openstack: Allow discovery in init-local using dhclient in a sandbox.

Network has not yet been configured in the init-local stage so the
openstack datasource will use dhcp-client to temporarily obtain an ipv4
address and query the metadata service at http://169.254.169.254 to get
network_data.json configuration. If present, the datasource will return
network_config version 1 config based on that network_data.json content.
Previously OpenStack datasource only setup dhcp on the fallback interface
so this represents a change in behavior to react to the full config
provided by openstack.

Also significant to OpenStack is the separation of a _crawl_data operation
from get_data(). crawl_data walks the available metadata services and
returns a dict of discovered content. get_data consumes the crawled_data,
 caches it in the datasource and reacts to that data.
/run/cloud-init/instance-data.json now published network_data.json or
ec2_metadata key if that data is present on any datasource.

The main reasons for the separation of crawl from get_data:
 * Enable performance metrics of cloud-init's metadata crawls on each
 * Enable cloud-init modules and scripts to query and consume metadata
   content which may have updated/changed after cloud-init's initial cache
   during instance boot. (Think hotplug)

Also generalize common logic to base DataSource class/module:
 * Move to a common UNSET variable up into base datasource module fix EC2,
   ConfigDrive, OpenStack, SmartOS to use the global.
 * Drop get_url_settings from Ec2, CloudStack and OpenStack and generalize
   DataSource.get_url_params(). Allow subclasses to override url_max_wait,
   url_timeout and url_retries params.
 * Rename get_network_metadata bool to perform_dhcp_setup as it designates
   whether EphemeralDHCPv4 setup is required before crawling metadata.

LP: #1749717

7 files changed, 299 insertions, 119 deletions

diff --git a/cloudinit/sources/DataSourceCloudStack.py b/cloudinit/sources/DataSourceCloudStack.py
index 0df545fc..d4b758f2 100644
--- a/cloudinit/sources/DataSourceCloudStack.py
+++ b/cloudinit/sources/DataSourceCloudStack.py
@@ -68,6 +68,10 @@ class DataSourceCloudStack(sources.DataSource):
 
     dsname = 'CloudStack'
 
+    # Setup read_url parameters per get_url_params.
+    url_max_wait = 120
+    url_timeout = 50
+
     def __init__(self, sys_cfg, distro, paths):
         sources.DataSource.__init__(self, sys_cfg, distro, paths)
         self.seed_dir = os.path.join(paths.seed_dir, 'cs')
@@ -80,33 +84,18 @@ class DataSourceCloudStack(sources.DataSource):
         self.metadata_address = "http://%s/" % (self.vr_addr,)
         self.cfg = {}
 
-    def _get_url_settings(self):
-        mcfg = self.ds_cfg
-        max_wait = 120
-        try:
-            max_wait = int(mcfg.get("max_wait", max_wait))
-        except Exception:
-            util.logexc(LOG, "Failed to get max wait. using %s", max_wait)
+    def wait_for_metadata_service(self):
+        url_params = self.get_url_params()
 
-        if max_wait == 0:
+        if url_params.max_wait_seconds <= 0:
             return False
 
-        timeout = 50
-        try:
-            timeout = int(mcfg.get("timeout", timeout))
-        except Exception:
-            util.logexc(LOG, "Failed to get timeout, using %s", timeout)
-
-        return (max_wait, timeout)
-
-    def wait_for_metadata_service(self):
-        (max_wait, timeout) = self._get_url_settings()
-
         urls = [uhelp.combine_url(self.metadata_address,
                                   'latest/meta-data/instance-id')]
         start_time = time.time()
-        url = uhelp.wait_for_url(urls=urls, max_wait=max_wait,
-                                 timeout=timeout, status_cb=LOG.warn)
+        url = uhelp.wait_for_url(
+            urls=urls, max_wait=url_params.max_wait_seconds,
+            timeout=url_params.timeout_seconds, status_cb=LOG.warn)
 
         if url:
             LOG.debug("Using metadata source: '%s'", url)
diff --git a/cloudinit/sources/DataSourceConfigDrive.py b/cloudinit/sources/DataSourceConfigDrive.py
index 121cf215..4cb28977 100644
--- a/cloudinit/sources/DataSourceConfigDrive.py
+++ b/cloudinit/sources/DataSourceConfigDrive.py
@@ -43,7 +43,7 @@ class DataSourceConfigDrive(openstack.SourceMixin, sources.DataSource):
         self.version = None
         self.ec2_metadata = None
         self._network_config = None
-        self.network_json = None
+        self.network_json = sources.UNSET
         self.network_eni = None
         self.known_macs = None
         self.files = {}
@@ -149,7 +149,7 @@ class DataSourceConfigDrive(openstack.SourceMixin, sources.DataSource):
     @property
     def network_config(self):
         if self._network_config is None:
-            if self.network_json is not None:
+            if self.network_json not in (None, sources.UNSET):
                 LOG.debug("network config provided via network_json")
                 self._network_config = openstack.convert_net_json(
                     self.network_json, known_macs=self.known_macs)
diff --git a/cloudinit/sources/DataSourceEc2.py b/cloudinit/sources/DataSourceEc2.py
index 21e9ef84..968ab3f7 100644
--- a/cloudinit/sources/DataSourceEc2.py
+++ b/cloudinit/sources/DataSourceEc2.py
@@ -27,8 +27,6 @@ SKIP_METADATA_URL_CODES = frozenset([uhelp.NOT_FOUND])
 STRICT_ID_PATH = ("datasource", "Ec2", "strict_id")
 STRICT_ID_DEFAULT = "warn"
 
-_unset = "_unset"
-
 
 class Platforms(object):
     # TODO Rename and move to cloudinit.cloud.CloudNames
@@ -59,15 +57,16 @@ class DataSourceEc2(sources.DataSource):
     # for extended metadata content. IPv6 support comes in 2016-09-02
     extended_metadata_versions = ['2016-09-02']
 
+    # Setup read_url parameters per get_url_params.
+    url_max_wait = 120
+    url_timeout = 50
+
     _cloud_platform = None
 
-    _network_config = _unset  # Used for caching calculated network config v1
+    _network_config = sources.UNSET  # Used to cache calculated network cfg v1
 
     # Whether we want to get network configuration from the metadata service.
-    get_network_metadata = False
-
-    # Track the discovered fallback nic for use in configuration generation.
-    _fallback_interface = None
+    perform_dhcp_setup = False
 
     def __init__(self, sys_cfg, distro, paths):
         super(DataSourceEc2, self).__init__(sys_cfg, distro, paths)
@@ -98,7 +97,7 @@ class DataSourceEc2(sources.DataSource):
         elif self.cloud_platform == Platforms.NO_EC2_METADATA:
             return False
 
-        if self.get_network_metadata:  # Setup networking in init-local stage.
+        if self.perform_dhcp_setup:  # Setup networking in init-local stage.
             if util.is_FreeBSD():
                 LOG.debug("FreeBSD doesn't support running dhclient with -sf")
                 return False
@@ -158,27 +157,11 @@ class DataSourceEc2(sources.DataSource):
         else:
             return self.metadata['instance-id']
 
-    def _get_url_settings(self):
-        mcfg = self.ds_cfg
-        max_wait = 120
-        try:
-            max_wait = int(mcfg.get("max_wait", max_wait))
-        except Exception:
-            util.logexc(LOG, "Failed to get max wait. using %s", max_wait)
-
-        timeout = 50
-        try:
-            timeout = max(0, int(mcfg.get("timeout", timeout)))
-        except Exception:
-            util.logexc(LOG, "Failed to get timeout, using %s", timeout)
-
-        return (max_wait, timeout)
-
     def wait_for_metadata_service(self):
         mcfg = self.ds_cfg
 
-        (max_wait, timeout) = self._get_url_settings()
-        if max_wait <= 0:
+        url_params = self.get_url_params()
+        if url_params.max_wait_seconds <= 0:
             return False
 
         # Remove addresses from the list that wont resolve.
@@ -205,7 +188,8 @@ class DataSourceEc2(sources.DataSource):
 
         start_time = time.time()
         url = uhelp.wait_for_url(
-            urls=urls, max_wait=max_wait, timeout=timeout, status_cb=LOG.warn)
+            urls=urls, max_wait=url_params.max_wait_seconds,
+            timeout=url_params.timeout_seconds, status_cb=LOG.warn)
 
         if url:
             self.metadata_address = url2base[url]
@@ -310,11 +294,11 @@ class DataSourceEc2(sources.DataSource):
     @property
     def network_config(self):
         """Return a network config dict for rendering ENI or netplan files."""
-        if self._network_config != _unset:
+        if self._network_config != sources.UNSET:
             return self._network_config
 
         if self.metadata is None:
-            # this would happen if get_data hadn't been called. leave as _unset
+            # this would happen if get_data hadn't been called. leave as UNSET
             LOG.warning(
                 "Unexpected call to network_config when metadata is None.")
             return None
@@ -353,9 +337,7 @@ class DataSourceEc2(sources.DataSource):
                 self._fallback_interface = _legacy_fbnic
                 self.fallback_nic = None
             else:
-                self._fallback_interface = net.find_fallback_nic()
-                if self._fallback_interface is None:
-                    LOG.warning("Did not find a fallback interface on EC2.")
+                return super(DataSourceEc2, self).fallback_interface
         return self._fallback_interface
 
     def _crawl_metadata(self):
@@ -390,7 +372,7 @@ class DataSourceEc2Local(DataSourceEc2):
     metadata service. If the metadata service provides network configuration
     then render the network configuration for that instance based on metadata.
     """
-    get_network_metadata = True  # Get metadata network config if present
+    perform_dhcp_setup = True  # Use dhcp before querying metadata
 
     def get_data(self):
         supported_platforms = (Platforms.AWS,)
diff --git a/cloudinit/sources/DataSourceOpenStack.py b/cloudinit/sources/DataSourceOpenStack.py
index fb166ae1..1a12a3f1 100644
--- a/cloudinit/sources/DataSourceOpenStack.py
+++ b/cloudinit/sources/DataSourceOpenStack.py
@@ -7,6 +7,7 @@
 import time
 
 from cloudinit import log as logging
+from cloudinit.net.dhcp import EphemeralDHCPv4, NoDHCPLeaseError
 from cloudinit import sources
 from cloudinit import url_helper
 from cloudinit import util
@@ -27,46 +28,25 @@ class DataSourceOpenStack(openstack.SourceMixin, sources.DataSource):
 
     dsname = "OpenStack"
 
+    _network_config = sources.UNSET  # Used to cache calculated network cfg v1
+
+    # Whether we want to get network configuration from the metadata service.
+    perform_dhcp_setup = False
+
     def __init__(self, sys_cfg, distro, paths):
         super(DataSourceOpenStack, self).__init__(sys_cfg, distro, paths)
         self.metadata_address = None
         self.ssl_details = util.fetch_ssl_details(self.paths)
         self.version = None
         self.files = {}
-        self.ec2_metadata = None
+        self.ec2_metadata = sources.UNSET
+        self.network_json = sources.UNSET
 
     def __str__(self):
         root = sources.DataSource.__str__(self)
         mstr = "%s [%s,ver=%s]" % (root, self.dsmode, self.version)
         return mstr
 
-    def _get_url_settings(self):
-        # TODO(harlowja): this is shared with ec2 datasource, we should just
-        # move it to a shared location instead...
-        # Note: the defaults here are different though.
-
-        # max_wait < 0 indicates do not wait
-        max_wait = -1
-        timeout = 10
-        retries = 5
-
-        try:
-            max_wait = int(self.ds_cfg.get("max_wait", max_wait))
-        except Exception:
-            util.logexc(LOG, "Failed to get max wait. using %s", max_wait)
-
-        try:
-            timeout = max(0, int(self.ds_cfg.get("timeout", timeout)))
-        except Exception:
-            util.logexc(LOG, "Failed to get timeout, using %s", timeout)
-
-        try:
-            retries = int(self.ds_cfg.get("retries", retries))
-        except Exception:
-            util.logexc(LOG, "Failed to get retries. using %s", retries)
-
-        return (max_wait, timeout, retries)
-
     def wait_for_metadata_service(self):
         urls = self.ds_cfg.get("metadata_urls", [DEF_MD_URL])
         filtered = [x for x in urls if util.is_resolvable_url(x)]
@@ -86,10 +66,11 @@ class DataSourceOpenStack(openstack.SourceMixin, sources.DataSource):
             md_urls.append(md_url)
             url2base[md_url] = url
 
-        (max_wait, timeout, _retries) = self._get_url_settings()
+        url_params = self.get_url_params()
         start_time = time.time()
-        avail_url = url_helper.wait_for_url(urls=md_urls, max_wait=max_wait,
-                                            timeout=timeout)
+        avail_url = url_helper.wait_for_url(
+            urls=md_urls, max_wait=url_params.max_wait_seconds,
+            timeout=url_params.timeout_seconds)
         if avail_url:
             LOG.debug("Using metadata source: '%s'", url2base[avail_url])
         else:
@@ -99,38 +80,64 @@ class DataSourceOpenStack(openstack.SourceMixin, sources.DataSource):
         self.metadata_address = url2base.get(avail_url)
         return bool(avail_url)
 
-    def _get_data(self):
-        try:
-            if not self.wait_for_metadata_service():
-                return False
-        except IOError:
-            return False
+    def check_instance_id(self, sys_cfg):
+        # quickly (local check only) if self.instance_id is still valid
+        return sources.instance_id_matches_system_uuid(self.get_instance_id())
 
-        (_max_wait, timeout, retries) = self._get_url_settings()
+    @property
+    def network_config(self):
+        """Return a network config dict for rendering ENI or netplan files."""
+        if self._network_config != sources.UNSET:
+            return self._network_config
+
+        # RELEASE_BLOCKER: SRU to Xenial and Artful SRU should not provide
+        # network_config by default unless configured in /etc/cloud/cloud.cfg*.
+        # Patch Xenial and Artful before release to default to False.
+        if util.is_false(self.ds_cfg.get('apply_network_config', True)):
+            self._network_config = None
+            return self._network_config
+        if self.network_json == sources.UNSET:
+            # this would happen if get_data hadn't been called. leave as UNSET
+            LOG.warning(
+                'Unexpected call to network_config when network_json is None.')
+            return None
+
+        LOG.debug('network config provided via network_json')
+        self._network_config = openstack.convert_net_json(
+            self.network_json, known_macs=None)
+        return self._network_config
 
-        try:
-            results = util.log_time(LOG.debug,
-                                    'Crawl of openstack metadata service',
-                                    read_metadata_service,
-                                    args=[self.metadata_address],
-                                    kwargs={'ssl_details': self.ssl_details,
-                                            'retries': retries,
-                                            'timeout': timeout})
-        except openstack.NonReadable:
-            return False
-        except (openstack.BrokenMetadata, IOError):
-            util.logexc(LOG, "Broken metadata address %s",
-                        self.metadata_address)
-            return False
+    def _get_data(self):
+        """Crawl metadata, parse and persist that data for this instance.
+
+        @return: True when metadata discovered indicates OpenStack datasource.
+            False when unable to contact metadata service or when metadata
+            format is invalid or disabled.
+        """
+        if self.perform_dhcp_setup:  # Setup networking in init-local stage.
+            try:
+                with EphemeralDHCPv4(self.fallback_interface):
+                    results = util.log_time(
+                        logfunc=LOG.debug, msg='Crawl of metadata service',
+                        func=self._crawl_metadata)
+            except (NoDHCPLeaseError, sources.InvalidMetaDataException) as e:
+                util.logexc(LOG, str(e))
+                return False
+        else:
+            try:
+                results = self._crawl_metadata()
+            except sources.InvalidMetaDataException as e:
+                util.logexc(LOG, str(e))
+                return False
 
         self.dsmode = self._determine_dsmode([results.get('dsmode')])
         if self.dsmode == sources.DSMODE_DISABLED:
             return False
-
         md = results.get('metadata', {})
         md = util.mergemanydict([md, DEFAULT_METADATA])
         self.metadata = md
         self.ec2_metadata = results.get('ec2-metadata')
+        self.network_json = results.get('networkdata')
         self.userdata_raw = results.get('userdata')
         self.version = results['version']
         self.files.update(results.get('files', {}))
@@ -145,9 +152,50 @@ class DataSourceOpenStack(openstack.SourceMixin, sources.DataSource):
 
         return True
 
-    def check_instance_id(self, sys_cfg):
-        # quickly (local check only) if self.instance_id is still valid
-        return sources.instance_id_matches_system_uuid(self.get_instance_id())
+    def _crawl_metadata(self):
+        """Crawl metadata service when available.
+
+        @returns: Dictionary with all metadata discovered for this datasource.
+        @raise: InvalidMetaDataException on unreadable or broken
+            metadata.
+        """
+        try:
+            if not self.wait_for_metadata_service():
+                raise sources.InvalidMetaDataException(
+                    'No active metadata service found')
+        except IOError as e:
+            raise sources.InvalidMetaDataException(
+                'IOError contacting metadata service: {error}'.format(
+                    error=str(e)))
+
+        url_params = self.get_url_params()
+
+        try:
+            result = util.log_time(
+                LOG.debug, 'Crawl of openstack metadata service',
+                read_metadata_service, args=[self.metadata_address],
+                kwargs={'ssl_details': self.ssl_details,
+                        'retries': url_params.num_retries,
+                        'timeout': url_params.timeout_seconds})
+        except openstack.NonReadable as e:
+            raise sources.InvalidMetaDataException(str(e))
+        except (openstack.BrokenMetadata, IOError):
+            msg = 'Broken metadata address {addr}'.format(
+                addr=self.metadata_address)
+            raise sources.InvalidMetaDataException(msg)
+        return result
+
+
+class DataSourceOpenStackLocal(DataSourceOpenStack):
+    """Run in init-local using a dhcp discovery prior to metadata crawl.
+
+    In init-local, no network is available. This subclass sets up minimal
+    networking with dhclient on a viable nic so that it can talk to the
+    metadata service. If the metadata service provides network configuration
+    then render the network configuration for that instance based on metadata.
+    """
+
+    perform_dhcp_setup = True  # Get metadata network config if present
 
 
 def read_metadata_service(base_url, ssl_details=None,
@@ -159,6 +207,7 @@ def read_metadata_service(base_url, ssl_details=None,
 
 # Used to match classes to dependencies
 datasources = [
+    (DataSourceOpenStackLocal, (sources.DEP_FILESYSTEM,)),
     (DataSourceOpenStack, (sources.DEP_FILESYSTEM, sources.DEP_NETWORK)),
 ]
 
diff --git a/cloudinit/sources/DataSourceSmartOS.py b/cloudinit/sources/DataSourceSmartOS.py
index fcb46b14..c91e4d59 100644
--- a/cloudinit/sources/DataSourceSmartOS.py
+++ b/cloudinit/sources/DataSourceSmartOS.py
@@ -165,9 +165,8 @@ class DataSourceSmartOS(sources.DataSource):
 
     dsname = "Joyent"
 
-    _unset = "_unset"
-    smartos_type = _unset
-    md_client = _unset
+    smartos_type = sources.UNSET
+    md_client = sources.UNSET
 
     def __init__(self, sys_cfg, distro, paths):
         sources.DataSource.__init__(self, sys_cfg, distro, paths)
@@ -189,12 +188,12 @@ class DataSourceSmartOS(sources.DataSource):
         return "%s [client=%s]" % (root, self.md_client)
 
     def _init(self):
-        if self.smartos_type == self._unset:
+        if self.smartos_type == sources.UNSET:
             self.smartos_type = get_smartos_environ()
             if self.smartos_type is None:
                 self.md_client = None
 
-        if self.md_client == self._unset:
+        if self.md_client == sources.UNSET:
             self.md_client = jmc_client_factory(
                 smartos_type=self.smartos_type,
                 metadata_sockfile=self.ds_cfg['metadata_sockfile'],
diff --git a/cloudinit/sources/__init__.py b/cloudinit/sources/__init__.py
index df0b374a..90d74575 100644
--- a/cloudinit/sources/__init__.py
+++ b/cloudinit/sources/__init__.py
@@ -9,6 +9,7 @@
 # This file is part of cloud-init. See LICENSE file for license information.
 
 import abc
+from collections import namedtuple
 import copy
 import json
 import os
@@ -17,6 +18,7 @@ import six
 from cloudinit.atomic_helper import write_json
 from cloudinit import importer
 from cloudinit import log as logging
+from cloudinit import net
 from cloudinit import type_utils
 from cloudinit import user_data as ud
 from cloudinit import util
@@ -41,6 +43,8 @@ INSTANCE_JSON_FILE = 'instance-data.json'
 # Key which can be provide a cloud's official product name to cloud-init
 METADATA_CLOUD_NAME_KEY = 'cloud-name'
 
+UNSET = "_unset"
+
 LOG = logging.getLogger(__name__)
 
 
@@ -48,6 +52,11 @@ class DataSourceNotFoundException(Exception):
     pass
 
 
+class InvalidMetaDataException(Exception):
+    """Raised when metadata is broken, unavailable or disabled."""
+    pass
+
+
 def process_base64_metadata(metadata, key_path=''):
     """Strip ci-b64 prefix and return metadata with base64-encoded-keys set."""
     md_copy = copy.deepcopy(metadata)
@@ -68,6 +77,10 @@ def process_base64_metadata(metadata, key_path=''):
     return md_copy
 
 
+URLParams = namedtuple(
+    'URLParms', ['max_wait_seconds', 'timeout_seconds', 'num_retries'])
+
+
 @six.add_metaclass(abc.ABCMeta)
 class DataSource(object):
 
@@ -81,6 +94,14 @@ class DataSource(object):
     # Cached cloud_name as determined by _get_cloud_name
     _cloud_name = None
 
+    # Track the discovered fallback nic for use in configuration generation.
+    _fallback_interface = None
+
+    # read_url_params
+    url_max_wait = -1   # max_wait < 0 means do not wait
+    url_timeout = 10    # timeout for each metadata url read attempt
+    url_retries = 5     # number of times to retry url upon 404
+
     def __init__(self, sys_cfg, distro, paths, ud_proc=None):
         self.sys_cfg = sys_cfg
         self.distro = distro
@@ -128,6 +149,14 @@ class DataSource(object):
                 'meta-data': self.metadata,
                 'user-data': self.get_userdata_raw(),
                 'vendor-data': self.get_vendordata_raw()}}
+        if hasattr(self, 'network_json'):
+            network_json = getattr(self, 'network_json')
+            if network_json != UNSET:
+                instance_data['ds']['network_json'] = network_json
+        if hasattr(self, 'ec2_metadata'):
+            ec2_metadata = getattr(self, 'ec2_metadata')
+            if ec2_metadata != UNSET:
+                instance_data['ds']['ec2_metadata'] = ec2_metadata
         instance_data.update(
             self._get_standardized_metadata())
         try:
@@ -149,6 +178,42 @@ class DataSource(object):
             'Subclasses of DataSource must implement _get_data which'
             ' sets self.metadata, vendordata_raw and userdata_raw.')
 
+    def get_url_params(self):
+        """Return the Datasource's prefered url_read parameters.
+
+        Subclasses may override url_max_wait, url_timeout, url_retries.
+
+        @return: A URLParams object with max_wait_seconds, timeout_seconds,
+            num_retries.
+        """
+        max_wait = self.url_max_wait
+        try:
+            max_wait = int(self.ds_cfg.get("max_wait", self.url_max_wait))
+        except ValueError:
+            util.logexc(
+                LOG, "Config max_wait '%s' is not an int, using default '%s'",
+                self.ds_cfg.get("max_wait"), max_wait)
+
+        timeout = self.url_timeout
+        try:
+            timeout = max(
+                0, int(self.ds_cfg.get("timeout", self.url_timeout)))
+        except ValueError:
+            timeout = self.url_timeout
+            util.logexc(
+                LOG, "Config timeout '%s' is not an int, using default '%s'",
+                self.ds_cfg.get('timeout'), timeout)
+
+        retries = self.url_retries
+        try:
+            retries = int(self.ds_cfg.get("retries", self.url_retries))
+        except Exception:
+            util.logexc(
+                LOG, "Config retries '%s' is not an int, using default '%s'",
+                self.ds_cfg.get('retries'), retries)
+
+        return URLParams(max_wait, timeout, retries)
+
     def get_userdata(self, apply_filter=False):
         if self.userdata is None:
             self.userdata = self.ud_proc.process(self.get_userdata_raw())
@@ -162,6 +227,17 @@ class DataSource(object):
         return self.vendordata
 
     @property
+    def fallback_interface(self):
+        """Determine the network interface used during local network config."""
+        if self._fallback_interface is None:
+            self._fallback_interface = net.find_fallback_nic()
+            if self._fallback_interface is None:
+                LOG.warning(
+                    "Did not find a fallback interface on %s.",
+                    self.cloud_name)
+        return self._fallback_interface
+
+    @property
     def cloud_name(self):
         """Return lowercase cloud name as determined by the datasource.
 
diff --git a/cloudinit/sources/tests/test_init.py b/cloudinit/sources/tests/test_init.py
index 452e9219..d5bc98a4 100644
--- a/cloudinit/sources/tests/test_init.py
+++ b/cloudinit/sources/tests/test_init.py
@@ -17,6 +17,7 @@ from cloudinit import util
 class DataSourceTestSubclassNet(DataSource):
 
     dsname = 'MyTestSubclass'
+    url_max_wait = 55
 
     def __init__(self, sys_cfg, distro, paths, custom_userdata=None):
         super(DataSourceTestSubclassNet, self).__init__(
@@ -70,8 +71,7 @@ class TestDataSource(CiTestCase):
         """Init uses DataSource.dsname for sourcing ds_cfg."""
         sys_cfg = {'datasource': {'MyTestSubclass': {'key2': False}}}
         distro = 'distrotest'  # generally should be a Distro object
-        paths = Paths({})
-        datasource = DataSourceTestSubclassNet(sys_cfg, distro, paths)
+        datasource = DataSourceTestSubclassNet(sys_cfg, distro, self.paths)
         self.assertEqual({'key2': False}, datasource.ds_cfg)
 
     def test_str_is_classname(self):
@@ -81,6 +81,91 @@ class TestDataSource(CiTestCase):
             'DataSourceTestSubclassNet',
             str(DataSourceTestSubclassNet('', '', self.paths)))
 
+    def test_datasource_get_url_params_defaults(self):
+        """get_url_params default url config settings for the datasource."""
+        params = self.datasource.get_url_params()
+        self.assertEqual(params.max_wait_seconds, self.datasource.url_max_wait)
+        self.assertEqual(params.timeout_seconds, self.datasource.url_timeout)
+        self.assertEqual(params.num_retries, self.datasource.url_retries)
+
+    def test_datasource_get_url_params_subclassed(self):
+        """Subclasses can override get_url_params defaults."""
+        sys_cfg = {'datasource': {'MyTestSubclass': {'key2': False}}}
+        distro = 'distrotest'  # generally should be a Distro object
+        datasource = DataSourceTestSubclassNet(sys_cfg, distro, self.paths)
+        expected = (datasource.url_max_wait, datasource.url_timeout,
+                    datasource.url_retries)
+        url_params = datasource.get_url_params()
+        self.assertNotEqual(self.datasource.get_url_params(), url_params)
+        self.assertEqual(expected, url_params)
+
+    def test_datasource_get_url_params_ds_config_override(self):
+        """Datasource configuration options can override url param defaults."""
+        sys_cfg = {
+            'datasource': {
+                'MyTestSubclass': {
+                    'max_wait': '1', 'timeout': '2', 'retries': '3'}}}
+        datasource = DataSourceTestSubclassNet(
+            sys_cfg, self.distro, self.paths)
+        expected = (1, 2, 3)
+        url_params = datasource.get_url_params()
+        self.assertNotEqual(
+            (datasource.url_max_wait, datasource.url_timeout,
+             datasource.url_retries),
+            url_params)
+        self.assertEqual(expected, url_params)
+
+    def test_datasource_get_url_params_is_zero_or_greater(self):
+        """get_url_params ignores timeouts with a value below 0."""
+        # Set an override that is below 0 which gets ignored.
+        sys_cfg = {'datasource': {'_undef': {'timeout': '-1'}}}
+        datasource = DataSource(sys_cfg, self.distro, self.paths)
+        (_max_wait, timeout, _retries) = datasource.get_url_params()
+        self.assertEqual(0, timeout)
+
+    def test_datasource_get_url_uses_defaults_on_errors(self):
+        """On invalid system config values for url_params defaults are used."""
+        # All invalid values should be logged
+        sys_cfg = {'datasource': {
+            '_undef': {
+                'max_wait': 'nope', 'timeout': 'bug', 'retries': 'nonint'}}}
+        datasource = DataSource(sys_cfg, self.distro, self.paths)
+        url_params = datasource.get_url_params()
+        expected = (datasource.url_max_wait, datasource.url_timeout,
+                    datasource.url_retries)
+        self.assertEqual(expected, url_params)
+        logs = self.logs.getvalue()
+        expected_logs = [
+            "Config max_wait 'nope' is not an int, using default '-1'",
+            "Config timeout 'bug' is not an int, using default '10'",
+            "Config retries 'nonint' is not an int, using default '5'",
+        ]
+        for log in expected_logs:
+            self.assertIn(log, logs)
+
+    @mock.patch('cloudinit.sources.net.find_fallback_nic')
+    def test_fallback_interface_is_discovered(self, m_get_fallback_nic):
+        """The fallback_interface is discovered via find_fallback_nic."""
+        m_get_fallback_nic.return_value = 'nic9'
+        self.assertEqual('nic9', self.datasource.fallback_interface)
+
+    @mock.patch('cloudinit.sources.net.find_fallback_nic')
+    def test_fallback_interface_logs_undiscovered(self, m_get_fallback_nic):
+        """Log a warning when fallback_interface can not discover the nic."""
+        self.datasource._cloud_name = 'MySupahCloud'
+        m_get_fallback_nic.return_value = None  # Couldn't discover nic
+        self.assertIsNone(self.datasource.fallback_interface)
+        self.assertEqual(
+            'WARNING: Did not find a fallback interface on MySupahCloud.\n',
+            self.logs.getvalue())
+
+    @mock.patch('cloudinit.sources.net.find_fallback_nic')
+    def test_wb_fallback_interface_is_cached(self, m_get_fallback_nic):
+        """The fallback_interface is cached and won't be rediscovered."""
+        self.datasource._fallback_interface = 'nic10'
+        self.assertEqual('nic10', self.datasource.fallback_interface)
+        m_get_fallback_nic.assert_not_called()
+
     def test__get_data_unimplemented(self):
         """Raise an error when _get_data is not implemented."""
         with self.assertRaises(NotImplementedError) as context_manager: