From 1979ea3e3440335632af8e7e58dd34aae52a2b96 Mon Sep 17 00:00:00 2001
From: Ben Howard <ben.howard@canonical.com>
Date: Wed, 11 Sep 2013 14:55:32 -0600
Subject: Initial cut at disk partition support.

---
 doc/examples/cloud-config-disk-setup.txt | 178 +++++++++++++++++++++++++++++++
 1 file changed, 178 insertions(+)
 create mode 100644 doc/examples/cloud-config-disk-setup.txt

(limited to 'doc/examples')

diff --git a/doc/examples/cloud-config-disk-setup.txt b/doc/examples/cloud-config-disk-setup.txt
new file mode 100644
index 00000000..518936d1
--- /dev/null
+++ b/doc/examples/cloud-config-disk-setup.txt
@@ -0,0 +1,178 @@
+Cloud-init supports the creation of simple partition tables and filesystems
+on devices.
+
+default disk definitions
+------------------------
+
+ disk_setup:
+    ephmeral0:
+        type: 'mbr'
+        layout: True
+        overwrite: False
+ fs_setup:
+    ephemeral0:
+        filesystem: 'ext3'
+        device: 'ephemeral0'
+        partition: 'auto'
+
+The default definition is used to make sure that the ephemeral storage is
+setup properly.
+
+"disk_setup": disk parititioning
+--------------------------------
+
+The disk_setup directive instructs Cloud-init to partition a disk. The format is:
+
+ disk_setup:
+    ephmeral0:
+        type: 'mbr'
+        layout: 'auto'
+    /dev/xvdh:
+        type: 'mbr'
+        layout:
+            - 33
+            - [33, 82]
+            - 33
+        overwrite: True
+
+The format is a list of dicts of dicts. The first value is the name of the
+device and the subsiquent values define how to create and layout the partition.
+
+The general format is:
+    disk_setup:
+        <DEVICE>:
+            type: 'mbr'
+            layout: <LAYOUT|BOOL>
+            overwrite: <BOOL>
+
+Where:
+    <DEVICE>: The name of the device. 'ephemeralX' and 'swap' are special
+                values which are specific to the cloud. For these devices
+                Cloud-init will look up what the real devices is and then
+                use it.
+
+                For other devices, the kernel device name is used. At this
+                time only simply kernel devices are supported, meaning
+                that device mapper and other targets may not work.
+
+                Note: At this time, there is no handling or setup of
+                device mapper targets.
+
+    type=<TYPE>: Currently the following are supported:
+                    'mbr': default and setups a MS-DOS partition table
+
+                Note: At this time only 'mbr' partition tables are allowed.
+                    It is anticipated in the future that we'll have GPT as
+                    option in the future, or even "RAID" to create a mdadm
+                    RAID.
+
+    layout={...}: The device layout. This is a list of values, with the
+                percentage of disk that partition will take.
+                Valid options are:
+                    [<SIZE>, [<SIZE>, <PART_TYPE]]
+
+                Where <SIZE> is the _percentage_ of the disk to use, while
+                <PART_TYPE> is the numerical value of the partition type.
+
+                The following setups two partitions, with the first
+                partition having a swap label, taking 1/3 of the disk space
+                and the remainder being used as the second partition.
+                    /dev/xvdh':
+                        type: 'mbr'
+                        layout:
+                            - [33,82]
+                            - 66
+                        overwrite: True
+
+                When layout is "true" it means single partition the entire
+                device.
+
+                When layout is "false" it means don't partition or ignore
+                existing partitioning.
+
+                If layout is set to "true" and overwrite is set to "false",
+                it will skip partitioning the device without a failure.
+
+    overwrite=<BOOL>: This describes whether to ride with saftey's on and
+                everything holstered.
+
+                'false' is the default, which means that:
+                    1. The device will be checked for a partition table
+                    2. The device will be checked for a file system
+                    3. If either a partition of file system is found, then
+                        the operation will be _skipped_.
+
+                'true' is cowboy mode. There are no checks and things are
+                    done blindly. USE with caution, you can do things you
+                    really, really don't want to do.
+
+
+fs_setup: Setup the file system
+-------------------------------
+
+fs_setup describes the how the file systems are supposed to look.
+
+ fs_setup:
+    ephemeral0:
+        filesystem: 'ext3'
+        device: 'ephemeral0'
+        partition: 'auto'
+    mylabl2:
+        filesystem: 'ext4'
+        device: '/dev/xvda1'
+    special:
+        cmd: mkfs -t %(FILESYSTEM)s -L %(LABEL)s %(DEVICE)s
+        filesystem: 'btrfs'
+        device: '/dev/xvdh'
+
+The general format is:
+    fs_setup:
+        <LABEL>:
+            filesystem: <FS_TYPE>
+            device: <DEVICE>
+            partition: <PART_VALUE>
+            overwrite: <OVERWRITE>
+
+Where:
+    <LABEL>: The file system label to be used.
+
+    <FS_TYPE>: The file system type. It is assumed that the there
+        will be a "mkfs.<FS_TYPE>" that behaves likes "mkfs". On a standard
+        Ubuntu Cloud Image, this means that you have the option of ext{2,3,4},
+        and vfat by default.
+
+    <DEVICE>: The device name. Special names of 'ephemeralX' or 'swap'
+        are allowed and the actual device is acquired from the cloud datasource.
+        When using 'ephemeralX' (i.e. ephemeral0), make sure to leave the
+        label as 'ephemeralX' otherwise there may be issues with the mounting
+        of the ephemeral storage layer.
+
+    <PART_VALUE>: The valid options are:
+        "auto": auto is a special in the sense that you are telling cloud-init
+            not to care whether there is a partition or not. Auto will put the
+            first partition that does not contain a file system already. In
+            the absence of a partition table, it will put it directly on the
+            disk.
+
+        "none": Put the partition directly on the disk.
+
+        <NUM>: where NUM is the actual partition number.
+
+    <OVERWRITE>: Defines whether or not to overwrite any existing
+        filesystem.
+
+        "true": Indiscriminately destroy any pre-existing file system. Use at
+            your own peril.
+
+        "false": If an existing file system exists, skip the creation.
+
+        "force": Recreate the file system, even it already exists
+
+
+Behavior Caveat: The default behavior is to _check_ if the file system exists.
+    If a file system matches the specification, then the operation is a no-op.
+
+    For 'ephemeralX' or 'swap' labeled filesystems, the operation will be a
+    no-op if a file system of the same type is present, regardless of the label.
+    This is to accommodate Clouds like EC2 that present a blank file system with
+    out a label.
-- 
cgit v1.2.3


From d1bad8880c2219b9d7a648169bbe7a0a27c03be2 Mon Sep 17 00:00:00 2001
From: Ben Howard <ben.howard@canonical.com>
Date: Thu, 19 Sep 2013 16:49:50 -0600
Subject: Fixes for the MP.

Changed cc_disk_setup to handle the file systems as a label, no longer
passing "log" around.

Tidied up the documentation to reflect the changes and made grammer,
spelling and improved the content a little.

Added disk_setup to the default modules list.
---
 cloudinit/config/cc_disk_setup.py               | 241 ++++++++++++++----------
 cloudinit/sources/DataSourceSmartOS.py          |  72 +++++--
 config/cloud.cfg                                |  15 +-
 doc/examples/cloud-config-disk-setup.txt        | 100 ++++++----
 doc/sources/smartos/README.rst                  |  27 ++-
 tests/unittests/test_datasource/test_smartos.py |  35 ++++
 6 files changed, 317 insertions(+), 173 deletions(-)

(limited to 'doc/examples')

diff --git a/cloudinit/config/cc_disk_setup.py b/cloudinit/config/cc_disk_setup.py
index 6af2c9b1..fb404c5d 100644
--- a/cloudinit/config/cc_disk_setup.py
+++ b/cloudinit/config/cc_disk_setup.py
@@ -18,14 +18,11 @@
 #    along with this program.  If not, see <http://www.gnu.org/licenses/>.
 from cloudinit import util
 from cloudinit.settings import PER_INSTANCE
-import re
-import traceback
+import logging
+import shlex
 
 frequency = PER_INSTANCE
 
-virtal_devices = ["ephemeral0", "swap"]
-defmnts = ["ephemeral0", "swap"]
-
 # Define the commands to use
 UDEVADM_CMD = util.which('udevadm')
 SFDISK_CMD = util.which("sfdisk")
@@ -33,46 +30,47 @@ LSBLK_CMD = util.which("lsblk")
 BLKID_CMD = util.which("blkid")
 BLKDEV_CMD = util.which("blockdev")
 
+LOG = logging.getLogger(__name__)
+
+
 def handle(_name, cfg, cloud, log, _args):
     """
     Call util.prep_disk for disk_setup cloud-config.
-    The format is:
-
-    disk_setup:
-        ephmeral0: {type='mbr', layout='True', overwrite='False'}
-        /dev/xvdj: {type='None'}
-        /dev/xvdh: {type='mbr', layout:[(33,83),66], overwrite='True'}
-
-    fs_setup:
-        ephemeral0: {filesystem='ext3', device='ephemeral0', partition='auto'}
-        mylabel2: {filesystem='ext3', device='/dev/xvda1', partition='None'}
-        special1: {cmd="mkfs -t %(FILESYSTEM)s -L %(LABEL)s %(DEVICE)s", filesystem='btrfs', device='/dev/xvda1'}
-
-
+    See doc/examples/cloud-config_disk-setup.txt for documentation on the
+    format.
     """
-
     disk_setup = cfg.get("disk_setup")
     if isinstance(disk_setup, dict):
         log.info("Partitioning disks.")
         for disk, definition in disk_setup.items():
             if not isinstance(definition, dict):
-                log.debug("Invalid disk definition for %s" % disk)
+                log.warn("Invalid disk definition for %s" % disk)
                 continue
 
-            util.log_time(logfunc=log.info,
+            try:
+                log.debug("Creating new partition table/disk")
+                util.log_time(logfunc=LOG.debug,
                               msg="Creating partition on %s" % disk,
-                              func=mkpart, args=(disk, cloud, definition, log))
+                              func=mkpart, args=(disk, cloud, definition))
+            except Exception as e:
+                util.logexc(LOG, "Failed partitioning operation\n%s" % e)
 
     fs_setup = cfg.get("fs_setup")
-    if isinstance(fs_setup, dict):
-        log.info("Setting up filesystems")
-        for label, definition in fs_setup.items():
+    if isinstance(fs_setup, list):
+        log.info("Creating file systems.")
+        for definition in fs_setup:
             if not isinstance(definition, dict):
-               log.debug("Invalid filesystem definition for %s" % label)
-               continue
+                log.warn("Invalid file system definition: %s" % definition)
+                continue
 
-            util.log_time(logfunc=log.debug, msg="Creating fs for %s" % label,
-                         func=mkfs, args=(label, cloud, definition, log))
+            try:
+                log.debug("Creating new filesystem.")
+                device = definition.get('device')
+                util.log_time(logfunc=LOG.debug,
+                              msg="Creating fs for %s" % device,
+                              func=mkfs, args=(cloud, definition))
+            except Exception as e:
+                util.logexc(LOG, "Failed during filesystem operation\n%s" % e)
 
 
 def is_default_device(name, cloud, fallback=None):
@@ -82,10 +80,11 @@ def is_default_device(name, cloud, fallback=None):
     fallback if so defined.
     """
 
+    _dev = None
     try:
         _dev = cloud.device_name_to_device(name)
     except Exception as e:
-        print e
+        util.logexc(LOG, "Failed to find mapping for %s" % e)
 
     if _dev:
         return _dev
@@ -96,34 +95,19 @@ def is_default_device(name, cloud, fallback=None):
     return name
 
 
-def check_value(key, dct, default=None):
-    """
-    Convience function for getting value out of a dict.
-    """
-
-    if key in dct:
-        return dct[key]
-    if default:
-        return default
-    return None
-
-
 def value_splitter(values, start=None):
     """
     Returns the key/value pairs of output sent as string
     like:  FOO='BAR' HOME='127.0.0.1'
     """
-    _values = values.split()
+    _values = shlex.split(values)
     if start:
         _values = _values[start:]
 
     for key, value in [x.split('=') for x in _values]:
-        if value == '""':
-            value = None
-        elif '"' in value:
-            value = value.replace('"','')
         yield key, value
 
+
 def device_type(device):
     """
     Return the device type of the device by calling lsblk.
@@ -148,7 +132,13 @@ def is_device_valid(name, partition=False):
     """
     Check if the device is a valid device.
     """
-    d_type = device_type(name)
+    d_type = ""
+    try:
+        d_type = device_type(name)
+    except:
+        LOG.warn("Query against device %s failed" % name)
+        return False
+
     if partition and d_type == 'part':
         return True
     elif not partition and d_type == 'disk':
@@ -171,7 +161,6 @@ def check_fs(device):
     try:
         out, _err = util.subp(blkid_cmd, rcs=[0, 2])
     except Exception as e:
-        util.logexc(e)
         raise Exception("Failed during disk check for %s\n%s" % (device, e))
 
     if out:
@@ -195,7 +184,8 @@ def is_filesystem(device):
     return fs_type
 
 
-def find_device_node(device, fs_type=None, label=None, valid_targets=None):
+def find_device_node(device, fs_type=None, label=None, valid_targets=None,
+                     label_match=True):
     """
     Find a device that is either matches the spec, or the first
 
@@ -208,7 +198,8 @@ def find_device_node(device, fs_type=None, label=None, valid_targets=None):
     if not valid_targets:
         valid_targets = ['disk', 'part']
 
-    lsblk_cmd = [LSBLK_CMD, '--pairs', '--out', 'NAME,TYPE,FSTYPE,LABEL', device]
+    lsblk_cmd = [LSBLK_CMD, '--pairs', '--out', 'NAME,TYPE,FSTYPE,LABEL',
+                 device]
     info = None
     try:
         info, _err = util.subp(lsblk_cmd)
@@ -216,7 +207,7 @@ def find_device_node(device, fs_type=None, label=None, valid_targets=None):
         raise Exception("Failed during disk check for %s\n%s" % (device, e))
 
     raw_device_used = False
-    parts = [ x for x in (info.strip()).splitlines() if len(x.split()) > 0 ]
+    parts = [x for x in (info.strip()).splitlines() if len(x.split()) > 0]
 
     for part in parts:
         d = {'name': None,
@@ -228,7 +219,8 @@ def find_device_node(device, fs_type=None, label=None, valid_targets=None):
         for key, value in value_splitter(part):
             d[key.lower()] = value
 
-        if d['fstype'] == fs_type and d['label'] == label:
+        if d['fstype'] == fs_type and \
+            ((label_match and d['label'] == label) or not label_match):
             # If we find a matching device, we return that
             return ('/dev/%s' % d['name'], True)
 
@@ -247,6 +239,7 @@ def find_device_node(device, fs_type=None, label=None, valid_targets=None):
     if not raw_device_used:
         return (device, False)
 
+    LOG.warn("Failed to find device during available device search.")
     return (None, False)
 
 
@@ -262,6 +255,8 @@ def is_disk_used(device):
         info, _err = util.subp(lsblk_cmd)
     except Exception as e:
         # if we error out, we can't use the device
+        util.logexc(LOG,
+                    "Error checking for filesystem on %s\n%s" % (device, e))
         return True
 
     # If there is any output, then the device has something
@@ -313,6 +308,7 @@ def get_dyn_func(*args):
     except KeyError:
         raise Exception("No such function %s to call!" % func_name)
 
+
 def check_partition_mbr_layout(device, layout):
     """
     Returns true if the partition layout matches the one on the disk
@@ -363,7 +359,7 @@ def check_partition_mbr_layout(device, layout):
             # This just makes sure that the number of requested
             # partitions and the type labels are right
             for x in range(1, len(layout) + 1):
-                if isinstance(layout[x-1], tuple):
+                if isinstance(layout[x - 1], tuple):
                     _, part_type = layout[x]
                     if int(found_layout[x]) != int(part_type):
                         return False
@@ -455,10 +451,13 @@ def read_parttbl(device):
     reliable way to probe the partition table.
     """
     blkdev_cmd = [BLKDEV_CMD, '--rereadpt', device]
+    udev_cmd = [UDEVADM_CMD, 'settle']
     try:
+        util.subp(udev_cmd)
         util.subp(blkdev_cmd)
+        util.subp(udev_cmd)
     except Exception as e:
-        raise Exception("Failed on call to partprobe\n%s" % e)
+        util.logexc(LOG, "Failed reading the partition table %s" % e)
 
 
 def exec_mkpart_mbr(device, layout):
@@ -475,6 +474,7 @@ def exec_mkpart_mbr(device, layout):
 
     read_parttbl(device)
 
+
 def exec_mkpart(table_type, device, layout):
     """
     Fetches the function for creating the table type.
@@ -488,7 +488,7 @@ def exec_mkpart(table_type, device, layout):
     return get_dyn_func("exec_mkpart_%s", table_type, device, layout)
 
 
-def mkpart(device, cloud, definition, log):
+def mkpart(device, cloud, definition):
     """
     Creates the partition table.
 
@@ -504,14 +504,14 @@ def mkpart(device, cloud, definition, log):
                 device: the device to work on.
     """
 
-    log.debug("Checking values for %s definition" % device)
-    overwrite = check_value('overwrite', definition, False)
-    layout = check_value('layout', definition, False)
-    table_type = check_value('type', definition, 'mbr')
+    LOG.debug("Checking values for %s definition" % device)
+    overwrite = definition.get('overwrite', False)
+    layout = definition.get('layout', False)
+    table_type = definition.get('table_type', 'mbr')
     _device = is_default_device(device, cloud)
 
     # Check if the default device is a partition or not
-    log.debug("Checking against default devices")
+    LOG.debug("Checking against default devices")
     if _device and (_device != device):
         if not is_device_valid(_device):
             _device = _device[:-1]
@@ -519,40 +519,43 @@ def mkpart(device, cloud, definition, log):
         if not is_device_valid(_device):
             raise Exception("Unable to find backing block device for %s" % \
                             device)
+        else:
+            LOG.debug("Mapped %s to physical device %s" % (device, _device))
+            device = _device
 
     if (isinstance(layout, bool) and not layout) or not layout:
-        log.debug("Device is not to be partitioned, skipping")
+        LOG.debug("Device is not to be partitioned, skipping")
         return  # Device is not to be partitioned
 
     # This prevents you from overwriting the device
-    log.debug("Checking if device %s is a valid device" % device)
+    LOG.debug("Checking if device %s is a valid device" % device)
     if not is_device_valid(device):
         raise Exception("Device %s is not a disk device!" % device)
 
-    log.debug("Checking if device layout matches")
+    LOG.debug("Checking if device layout matches")
     if check_partition_layout(table_type, device, layout):
-        log.debug("Device partitioning layout matches")
+        LOG.debug("Device partitioning layout matches")
         return True
 
-    log.debug("Checking if device is safe to partition")
+    LOG.debug("Checking if device is safe to partition")
     if not overwrite and (is_disk_used(device) or is_filesystem(device)):
-        log.debug("Skipping partitioning on configured device %s" % device)
+        LOG.debug("Skipping partitioning on configured device %s" % device)
         return
 
-    log.debug("Checking for device size")
+    LOG.debug("Checking for device size")
     device_size = get_hdd_size(device)
 
-    log.debug("Calculating partition layout")
+    LOG.debug("Calculating partition layout")
     part_definition = get_partition_layout(table_type, device_size, layout)
-    log.debug("   Layout is: %s" % part_definition)
+    LOG.debug("   Layout is: %s" % part_definition)
 
-    log.debug("Creating partition table on %s" % device)
+    LOG.debug("Creating partition table on %s" % device)
     exec_mkpart(table_type, device, part_definition)
 
-    log.debug("Partition table created for %s" % device)
+    LOG.debug("Partition table created for %s" % device)
 
 
-def mkfs(label, cloud, fs_cfg, log):
+def mkfs(cloud, fs_cfg):
     """
     Create a file system on the device.
 
@@ -568,68 +571,94 @@ def mkfs(label, cloud, fs_cfg, log):
                             first free device or the first device which
                             matches both label and type will be used.
 
+                            'any' means the first filesystem that matches
+                            on the device.
+
             When 'cmd' is provided then no other parameter is required.
     """
-    device = check_value('device', fs_cfg)
-    partition = str(check_value('partition', fs_cfg))
-    fs_type = check_value('filesystem', fs_cfg)
-    fs_cmd = check_value('cmd', fs_cfg, [])
-    fs_opts = check_value('extra_opts', fs_cfg, [])
-    overwrite = check_value('overwrite', fs_cfg, False)
+    fs_cfg['partition'] = 'any'
+    label = fs_cfg.get('label')
+    device = fs_cfg.get('device')
+    partition = str(fs_cfg.get('partition'))
+    fs_type = fs_cfg.get('filesystem')
+    fs_cmd = fs_cfg.get('cmd', [])
+    fs_opts = fs_cfg.get('extra_opts', [])
+    overwrite = fs_cfg.get('overwrite', False)
 
     # This allows you to define the default ephemeral or swap
-    log.debug("Checking %s label against default devices" % label)
-    device = is_default_device(label, cloud, fallback=device)
+    LOG.debug("Checking %s against default devices" % device)
+    _device = is_default_device(label, cloud, fallback=device)
+    if _device and (_device != device):
+        if not is_device_valid(_device):
+            raise Exception("Unable to find backing block device for %s" % \
+                            device)
+        else:
+            LOG.debug("Mapped %s to physical device %s" % (device, _device))
+            device = _device
 
     if not partition or partition.isdigit():
         # Handle manual definition of partition
         if partition.isdigit():
             device = "%s%s" % (device, partition)
-            log.debug("Manual request of partition %s for %s" % (
-                         partition, device))
+            LOG.debug("Manual request of partition %s for %s" % (
+                      partition, device))
 
         # Check to see if the fs already exists
-        log.debug("Checking device %s" % device)
+        LOG.debug("Checking device %s" % device)
         check_label, check_fstype, _ = check_fs(device)
-        log.debug("Device %s has %s %s" % (device, check_label, check_fstype))
+        LOG.debug("Device %s has %s %s" % (device, check_label, check_fstype))
 
         if check_label == label and check_fstype == fs_type:
-            log.debug("Existing file system found at %s" % device)
+            LOG.debug("Existing file system found at %s" % device)
 
             if not overwrite:
-                log.debug("Device %s has required file system" % device)
+                LOG.warn("Device %s has required file system" % device)
                 return
             else:
-                log.debug("Destroying filesystem on %s" % device)
+                LOG.warn("Destroying filesystem on %s" % device)
 
         else:
-            log.debug("Device %s is cleared for formating" % device)
+            LOG.debug("Device %s is cleared for formating" % device)
 
-    elif partition and partition == 'auto':
+    elif partition and str(partition).lower() in ('auto', 'any'):
         # For auto devices, we match if the filesystem does exist
-        log.debug("Identifying device to create %s filesytem on" % label)
-        device, reuse = find_device_node(device, fs_type=fs_type, label=label)
-        log.debug("Device identified as %s" % device)
+        odevice = device
+        LOG.debug("Identifying device to create %s filesytem on" % label)
+
+        # any mean pick the first match on the device with matching fs_type
+        label_match = True
+        if partition.lower() == 'any':
+            label_match = False
+
+        device, reuse = find_device_node(device, fs_type=fs_type, label=label,
+                                         label_match=label_match)
+        LOG.debug("Automatic device for %s identified as %s" % (
+                  odevice, device))
 
         if reuse:
-            log.debug("Found filesystem match, skipping formating.")
+            LOG.debug("Found filesystem match, skipping formating.")
+            return
+
+        if not device:
+            LOG.debug("No device aviable that matches request.")
+            LOG.debug("Skipping fs creation for %s" % fs_cfg)
             return
 
     else:
-        log.debug("Error in device identification handling.")
+        LOG.debug("Error in device identification handling.")
         return
 
-
-    log.debug("File system %s will be created on %s" % (label, device))
+    LOG.debug("File system %s will be created on %s" % (label, device))
 
     # Make sure the device is defined
     if not device:
-        raise Exception("Device identification error for %s" % label)
+        LOG.critical("Device is not known: %s" % fs_cfg)
+        return
 
     # Check that we can create the FS
     if not label or not fs_type:
-        log.debug("Command to create filesystem %s is bad. Skipping." % \
-                     label)
+        LOG.debug("Command to create filesystem %s is bad. Skipping." % \
+                  label)
 
     # Create the commands
     if fs_cmd:
@@ -638,22 +667,26 @@ def mkfs(label, cloud, fs_cfg, log):
                                   'device': device,
                                  }
     else:
+        # Find the mkfs command
         mkfs_cmd = util.which("mkfs.%s" % fs_type)
         if not mkfs_cmd:
             mkfs_cmd = util.which("mk%s" % fs_type)
 
         if not mkfs_cmd:
-            log.debug("Unable to locate command to create filesystem.")
+            LOG.critical("Unable to locate command to create filesystem.")
             return
 
-        fs_cmd = [mkfs_cmd, "-L", label, device]
+        fs_cmd = [mkfs_cmd, device]
+
+        if label:
+            fs_cmd.extend(["-L", label])
 
     # Add the extends FS options
     if fs_opts:
         fs_cmd.extend(fs_opts)
 
-    log.debug("Creating file system %s on %s" % (label, device))
-    print fs_cmd
+    LOG.debug("Creating file system %s on %s" % (label, device))
+    LOG.debug("     Using cmd: %s" % "".join(fs_cmd))
     try:
         util.subp(fs_cmd)
     except Exception as e:
diff --git a/cloudinit/sources/DataSourceSmartOS.py b/cloudinit/sources/DataSourceSmartOS.py
index d348d20b..da1eec79 100644
--- a/cloudinit/sources/DataSourceSmartOS.py
+++ b/cloudinit/sources/DataSourceSmartOS.py
@@ -35,8 +35,7 @@ import os
 import os.path
 import serial
 
-DEF_TTY_LOC = '/dev/ttyS1'
-DEF_TTY_TIMEOUT = 60
+
 LOG = logging.getLogger(__name__)
 
 SMARTOS_ATTRIB_MAP = {
@@ -49,24 +48,61 @@ SMARTOS_ATTRIB_MAP = {
     'motd_sys_info': ('motd_sys_info', True),
 }
 
-# These are values which will never be base64 encoded.
-# They come from the cloud platform, not user
-SMARTOS_NO_BASE64 = ['root_authorized_keys', 'motd_sys_info',
-                     'iptables_disable']
+DS_NAME = 'SmartOS'
+DS_CFG_PATH = ['datasource', DS_NAME]
+# BUILT-IN DATASOURCE CONFIGURATION
+#  The following is the built-in configuration. If the values
+#  are not set via the system configuration, then these default
+#  will be used:
+#    serial_device: which serial device to use for the meta-data
+#    seed_timeout: how long to wait on the device
+#    no_base64_decode: values which are not base64 encoded and
+#            are fetched directly from SmartOS, not meta-data values
+#    base64_keys: meta-data keys that are delivered in base64
+#    base64_all: with the exclusion of no_base64_decode values,
+#            treat all meta-data as base64 encoded
+#    disk_setup: describes how to partition the ephemeral drive
+#    fs_setup: describes how to format the ephemeral drive
+#
+BUILTIN_DS_CONFIG = {
+    'serial_device': '/dev/ttyS1',
+    'seed_timeout': 60,
+    'no_base64_decode': ['root_authorized_keys',
+                         'motd_sys_info',
+                         'iptables_disable'],
+    'base64_keys': [],
+    'base64_all': False,
+    'ephemeral_disk': '/dev/vdb',
+    'disk_setup': {
+        'ephemeral0': {'table_type': 'mbr',
+                       'layout': True,
+                       'overwrite': False}
+         },
+    'fs_setup': [{'label': 'ephemeral0', 'filesystem': 'ext3',
+                  'device': '/dev/xvdb', 'partition': 'auto'}],
+}
 
 
 class DataSourceSmartOS(sources.DataSource):
     def __init__(self, sys_cfg, distro, paths):
         sources.DataSource.__init__(self, sys_cfg, distro, paths)
-        self.seed_dir = os.path.join(paths.seed_dir, 'sdc')
         self.is_smartdc = None
 
-        self.seed = self.ds_cfg.get("serial_device", DEF_TTY_LOC)
-        self.seed_timeout = self.ds_cfg.get("serial_timeout", DEF_TTY_TIMEOUT)
-        self.smartos_no_base64 = self.ds_cfg.get('no_base64_decode',
-                                                 SMARTOS_NO_BASE64)
-        self.b64_keys = self.ds_cfg.get('base64_keys', [])
-        self.b64_all = self.ds_cfg.get('base64_all', False)
+        self.ds_cfg = util.mergemanydict([
+            self.ds_cfg,
+            util.get_cfg_by_path(sys_cfg, DS_CFG_PATH, {}),
+            BUILTIN_DS_CONFIG])
+
+        self.metadata = {}
+        self.cfg = {}
+        self.cfg['disk_setup'] = self.ds_cfg.get('disk_setup')
+        self.cfg['fs_setup'] = self.ds_cfg.get('fs_setup')
+
+        self.seed = self.ds_cfg.get("serial_device")
+        self.seed_timeout = self.ds_cfg.get("serial_timeout")
+        self.smartos_no_base64 = self.ds_cfg.get('no_base64_decode')
+        self.b64_keys = self.ds_cfg.get('base64_keys')
+        self.b64_all = self.ds_cfg.get('base64_all')
 
     def __str__(self):
         root = sources.DataSource.__str__(self)
@@ -79,7 +115,6 @@ class DataSourceSmartOS(sources.DataSource):
         if not os.path.exists(self.seed):
             LOG.debug("Host does not appear to be on SmartOS")
             return False
-        self.seed = self.seed
 
         dmi_info = dmi_data()
         if dmi_info is False:
@@ -114,10 +149,17 @@ class DataSourceSmartOS(sources.DataSource):
         elif md['user-script']:
             ud = md['user-script']
 
-        self.metadata = md
+        self.metadata = util.mergemanydict([md, self.metadata])
         self.userdata_raw = ud
         return True
 
+    def device_name_to_device(self, name):
+        if 'ephemeral0' in name:
+            return self.ds_cfg['ephemeral_disk']
+
+    def get_config_obj(self):
+        return self.cfg
+
     def get_instance_id(self):
         return self.metadata['instance-id']
 
diff --git a/config/cloud.cfg b/config/cloud.cfg
index 8b52ca39..382cfaf8 100644
--- a/config/cloud.cfg
+++ b/config/cloud.cfg
@@ -14,20 +14,6 @@ disable_root: true
 # This will cause the set+update hostname module to not operate (if true)
 preserve_hostname: false
 
-# This sets the default for creating the ephemeral0 on clouds
-# that support it.
-disk_setup:
-    ephemeral0
-         type: 'mbr'
-         layout: True
-         overwrite: False
-
-fs_setup:
-    ephemeral0
-        filesystem: 'ext3'
-        device: ephemeral0
-        partition: 'auto'
-
 # Example datasource config
 # datasource: 
 #    Ec2: 
@@ -56,6 +42,7 @@ cloud_config_modules:
 # Emit the cloud config ready event
 # this can be used by upstart jobs for 'start on cloud-config'.
  - emit_upstart
+ - disk_setup
  - mounts
  - ssh-import-id
  - locale
diff --git a/doc/examples/cloud-config-disk-setup.txt b/doc/examples/cloud-config-disk-setup.txt
index 518936d1..db2c52a7 100644
--- a/doc/examples/cloud-config-disk-setup.txt
+++ b/doc/examples/cloud-config-disk-setup.txt
@@ -1,24 +1,63 @@
-Cloud-init supports the creation of simple partition tables and filesystems
+Cloud-init supports the creation of simple partition tables and file systems
 on devices.
 
-default disk definitions
-------------------------
+Default disk definitions for AWS
+--------------------------------
+(Not implemented yet, but provided for future documentation)
 
  disk_setup:
     ephmeral0:
         type: 'mbr'
         layout: True
         overwrite: False
+
  fs_setup:
-    ephemeral0:
-        filesystem: 'ext3'
-        device: 'ephemeral0'
-        partition: 'auto'
+    - label: None,
+      filesystem: ext3
+      device: ephemeral0
+      partition: auto
+
+Default disk definitions for Windows Azure
+------------------------------------------
+(Not implemented yet due to conflict with WALinuxAgent in Ubuntu)
+
+disk_setup:
+    /dev/sdb:
+         type: mbr
+         layout: True
+         overwrite: False
+
+fs_setup:
+    - label: ephemeral0
+      filesystem: ext3
+      device: ephemeral0
+      partition: any
+
+
+Default disk definitions for SmartOS
+------------------------------------
+
+ephemeral_disk: /dev/vdb
+disk_setup:
+    /dev/vdb:
+         type: mbr
+         layout: True
+         overwrite: False
+
+fs_setup:
+    - label: ephemeral0
+      filesystem: ext3
+      device: /dev/vdb
+      partition: 1
+
+Cavaut for SmartOS: if ephemeral disk is not defined, then the disk will
+    not be automatically added to the mounts.
+
 
 The default definition is used to make sure that the ephemeral storage is
 setup properly.
 
-"disk_setup": disk parititioning
+"disk_setup": disk partitioning
 --------------------------------
 
 The disk_setup directive instructs Cloud-init to partition a disk. The format is:
@@ -36,7 +75,7 @@ The disk_setup directive instructs Cloud-init to partition a disk. The format is
         overwrite: True
 
 The format is a list of dicts of dicts. The first value is the name of the
-device and the subsiquent values define how to create and layout the partition.
+device and the subsequent values define how to create and layout the partition.
 
 The general format is:
     disk_setup:
@@ -113,28 +152,29 @@ fs_setup: Setup the file system
 fs_setup describes the how the file systems are supposed to look.
 
  fs_setup:
-    ephemeral0:
-        filesystem: 'ext3'
-        device: 'ephemeral0'
-        partition: 'auto'
-    mylabl2:
-        filesystem: 'ext4'
-        device: '/dev/xvda1'
-    special:
-        cmd: mkfs -t %(FILESYSTEM)s -L %(LABEL)s %(DEVICE)s
-        filesystem: 'btrfs'
-        device: '/dev/xvdh'
+    - label: ephemeral0
+      filesystem: 'ext3'
+      device: 'ephemeral0'
+      partition: 'auto'
+    - label:  mylabl2
+      filesystem: 'ext4'
+      device: '/dev/xvda1'
+    - special:
+      cmd: mkfs -t %(FILESYSTEM)s -L %(LABEL)s %(DEVICE)s
+      filesystem: 'btrfs'
+      device: '/dev/xvdh'
 
 The general format is:
     fs_setup:
-        <LABEL>:
-            filesystem: <FS_TYPE>
-            device: <DEVICE>
-            partition: <PART_VALUE>
-            overwrite: <OVERWRITE>
+        - label: <LABEL>
+          filesystem: <FS_TYPE>
+          device: <DEVICE>
+          partition: <PART_VALUE>
+          overwrite: <OVERWRITE>
 
 Where:
-    <LABEL>: The file system label to be used.
+    <LABEL>: The file system label to be used. If set to None, no label is
+        used.
 
     <FS_TYPE>: The file system type. It is assumed that the there
         will be a "mkfs.<FS_TYPE>" that behaves likes "mkfs". On a standard
@@ -166,13 +206,5 @@ Where:
 
         "false": If an existing file system exists, skip the creation.
 
-        "force": Recreate the file system, even it already exists
-
-
 Behavior Caveat: The default behavior is to _check_ if the file system exists.
     If a file system matches the specification, then the operation is a no-op.
-
-    For 'ephemeralX' or 'swap' labeled filesystems, the operation will be a
-    no-op if a file system of the same type is present, regardless of the label.
-    This is to accommodate Clouds like EC2 that present a blank file system with
-    out a label.
diff --git a/doc/sources/smartos/README.rst b/doc/sources/smartos/README.rst
index fd4e496d..e2d3312e 100644
--- a/doc/sources/smartos/README.rst
+++ b/doc/sources/smartos/README.rst
@@ -5,11 +5,13 @@ SmartOS Datasource
 This datasource finds metadata and user-data from the SmartOS virtualization
 platform (i.e. Joyent).
 
+Please see http://smartos.org/ for information about SmartOS.
+
 SmartOS Platform
 ----------------
-The SmartOS virtualization platform meta-data to the instance via the second
-serial console. On Linux, this is /dev/ttyS1. The data is a provided via a
-simple protocol, where something queries for the userdata, where the console
+The SmartOS virtualization platform uses meta-data to the instance via the
+second serial console. On Linux, this is /dev/ttyS1. The data is a provided
+via a simple protocol: something queries for the data, the console responds
 responds with the status and if "SUCCESS" returns until a single ".\n".
 
 New versions of the SmartOS tooling will include support for base64 encoded data.
@@ -18,7 +20,7 @@ Userdata
 --------
 
 In SmartOS parlance, user-data is a actually meta-data. This userdata can be
-provided a key-value pairs.
+provided as key-value pairs.
 
 Cloud-init supports reading the traditional meta-data fields supported by the
 SmartOS tools. These are:
@@ -36,13 +38,13 @@ user-script
 SmartOS traditionally supports sending over a user-script for execution at the
 rc.local level. Cloud-init supports running user-scripts as if they were
 cloud-init user-data. In this sense, anything with a shell interpreter
-directive will run
+directive will run.
 
 user-data and user-script
 -------------------------
 
 In the event that a user defines the meta-data key of "user-data" it will
-always supercede any user-script data. This is for consistency.
+always supersede any user-script data. This is for consistency.
 
 base64
 ------
@@ -70,3 +72,16 @@ or not to base64 decode something:
   * no_base64_decode: This is a configuration setting
         (i.e. /etc/cloud/cloud.cfg.d) that sets which values should not be
         base64 decoded.
+
+ephemeral_disk:
+---------------
+
+In order to instruct Cloud-init which disk to auto-mount. By default,
+SmartOS only supports a single ephemeral disk.
+
+The default SmartOS configuration will prepare the ephemeral disk and format
+it for you. SmartOS does not, by default, prepare the ephemeral disk for you.
+
+If you change ephemeral_disk, you should also consider changing
+the default disk formatting parameters. See
+doc/examples/cloud-config-disk-setup.txt for information on using this.
diff --git a/tests/unittests/test_datasource/test_smartos.py b/tests/unittests/test_datasource/test_smartos.py
index f53715b0..56fe811e 100644
--- a/tests/unittests/test_datasource/test_smartos.py
+++ b/tests/unittests/test_datasource/test_smartos.py
@@ -261,6 +261,41 @@ class TestSmartOSDataSource(MockerTestCase):
         self.assertEquals(MOCK_RETURNS['enable_motd_sys_info'],
                           dsrc.metadata['motd_sys_info'])
 
+    def test_default_ephemeral(self):
+        # Test to make sure that the builtin config has the ephemeral
+        # configuration.
+        dsrc = self._get_ds()
+        cfg = dsrc.get_config_obj()
+
+        ret = dsrc.get_data()
+        self.assertTrue(ret)
+
+        assert 'disk_setup' in cfg
+        assert 'fs_setup' in cfg
+        self.assertIsInstance(cfg['disk_setup'], dict)
+        self.assertIsInstance(cfg['fs_setup'], list)
+
+    def test_override_builtin_ds(self):
+        # Test to make sure that the built-in DS is overriden
+        data = {}
+        data['disk_setup'] = {'test_dev': {}}
+        data['fs_setup'] = [{'label': 'test_dev'}]
+        data['serial_device'] = '/dev/ttyS2'
+        dsrc = self._get_ds(ds_cfg=data)
+        cfg = dsrc.get_config_obj()
+
+        ret = dsrc.get_data()
+        self.assertTrue(ret)
+
+        assert 'disk_setup' in cfg
+        assert 'fs_setup' in cfg
+        self.assertIsInstance(cfg['disk_setup'], dict)
+        self.assertIsInstance(cfg['fs_setup'], list)
+        assert 'test_dev' in cfg['disk_setup']
+        assert 'test_dev' in cfg['fs_setup'][0]['label']
+
+        self.assertEquals(data['serial_device'], dsrc.seed)
+
 
 def apply_patches(patches):
     ret = []
-- 
cgit v1.2.3