Fix MIME policy failure on python version upgrade (#934)

Python 3.6 added a new `policy` attribute to `MIMEMultipart`.
MIMEMultipart may be part of the cached object pickle of a datasource.
Upgrading from an old version of python to 3.6+ will cause the
datasource to be invalid after pickle load.

This commit uses the upgrade framework to attempt to access the mime
message and fail early (thus discarding the cache) if we cannot.
Commit 78e89b03 should fix this issue more generally.

4 files changed, 622 insertions, 0 deletions

diff --git a/cloudinit/sources/__init__.py b/cloudinit/sources/__init__.py
index a07c4b4f..9d25b0ee 100644
--- a/cloudinit/sources/__init__.py
+++ b/cloudinit/sources/__init__.py
@@ -75,6 +75,10 @@ NetworkConfigSource = namedtuple('NetworkConfigSource',
                                  _NETCFG_SOURCE_NAMES)(*_NETCFG_SOURCE_NAMES)
 
 
+class DatasourceUnpickleUserDataError(Exception):
+    """Raised when userdata is unable to be unpickled due to python upgrades"""
+
+
 class DataSourceNotFoundException(Exception):
     pass
 
@@ -239,6 +243,20 @@ class DataSource(CloudInitPickleMixin, metaclass=abc.ABCMeta):
             self.vendordata2 = None
         if not hasattr(self, 'vendordata2_raw'):
             self.vendordata2_raw = None
+        if hasattr(self, 'userdata') and self.userdata is not None:
+            # If userdata stores MIME data, on < python3.6 it will be
+            # missing the 'policy' attribute that exists on >=python3.6.
+            # Calling str() on the userdata will attempt to access this
+            # policy attribute. This will raise an exception, causing
+            # the pickle load to fail, so cloud-init will discard the cache
+            try:
+                str(self.userdata)
+            except AttributeError as e:
+                LOG.debug(
+                    "Unable to unpickle datasource: %s."
+                    " Ignoring current cache.", e
+                )
+                raise DatasourceUnpickleUserDataError() from e
 
     def __str__(self):
         return type_utils.obj_name(self)
diff --git a/cloudinit/stages.py b/cloudinit/stages.py
index 3688be2e..06e0d9b1 100644
--- a/cloudinit/stages.py
+++ b/cloudinit/stages.py
@@ -1070,6 +1070,8 @@ def _pkl_load(fname):
         return None
     try:
         return pickle.loads(pickle_contents)
+    except sources.DatasourceUnpickleUserDataError:
+        return None
     except Exception:
         util.logexc(LOG, "Failed loading pickled blob from %s", fname)
         return None
diff --git a/tests/integration_tests/assets/trusty_with_mime.pkl b/tests/integration_tests/assets/trusty_with_mime.pkl
new file mode 100644
index 00000000..a4089ecf
--- /dev/null
+++ b/tests/integration_tests/assets/trusty_with_mime.pkl
@@ -0,0 +1,572 @@
+ccopy_reg
+_reconstructor
+p1
+(ccloudinit.sources.DataSourceNoCloud
+DataSourceNoCloudNet
+p2
+c__builtin__
+object
+p3
+NtRp4
+(dp5
+S'paths'
+p6
+g1
+(ccloudinit.helpers
+Paths
+p7
+g3
+NtRp8
+(dp9
+S'lookups'
+p10
+(dp11
+S'cloud_config'
+p12
+S'cloud-config.txt'
+p13
+sS'userdata'
+p14
+S'user-data.txt.i'
+p15
+sS'vendordata'
+p16
+S'vendor-data.txt.i'
+p17
+sS'userdata_raw'
+p18
+S'user-data.txt'
+p19
+sS'boothooks'
+p20
+g20
+sS'scripts'
+p21
+g21
+sS'sem'
+p22
+g22
+sS'data'
+p23
+g23
+sS'vendor_scripts'
+p24
+S'scripts/vendor'
+p25
+sS'handlers'
+p26
+g26
+sS'obj_pkl'
+p27
+S'obj.pkl'
+p28
+sS'vendordata_raw'
+p29
+S'vendor-data.txt'
+p30
+sS'vendor_cloud_config'
+p31
+S'vendor-cloud-config.txt'
+p32
+ssS'template_tpl'
+p33
+S'/etc/cloud/templates/%s.tmpl'
+p34
+sS'cfgs'
+p35
+(dp36
+S'cloud_dir'
+p37
+S'/var/lib/cloud/'
+p38
+sS'templates_dir'
+p39
+S'/etc/cloud/templates/'
+p40
+sS'upstart_dir'
+p41
+S'/etc/init/'
+p42
+ssS'cloud_dir'
+p43
+g38
+sS'datasource'
+p44
+NsS'upstart_conf_d'
+p45
+g42
+sS'boot_finished'
+p46
+S'/var/lib/cloud/instance/boot-finished'
+p47
+sS'instance_link'
+p48
+S'/var/lib/cloud/instance'
+p49
+sS'seed_dir'
+p50
+S'/var/lib/cloud/seed'
+p51
+sbsS'supported_seed_starts'
+p52
+(S'http://'
+p53
+S'https://'
+p54
+S'ftp://'
+p55
+tp56
+sS'sys_cfg'
+p57
+(dp58
+S'output'
+p59
+(dp60
+S'all'
+p61
+S'| tee -a /var/log/cloud-init-output.log'
+p62
+ssS'users'
+p63
+(lp64
+S'default'
+p65
+asS'def_log_file'
+p66
+S'/var/log/cloud-init.log'
+p67
+sS'cloud_final_modules'
+p68
+(lp69
+S'rightscale_userdata'
+p70
+aS'scripts-vendor'
+p71
+aS'scripts-per-once'
+p72
+aS'scripts-per-boot'
+p73
+aS'scripts-per-instance'
+p74
+aS'scripts-user'
+p75
+aS'ssh-authkey-fingerprints'
+p76
+aS'keys-to-console'
+p77
+aS'phone-home'
+p78
+aS'final-message'
+p79
+aS'power-state-change'
+p80
+asS'disable_root'
+p81
+I01
+sS'syslog_fix_perms'
+p82
+S'syslog:adm'
+p83
+sS'log_cfgs'
+p84
+(lp85
+(lp86
+S'[loggers]\nkeys=root,cloudinit\n\n[handlers]\nkeys=consoleHandler,cloudLogHandler\n\n[formatters]\nkeys=simpleFormatter,arg0Formatter\n\n[logger_root]\nlevel=DEBUG\nhandlers=consoleHandler,cloudLogHandler\n\n[logger_cloudinit]\nlevel=DEBUG\nqualname=cloudinit\nhandlers=\npropagate=1\n\n[handler_consoleHandler]\nclass=StreamHandler\nlevel=WARNING\nformatter=arg0Formatter\nargs=(sys.stderr,)\n\n[formatter_arg0Formatter]\nformat=%(asctime)s - %(filename)s[%(levelname)s]: %(message)s\n\n[formatter_simpleFormatter]\nformat=[CLOUDINIT] %(filename)s[%(levelname)s]: %(message)s\n'
+p87
+aS'[handler_cloudLogHandler]\nclass=handlers.SysLogHandler\nlevel=DEBUG\nformatter=simpleFormatter\nargs=("/dev/log", handlers.SysLogHandler.LOG_USER)\n'
+p88
+aa(lp89
+g87
+aS"[handler_cloudLogHandler]\nclass=FileHandler\nlevel=DEBUG\nformatter=arg0Formatter\nargs=('/var/log/cloud-init.log',)\n"
+p90
+aasS'cloud_init_modules'
+p91
+(lp92
+S'migrator'
+p93
+aS'seed_random'
+p94
+aS'bootcmd'
+p95
+aS'write-files'
+p96
+aS'growpart'
+p97
+aS'resizefs'
+p98
+aS'set_hostname'
+p99
+aS'update_hostname'
+p100
+aS'update_etc_hosts'
+p101
+aS'ca-certs'
+p102
+aS'rsyslog'
+p103
+aS'users-groups'
+p104
+aS'ssh'
+p105
+asS'preserve_hostname'
+p106
+I00
+sS'_log'
+p107
+(lp108
+g87
+ag90
+ag88
+asS'datasource_list'
+p109
+(lp110
+S'NoCloud'
+p111
+aS'ConfigDrive'
+p112
+aS'OpenNebula'
+p113
+aS'Azure'
+p114
+aS'AltCloud'
+p115
+aS'OVF'
+p116
+aS'MAAS'
+p117
+aS'GCE'
+p118
+aS'OpenStack'
+p119
+aS'CloudSigma'
+p120
+aS'Ec2'
+p121
+aS'CloudStack'
+p122
+aS'SmartOS'
+p123
+aS'None'
+p124
+asS'vendor_data'
+p125
+(dp126
+S'prefix'
+p127
+(lp128
+sS'enabled'
+p129
+I01
+ssS'cloud_config_modules'
+p130
+(lp131
+S'emit_upstart'
+p132
+aS'disk_setup'
+p133
+aS'mounts'
+p134
+aS'ssh-import-id'
+p135
+aS'locale'
+p136
+aS'set-passwords'
+p137
+aS'grub-dpkg'
+p138
+aS'apt-pipelining'
+p139
+aS'apt-configure'
+p140
+aS'package-update-upgrade-install'
+p141
+aS'landscape'
+p142
+aS'timezone'
+p143
+aS'puppet'
+p144
+aS'chef'
+p145
+aS'salt-minion'
+p146
+aS'mcollective'
+p147
+aS'disable-ec2-metadata'
+p148
+aS'runcmd'
+p149
+aS'byobu'
+p150
+assg14
+(iemail.mime.multipart
+MIMEMultipart
+p151
+(dp152
+S'_headers'
+p153
+(lp154
+(S'Content-Type'
+p155
+S'multipart/mixed; boundary="===============4291038100093149247=="'
+tp156
+a(S'MIME-Version'
+p157
+S'1.0'
+p158
+tp159
+a(S'Number-Attachments'
+p160
+S'1'
+tp161
+asS'_payload'
+p162
+(lp163
+(iemail.mime.base
+MIMEBase
+p164
+(dp165
+g153
+(lp166
+(g157
+g158
+tp167
+a(S'Content-Type'
+p168
+S'text/x-not-multipart'
+tp169
+a(S'Content-Disposition'
+p170
+S'attachment; filename="part-001"'
+tp171
+asg162
+S''
+sS'_charset'
+p172
+NsS'_default_type'
+p173
+S'text/plain'
+p174
+sS'preamble'
+p175
+NsS'defects'
+p176
+(lp177
+sS'_unixfrom'
+p178
+NsS'epilogue'
+p179
+Nsbasg172
+Nsg173
+g174
+sg175
+Nsg176
+(lp180
+sg178
+Nsg179
+Nsbsg16
+S'#cloud-config\n{}\n\n'
+p181
+sg18
+S'Content-Type: multipart/mixed; boundary="===============1378281702283945349=="\nMIME-Version: 1.0\n\n--===============1378281702283945349==\nContent-Type: text/x-shellscript; charset="utf-8"\nMIME-Version: 1.0\nContent-Transfer-Encoding: base64\nContent-Disposition: attachment; filename="script1.sh"\n\nIyEvYmluL3NoCgplY2hvICdoaScgPiAvdmFyL3RtcC9oaQo=\n\n--===============1378281702283945349==\nContent-Type: text/x-shellscript; charset="utf-8"\nMIME-Version: 1.0\nContent-Transfer-Encoding: base64\nContent-Disposition: attachment; filename="script2.sh"\n\nIyEvYmluL2Jhc2gKCmVjaG8gJ2hpMicgPiAvdmFyL3RtcC9oaTIK\n\n--===============1378281702283945349==--\n\n#cloud-config\n# final_message: |\n#   This is my final message!\n#   $version\n#   $timestamp\n#   $datasource\n#   $uptime\n# updates:\n#   network:\n#     when: [\'hotplug\']\n'
+p182
+sg29
+NsS'dsmode'
+p183
+S'net'
+p184
+sS'seed'
+p185
+S'/var/lib/cloud/seed/nocloud-net'
+p186
+sS'cmdline_id'
+p187
+S'ds=nocloud-net'
+p188
+sS'ud_proc'
+p189
+g1
+(ccloudinit.user_data
+UserDataProcessor
+p190
+g3
+NtRp191
+(dp192
+g6
+g8
+sS'ssl_details'
+p193
+(dp194
+sbsg50
+g186
+sS'ds_cfg'
+p195
+(dp196
+sS'distro'
+p197
+g1
+(ccloudinit.distros.ubuntu
+Distro
+p198
+g3
+NtRp199
+(dp200
+S'osfamily'
+p201
+S'debian'
+p202
+sS'_paths'
+p203
+g8
+sS'name'
+p204
+S'ubuntu'
+p205
+sS'_runner'
+p206
+g1
+(ccloudinit.helpers
+Runners
+p207
+g3
+NtRp208
+(dp209
+g6
+g8
+sS'sems'
+p210
+(dp211
+sbsS'_cfg'
+p212
+(dp213
+S'paths'
+p214
+(dp215
+g37
+g38
+sg39
+g40
+sg41
+g42
+ssS'default_user'
+p216
+(dp217
+S'shell'
+p218
+S'/bin/bash'
+p219
+sS'name'
+p220
+S'ubuntu'
+p221
+sS'sudo'
+p222
+(lp223
+S'ALL=(ALL) NOPASSWD:ALL'
+p224
+asS'lock_passwd'
+p225
+I01
+sS'gecos'
+p226
+S'Ubuntu'
+p227
+sS'groups'
+p228
+(lp229
+S'adm'
+p230
+aS'audio'
+p231
+aS'cdrom'
+p232
+aS'dialout'
+p233
+aS'dip'
+p234
+aS'floppy'
+p235
+aS'netdev'
+p236
+aS'plugdev'
+p237
+aS'sudo'
+p238
+aS'video'
+p239
+assS'package_mirrors'
+p240
+(lp241
+(dp242
+S'arches'
+p243
+(lp244
+S'i386'
+p245
+aS'amd64'
+p246
+asS'failsafe'
+p247
+(dp248
+S'security'
+p249
+S'http://security.ubuntu.com/ubuntu'
+p250
+sS'primary'
+p251
+S'http://archive.ubuntu.com/ubuntu'
+p252
+ssS'search'
+p253
+(dp254
+S'security'
+p255
+(lp256
+sS'primary'
+p257
+(lp258
+S'http://%(ec2_region)s.ec2.archive.ubuntu.com/ubuntu/'
+p259
+aS'http://%(availability_zone)s.clouds.archive.ubuntu.com/ubuntu/'
+p260
+aS'http://%(region)s.clouds.archive.ubuntu.com/ubuntu/'
+p261
+assa(dp262
+S'arches'
+p263
+(lp264
+S'armhf'
+p265
+aS'armel'
+p266
+aS'default'
+p267
+asS'failsafe'
+p268
+(dp269
+S'security'
+p270
+S'http://ports.ubuntu.com/ubuntu-ports'
+p271
+sS'primary'
+p272
+S'http://ports.ubuntu.com/ubuntu-ports'
+p273
+ssasS'ssh_svcname'
+p274
+S'ssh'
+p275
+ssbsS'metadata'
+p276
+(dp277
+g183
+g184
+sS'local-hostname'
+p278
+S'me'
+p279
+sS'instance-id'
+p280
+S'me'
+p281
+ssb.
\ No newline at end of file
diff --git a/tests/integration_tests/modules/test_persistence.py b/tests/integration_tests/modules/test_persistence.py
new file mode 100644
index 00000000..00fdeaea
--- /dev/null
+++ b/tests/integration_tests/modules/test_persistence.py
@@ -0,0 +1,30 @@
+# This file is part of cloud-init. See LICENSE file for license information.
+"""Test the behavior of loading/discarding pickle data"""
+from pathlib import Path
+
+import pytest
+
+from tests.integration_tests.instances import IntegrationInstance
+from tests.integration_tests.util import (
+    ASSETS_DIR,
+    verify_ordered_items_in_text,
+)
+
+
+PICKLE_PATH = Path('/var/lib/cloud/instance/obj.pkl')
+TEST_PICKLE = ASSETS_DIR / 'trusty_with_mime.pkl'
+
+
+@pytest.mark.lxd_container
+def test_log_message_on_missing_version_file(client: IntegrationInstance):
+    client.push_file(TEST_PICKLE, PICKLE_PATH)
+    client.restart()
+    assert client.execute('cloud-init status --wait').ok
+    log = client.read_from_file('/var/log/cloud-init.log')
+    verify_ordered_items_in_text([
+        "Unable to unpickle datasource: 'MIMEMultipart' object has no "
+        "attribute 'policy'. Ignoring current cache.",
+        'no cache found',
+        'Searching for local data source',
+        'SUCCESS: found local data from DataSourceNoCloud'
+    ], log)