finalize handlers even when another handler fails.

Instead of previously initializing and not finalizing the handles that
completed successfully when a handler initializing or running failed we
should attempt to always give said handlers a chance to finalize (even
when another handler fails).

LP: #1203368

1 files changed, 3 insertions, 4 deletions

diff --git a/cloudinit/handlers/__init__.py b/cloudinit/handlers/__init__.py
index 497d68c5..1d450061 100644
--- a/cloudinit/handlers/__init__.py
+++ b/cloudinit/handlers/__init__.py
@@ -151,10 +151,9 @@ def walker_handle_handler(pdata, _ctype, _filename, payload):
     try:
         mod = fixup_handler(importer.import_module(modname))
         call_begin(mod, pdata['data'], frequency)
-        # Only register and increment
-        # after the above have worked (so we don't if it
-        # fails)
-        handlers.register(mod)
+        # Only register and increment after the above have worked, so we don't
+        # register if it fails starting.
+        handlers.register(mod, initialized=True)
         pdata['handlercount'] = curcount + 1
     except:
         util.logexc(LOG, "Failed at registering python file: %s (part "