapos/testing/testcase.py

@@ -129,7 +129,7 @@ def makeModuleSetUpAndTestCaseClass(
   logger = logging.getLogger()
   console_handler = logging.StreamHandler()
   console_handler.setLevel(
-      logging.DEBUG if (verbose or debug) else logging.WARNING)
+      logging.DEBUG if verbose else logging.WARNING)
   logger.addHandler(console_handler)
 
   if debug:
@@ -414,7 +414,7 @@ def installSoftwareUrlList(cls, software_url_list, max_retry=10, debug=False):
       cls._copySnapshot(path, name)
 
   try:
-    cls.logger.debug("Starting")
+    cls.logger.debug("Starting SlapOS")
     cls.slap.start()
     for software_url in software_url_list:
       cls.logger.debug("Supplying %s", software_url)
@@ -570,7 +570,7 @@ class SlapOSInstanceTestCase(unittest.TestCase):
     snapshot_name = "{}.{}.setUpClass".format(cls.__module__, cls.__name__)
 
     try:
-      cls.logger.debug("Starting")
+      cls.logger.debug("Starting setUpClass %s", cls)
       cls.slap.start()
       cls.logger.debug(
           "Formatting to remove old partitions XXX should not be needed because we delete ..."
@@ -736,8 +736,16 @@ class SlapOSInstanceTestCase(unittest.TestCase):
       cls._storeSystemSnapshot(
           "{}._cleanup request destroy".format(snapshot_name))
     try:
+      # To make debug usable, we tolerate report_max_retry-1 errors and
+      # only debug the last.
       for _ in range(3):
-        cls.slap.waitForReport(max_retry=cls.report_max_retry, debug=cls._debug)
+        if cls._debug and cls.report_max_retry:
+          try:
+            cls.slap.waitForReport(max_retry=cls.report_max_retry - 1)
+          except SlapOSNodeCommandError:
+            cls.slap.waitForReport(debug=True)
+        else:
+          cls.slap.waitForReport(max_retry=cls.report_max_retry, debug=cls._debug)
     except:
       cls.logger.exception("Error during actual destruction")
       cls._storeSystemSnapshot(
@@ -768,8 +776,16 @@ class SlapOSInstanceTestCase(unittest.TestCase):
               "{}._cleanup leaked_partitions request destruction".format(
                   snapshot_name))
       try:
+        # To make debug usable, we tolerate report_max_retry-1 errors and
+        # only debug the last.
         for _ in range(3):
-          cls.slap.waitForReport(max_retry=cls.report_max_retry, debug=cls._debug)
+          if cls._debug and cls.report_max_retry:
+            try:
+              cls.slap.waitForReport(max_retry=cls.report_max_retry - 1)
+            except SlapOSNodeCommandError:
+              cls.slap.waitForReport(debug=True)
+          else:
+            cls.slap.waitForReport(max_retry=cls.report_max_retry, debug=cls._debug)
       except:
         cls.logger.exception(
             "Error during leaked partitions actual destruction")