azure: avoid re-running cloud-init when instance-id is byte-swapped (#84)

Azure stores the instance ID with an incorrect byte ordering for the
first three hyphen delimited parts. This results in invalid
is_new_instance checks forcing Azure datasource to recrawl the metadata
service.

When persisting instance-id from the metadata service, swap the
instance-id string byte order such that it is consistent with
that returned by dmi information. Check whether the instance-id
string is a byte-swapped match when determining correctly whether 
the Azure platform instance-id has actually changed. 

2 files changed, 40 insertions, 3 deletions

diff --git a/cloudinit/sources/DataSourceAzure.py b/cloudinit/sources/DataSourceAzure.py
index 87a848ce..24f448c5 100755
--- a/cloudinit/sources/DataSourceAzure.py
+++ b/cloudinit/sources/DataSourceAzure.py
@@ -33,7 +33,8 @@ from cloudinit.sources.helpers.azure import (
     get_boot_telemetry,
     get_system_info,
     report_diagnostic_event,
-    EphemeralDHCPv4WithReporting)
+    EphemeralDHCPv4WithReporting,
+    is_byte_swapped)
 
 LOG = logging.getLogger(__name__)
 
@@ -471,8 +472,7 @@ class DataSourceAzure(sources.DataSource):
         seed = _get_random_seed()
         if seed:
             crawled_data['metadata']['random_seed'] = seed
-        crawled_data['metadata']['instance-id'] = util.read_dmi_data(
-            'system-uuid')
+        crawled_data['metadata']['instance-id'] = self._iid()
 
         if perform_reprovision:
             LOG.info("Reporting ready to Azure after getting ReprovisionData")
@@ -558,6 +558,16 @@ class DataSourceAzure(sources.DataSource):
         # quickly (local check only) if self.instance_id is still valid
         return sources.instance_id_matches_system_uuid(self.get_instance_id())
 
+    def _iid(self, previous=None):
+        prev_iid_path = os.path.join(
+            self.paths.get_cpath('data'), 'instance-id')
+        iid = util.read_dmi_data('system-uuid')
+        if os.path.exists(prev_iid_path):
+            previous = util.load_file(prev_iid_path).strip()
+            if is_byte_swapped(previous, iid):
+                return previous
+        return iid
+
     @azure_ds_telemetry_reporter
     def setup(self, is_new_instance):
         if self._negotiated is False:
diff --git a/cloudinit/sources/helpers/azure.py b/cloudinit/sources/helpers/azure.py
index f5cdb3fd..fc760581 100755
--- a/cloudinit/sources/helpers/azure.py
+++ b/cloudinit/sources/helpers/azure.py
@@ -7,6 +7,7 @@ import re
 import socket
 import struct
 import time
+import textwrap
 
 from cloudinit.net import dhcp
 from cloudinit import stages
@@ -48,6 +49,32 @@ def azure_ds_telemetry_reporter(func):
     return impl
 
 
+def is_byte_swapped(previous_id, current_id):
+    """
+    Azure stores the instance ID with an incorrect byte ordering for the
+    first parts. This corrects the byte order such that it is consistent with
+    that returned by the metadata service.
+    """
+    if previous_id == current_id:
+        return False
+
+    def swap_bytestring(s, width=2):
+        dd = [byte for byte in textwrap.wrap(s, 2)]
+        dd.reverse()
+        return ''.join(dd)
+
+    parts = current_id.split('-')
+    swapped_id = '-'.join([
+            swap_bytestring(parts[0]),
+            swap_bytestring(parts[1]),
+            swap_bytestring(parts[2]),
+            parts[3],
+            parts[4]
+        ])
+
+    return previous_id == swapped_id
+
+
 @azure_ds_telemetry_reporter
 def get_boot_telemetry():
     """Report timestamps related to kernel initialization and systemd