initial vendordata support

This adds the ability for a datasource to provide "vendordata".
The difference here is that vendordata is from the vendor (cloud provider)
where user-data is from the user.  By enabling this channel, the vendor
can have input on how the instance is set up without modifying or needing
to understand the user-data.

vendordata is generally consumed exactly like user-data, but the user
has the ability to disable its consumption.

The only datasource supporting this at the moment is SmartOS.

1 files changed, 103 insertions, 15 deletions

diff --git a/cloudinit/stages.py b/cloudinit/stages.py
index 07c55802..593b72a2 100644
--- a/cloudinit/stages.py
+++ b/cloudinit/stages.py
@@ -123,6 +123,7 @@ class Init(object):
             os.path.join(c_dir, 'scripts', 'per-instance'),
             os.path.join(c_dir, 'scripts', 'per-once'),
             os.path.join(c_dir, 'scripts', 'per-boot'),
+            os.path.join(c_dir, 'scripts', 'vendor'),
             os.path.join(c_dir, 'seed'),
             os.path.join(c_dir, 'instances'),
             os.path.join(c_dir, 'handlers'),
@@ -319,6 +320,7 @@ class Init(object):
         if not self._write_to_cache():
             return
         self._store_userdata()
+        self._store_vendordata()
 
     def _store_userdata(self):
         raw_ud = "%s" % (self.datasource.get_userdata_raw())
@@ -326,11 +328,20 @@ class Init(object):
         processed_ud = "%s" % (self.datasource.get_userdata())
         util.write_file(self._get_ipath('userdata'), processed_ud, 0600)
 
-    def _default_userdata_handlers(self):
-        opts = {
+    def _store_vendordata(self):
+        raw_vd = "%s" % (self.datasource.get_vendordata_raw())
+        util.write_file(self._get_ipath('vendordata_raw'), raw_vd, 0600)
+        processed_vd = "%s" % (self.datasource.get_vendordata())
+        util.write_file(self._get_ipath('vendordata'), processed_vd, 0600)
+
+    def _default_handlers(self, opts=None):
+        if opts is None:
+            opts = {}
+
+        opts.update({
             'paths': self.paths,
             'datasource': self.datasource,
-        }
+        })
         # TODO(harlowja) Hmmm, should we dynamically import these??
         def_handlers = [
             cc_part.CloudConfigPartHandler(**opts),
@@ -340,7 +351,23 @@ class Init(object):
         ]
         return def_handlers
 
-    def consume_userdata(self, frequency=PER_INSTANCE):
+    def _default_userdata_handlers(self):
+        return self._default_handlers()
+
+    def _default_vendordata_handlers(self):
+        return self._default_handlers(
+            opts={'script_path': 'vendor_scripts',
+                  'cloud_config_path': 'vendor_cloud_config'})
+
+    def _do_handlers(self, data_msg, c_handlers_list, frequency,
+                     excluded=None):
+        """
+        Generalized handlers suitable for use with either vendordata
+        or userdata
+        """
+        if excluded is None:
+            excluded = []
+
         cdir = self.paths.get_cpath("handlers")
         idir = self._get_ipath("handlers")
 
@@ -352,12 +379,6 @@ class Init(object):
             if d and d not in sys.path:
                 sys.path.insert(0, d)
 
-        # Ensure datasource fetched before activation (just incase)
-        user_data_msg = self.datasource.get_userdata(True)
-
-        # This keeps track of all the active handlers
-        c_handlers = helpers.ContentHandlers()
-
         def register_handlers_in_dir(path):
             # Attempts to register any handler modules under the given path.
             if not path or not os.path.isdir(path):
@@ -382,13 +403,16 @@ class Init(object):
                     util.logexc(LOG, "Failed to register handler from %s",
                                 fname)
 
+        # This keeps track of all the active handlers
+        c_handlers = helpers.ContentHandlers()
+
         # Add any handlers in the cloud-dir
         register_handlers_in_dir(cdir)
 
         # Register any other handlers that come from the default set. This
         # is done after the cloud-dir handlers so that the cdir modules can
         # take over the default user-data handler content-types.
-        for mod in self._default_userdata_handlers():
+        for mod in c_handlers_list:
             types = c_handlers.register(mod, overwrite=False)
             if types:
                 LOG.debug("Added default handler for %s from %s", types, mod)
@@ -406,7 +430,7 @@ class Init(object):
                 handlers.call_begin(mod, data, frequency)
                 c_handlers.initialized.append(mod)
 
-        def walk_handlers():
+        def walk_handlers(excluded):
             # Walk the user data
             part_data = {
                 'handlers': c_handlers,
@@ -419,9 +443,9 @@ class Init(object):
                 # to help write there contents to files with numbered
                 # names...
                 'handlercount': 0,
+                'excluded': excluded,
             }
-            handlers.walk(user_data_msg, handlers.walker_callback,
-                          data=part_data)
+            handlers.walk(data_msg, handlers.walker_callback, data=part_data)
 
         def finalize_handlers():
             # Give callbacks opportunity to finalize
@@ -438,10 +462,16 @@ class Init(object):
 
         try:
             init_handlers()
-            walk_handlers()
+            walk_handlers(excluded)
         finally:
             finalize_handlers()
 
+    def consume_data(self, frequency=PER_INSTANCE):
+        # Consume the userdata first, because we need want to let the part
+        # handlers run first (for merging stuff)
+        self._consume_userdata(frequency)
+        self._consume_vendordata(frequency)
+
         # Perform post-consumption adjustments so that
         # modules that run during the init stage reflect
         # this consumed set.
@@ -453,6 +483,64 @@ class Init(object):
         # objects before the load of the userdata happened,
         # this is expected.
 
+    def _consume_vendordata(self, frequency=PER_INSTANCE):
+        """
+        Consume the vendordata and run the part handlers on it
+        """
+        # User-data should have been consumed first.
+        # So we merge the other available cloud-configs (everything except
+        # vendor provided), and check whether or not we should consume
+        # vendor data at all. That gives user or system a chance to override.
+        if not self.datasource.get_vendordata_raw():
+            LOG.debug("no vendordata from datasource")
+            return
+
+        _cc_merger = helpers.ConfigMerger(paths=self._paths,
+                                          datasource=self.datasource,
+                                          additional_fns=[],
+                                          base_cfg=self.cfg,
+                                          include_vendor=False)
+        vdcfg = _cc_merger.cfg.get('vendor_data', {})
+
+        if not isinstance(vdcfg, dict):
+            vdcfg = {'enabled': False}
+            LOG.warn("invalid 'vendor_data' setting. resetting to: %s", vdcfg)
+
+        enabled = vdcfg.get('enabled')
+        no_handlers = vdcfg.get('disabled_handlers', None)
+
+        if not util.is_true(enabled):
+            LOG.debug("vendordata consumption is disabled.")
+            return
+
+        LOG.debug("vendor data will be consumed. disabled_handlers=%s",
+                  no_handlers)
+
+        # Ensure vendordata source fetched before activation (just incase)
+        vendor_data_msg = self.datasource.get_vendordata()
+
+        # This keeps track of all the active handlers, while excluding what the
+        # users doesn't want run, i.e. boot_hook, cloud_config, shell_script
+        c_handlers_list = self._default_vendordata_handlers()
+
+        # Run the handlers
+        self._do_handlers(vendor_data_msg, c_handlers_list, frequency,
+                          excluded=no_handlers)
+
+    def _consume_userdata(self, frequency=PER_INSTANCE):
+        """
+        Consume the userdata and run the part handlers
+        """
+
+        # Ensure datasource fetched before activation (just incase)
+        user_data_msg = self.datasource.get_userdata(True)
+
+        # This keeps track of all the active handlers
+        c_handlers_list = self._default_handlers()
+
+        # Run the handlers
+        self._do_handlers(user_data_msg, c_handlers_list, frequency)
+
 
 class Modules(object):
     def __init__(self, init, cfg_files=None):