311 lines
13 KiB
Python
311 lines
13 KiB
Python
# (c) 2016-2018, Matt Davis <mdavis@ansible.com>
|
|
# (c) 2018, Sam Doran <sdoran@redhat.com>
|
|
# GNU General Public License v3.0+ (see COPYING or https://www.gnu.org/licenses/gpl-3.0.txt)
|
|
|
|
from __future__ import (absolute_import, division, print_function)
|
|
__metaclass__ = type
|
|
|
|
import random
|
|
import time
|
|
|
|
from datetime import datetime, timedelta
|
|
|
|
from ansible.errors import AnsibleError
|
|
from ansible.plugins.action import ActionBase
|
|
from ansible.module_utils._text import to_native, to_text
|
|
|
|
|
|
try:
|
|
from __main__ import display
|
|
except ImportError:
|
|
from ansible.utils.display import Display
|
|
display = Display()
|
|
|
|
|
|
class TimedOutException(Exception):
|
|
pass
|
|
|
|
|
|
class ActionModule(ActionBase):
|
|
TRANSFERS_FILES = False
|
|
_VALID_ARGS = frozenset(('connect_timeout', 'msg', 'post_reboot_delay', 'pre_reboot_delay', 'test_command'))
|
|
|
|
DEFAULT_REBOOT_TIMEOUT = 600
|
|
DEFAULT_CONNECT_TIMEOUT = None
|
|
DEFAULT_PRE_REBOOT_DELAY = 0
|
|
DEFAULT_POST_REBOOT_DELAY = 0
|
|
DEFAULT_TEST_COMMAND = 'whoami'
|
|
DEFAULT_BOOT_TIME_COMMAND = 'who -b'
|
|
DEFAULT_REBOOT_MESSAGE = 'Reboot initiated by Ansible'
|
|
DEFAULT_SHUTDOWN_COMMAND = 'shutdown'
|
|
DEFAULT_SUDOABLE = True
|
|
|
|
DEPRECATED_ARGS = {}
|
|
|
|
SHUTDOWN_COMMANDS = {
|
|
'linux': DEFAULT_SHUTDOWN_COMMAND,
|
|
'freebsd': DEFAULT_SHUTDOWN_COMMAND,
|
|
'sunos': '/usr/sbin/shutdown',
|
|
'darwin': '/sbin/shutdown',
|
|
}
|
|
|
|
SHUTDOWN_COMMAND_ARGS = {
|
|
'linux': '-r {delay_min} "{message}"',
|
|
'freebsd': '-r +{delay_sec}s "{message}"',
|
|
'sunos': '-y -g {delay_sec} -r "{message}"',
|
|
'darwin': '-r +{delay_min_macos} "{message}"'
|
|
}
|
|
|
|
def __init__(self, *args, **kwargs):
|
|
super(ActionModule, self).__init__(*args, **kwargs)
|
|
|
|
self._original_connection_timeout = None
|
|
self._previous_boot_time = None
|
|
|
|
def deprecated_args(self):
|
|
for arg, version in self.DEPRECATED_ARGS.items():
|
|
if self._task.args.get(arg) is not None:
|
|
display.warning("Since Ansible %s, %s is no longer a valid option for %s" % (version, arg, self._task.action))
|
|
|
|
def construct_command(self):
|
|
# Determine the system distribution in order to use the correct shutdown command arguments
|
|
uname_result = self._low_level_execute_command('uname')
|
|
distribution = uname_result['stdout'].strip().lower()
|
|
|
|
shutdown_command = self.SHUTDOWN_COMMANDS.get(distribution, self.SHUTDOWN_COMMANDS['linux'])
|
|
shutdown_command_args = self.SHUTDOWN_COMMAND_ARGS.get(distribution, self.SHUTDOWN_COMMAND_ARGS['linux'])
|
|
|
|
pre_reboot_delay = int(self._task.args.get('pre_reboot_delay', self.DEFAULT_PRE_REBOOT_DELAY))
|
|
if pre_reboot_delay < 0:
|
|
pre_reboot_delay = 0
|
|
|
|
# Convert seconds to minutes for Linux. If less that 60, set it to 0 except for macOS which will
|
|
# sever the connection too quickly if set to 0, so set that to 1.
|
|
# We could simplify this by setting them both to 1, but I think of all the time that
|
|
# people will lose waiting for that extra 1 minute delay and want to give them their
|
|
# lives back.
|
|
delay_min = pre_reboot_delay // 60
|
|
delay_min_macos = delay_min | 1
|
|
msg = self._task.args.get('msg', self.DEFAULT_REBOOT_MESSAGE)
|
|
|
|
shutdown_command_args = shutdown_command_args.format(delay_sec=pre_reboot_delay, delay_min=delay_min, delay_min_macos=delay_min_macos, message=msg)
|
|
|
|
reboot_command = '%s %s' % (shutdown_command, shutdown_command_args)
|
|
return reboot_command
|
|
|
|
def get_system_boot_time(self):
|
|
stdout = b''
|
|
stderr = b''
|
|
command_result = self._low_level_execute_command(self.DEFAULT_BOOT_TIME_COMMAND, sudoable=self.DEFAULT_SUDOABLE)
|
|
|
|
# For single board computers, e.g., Raspberry Pi, that lack a real time clock and are using fake-hwclock
|
|
# launched by systemd, the update of utmp/wtmp is not done correctly.
|
|
# Fall back to using uptime -s for those systems.
|
|
# https://github.com/systemd/systemd/issues/6057
|
|
if '1970-01-01 00:00' in command_result['stdout']:
|
|
stdout += command_result['stdout']
|
|
stderr += command_result['stderr']
|
|
command_result = self._low_level_execute_command('uptime -s', sudoable=self.DEFAULT_SUDOABLE)
|
|
|
|
# This is a last resort for bare Linux systems (e.g. OpenELEC) where 'who -b' or 'uptime -s' are not supported.
|
|
# Other options like parsing /proc/uptime or default uptime output are less reliable than this
|
|
if command_result['rc'] != 0:
|
|
stdout += command_result['stdout']
|
|
stderr += command_result['stderr']
|
|
command_result = self._low_level_execute_command('cat /proc/sys/kernel/random/boot_id', sudoable=self.DEFAULT_SUDOABLE)
|
|
|
|
if command_result['rc'] != 0:
|
|
stdout += command_result['stdout']
|
|
stderr += command_result['stderr']
|
|
raise AnsibleError("%s: failed to get host boot time info, rc: %d, stdout: %s, stderr: %s"
|
|
% (self._task.action, command_result['rc'], to_native(stdout), to_native(stderr)))
|
|
|
|
return command_result['stdout'].strip()
|
|
|
|
def check_boot_time(self):
|
|
display.vvv("%s: attempting to get system boot time" % self._task.action)
|
|
connect_timeout = self._task.args.get('connect_timeout', self._task.args.get('connect_timeout_sec', self.DEFAULT_CONNECT_TIMEOUT))
|
|
|
|
# override connection timeout from defaults to custom value
|
|
if connect_timeout:
|
|
try:
|
|
self._connection.set_option("connection_timeout", connect_timeout)
|
|
self._connection.reset()
|
|
except AttributeError:
|
|
display.warning("Connection plugin does not allow the connection timeout to be overridden")
|
|
|
|
# try and get boot time
|
|
try:
|
|
current_boot_time = self.get_system_boot_time()
|
|
except Exception as e:
|
|
raise e
|
|
|
|
# FreeBSD returns an empty string immediately before reboot so adding a length
|
|
# check to prevent prematurely assuming system has rebooted
|
|
if len(current_boot_time) == 0 or current_boot_time == self._previous_boot_time:
|
|
raise Exception("boot time has not changed")
|
|
|
|
def run_test_command(self, **kwargs):
|
|
test_command = self._task.args.get('test_command', self.DEFAULT_TEST_COMMAND)
|
|
display.vvv("%s: attempting post-reboot test command '%s'" % (self._task.action, test_command))
|
|
try:
|
|
command_result = self._low_level_execute_command(test_command, sudoable=self.DEFAULT_SUDOABLE)
|
|
except Exception:
|
|
# may need to reset the connection in case another reboot occurred
|
|
# which has invalidated our connection
|
|
try:
|
|
self._connection.reset()
|
|
except AttributeError:
|
|
pass
|
|
raise
|
|
|
|
result = {}
|
|
if command_result['rc'] != 0:
|
|
result['failed'] = True
|
|
result['msg'] = 'test command failed: %s %s' % (to_native(command_result['stderr'], to_native(command_result['stdout'])))
|
|
else:
|
|
result['msg'] = to_native(command_result['stdout'])
|
|
|
|
return result
|
|
|
|
def do_until_success_or_timeout(self, action, reboot_timeout, action_desc):
|
|
max_end_time = datetime.utcnow() + timedelta(seconds=reboot_timeout)
|
|
|
|
fail_count = 0
|
|
max_fail_sleep = 12
|
|
|
|
while datetime.utcnow() < max_end_time:
|
|
try:
|
|
action()
|
|
if action_desc:
|
|
display.debug('%s: %s success' % (self._task.action, action_desc))
|
|
return
|
|
except Exception as e:
|
|
# Use exponential backoff with a max timout, plus a little bit of randomness
|
|
random_int = random.randint(0, 1000) / 1000
|
|
fail_sleep = 2 ** fail_count + random_int
|
|
if fail_sleep > max_fail_sleep:
|
|
|
|
fail_sleep = max_fail_sleep + random_int
|
|
if action_desc:
|
|
display.debug("{0}: {1} fail '{2}', retrying in {3:.4} seconds...".format(self._task.action, action_desc, to_text(e), fail_sleep))
|
|
fail_count += 1
|
|
time.sleep(fail_sleep)
|
|
|
|
raise TimedOutException('Timed out waiting for %s' % (action_desc))
|
|
|
|
def perform_reboot(self):
|
|
display.debug("%s: rebooting server" % self._task.action)
|
|
|
|
remote_command = self.construct_command()
|
|
reboot_result = self._low_level_execute_command(remote_command, sudoable=self.DEFAULT_SUDOABLE)
|
|
result = {}
|
|
result['start'] = datetime.utcnow()
|
|
|
|
if reboot_result['rc'] != 0:
|
|
result['failed'] = True
|
|
result['rebooted'] = False
|
|
result['msg'] = "Shutdown command failed. Error was %s, %s" % (
|
|
to_native(reboot_result['stdout'].strip()), to_native(reboot_result['stderr'].strip()))
|
|
return result
|
|
|
|
result['failed'] = False
|
|
|
|
# attempt to store the original connection_timeout option var so it can be reset after
|
|
self._original_connection_timeout = None
|
|
try:
|
|
self._original_connection_timeout = self._connection.get_option('connection_timeout')
|
|
except AnsibleError:
|
|
display.debug("%s: connect_timeout connection option has not been set" % self._task.action)
|
|
|
|
post_reboot_delay = int(self._task.args.get('post_reboot_delay', self._task.args.get('post_reboot_delay_sec', self.DEFAULT_POST_REBOOT_DELAY)))
|
|
if post_reboot_delay < 0:
|
|
post_reboot_delay = 0
|
|
|
|
if post_reboot_delay != 0:
|
|
display.vvv("%s: waiting an additional %d seconds" % (self._task.action, post_reboot_delay))
|
|
time.sleep(post_reboot_delay)
|
|
|
|
return result
|
|
|
|
def validate_reboot(self):
|
|
display.debug('%s: Validating reboot' % self._task.action)
|
|
result = {}
|
|
|
|
try:
|
|
# keep on checking system boot_time with short connection responses
|
|
reboot_timeout = int(self._task.args.get('reboot_timeout', self._task.args.get('reboot_timeout_sec', self.DEFAULT_REBOOT_TIMEOUT)))
|
|
connect_timeout = self._task.args.get('connect_timeout', self._task.args.get('connect_timeout_sec', self.DEFAULT_CONNECT_TIMEOUT))
|
|
self.do_until_success_or_timeout(self.check_boot_time, reboot_timeout, action_desc="boot_time check")
|
|
|
|
if connect_timeout:
|
|
# reset the connection to clear the custom connection timeout
|
|
try:
|
|
self._connection.set_option("connection_timeout", connect_timeout)
|
|
self._connection.reset()
|
|
except (AnsibleError, AttributeError) as e:
|
|
display.debug("Failed to reset connection_timeout back to default: %s" % to_text(e))
|
|
|
|
# finally run test command to ensure everything is working
|
|
# FUTURE: add a stability check (system must remain up for N seconds) to deal with self-multi-reboot updates
|
|
self.do_until_success_or_timeout(self.run_test_command, reboot_timeout, action_desc="post-reboot test command")
|
|
|
|
result['rebooted'] = True
|
|
result['changed'] = True
|
|
|
|
except TimedOutException as toex:
|
|
result['failed'] = True
|
|
result['rebooted'] = True
|
|
result['msg'] = to_text(toex)
|
|
return result
|
|
|
|
return result
|
|
|
|
def run(self, tmp=None, task_vars=None):
|
|
self._supports_check_mode = True
|
|
self._supports_async = True
|
|
|
|
# If running with local connection, fail so we don't reboot ourself
|
|
if self._connection.transport == 'local':
|
|
msg = 'Running {0} with local connection would reboot the control node.'.format(self._task.action)
|
|
return dict(changed=False, elapsed=0, rebooted=False, failed=True, msg=msg)
|
|
|
|
if self._play_context.check_mode:
|
|
return dict(changed=True, elapsed=0, rebooted=True)
|
|
|
|
if task_vars is None:
|
|
task_vars = dict()
|
|
|
|
self.deprecated_args()
|
|
|
|
result = super(ActionModule, self).run(tmp, task_vars)
|
|
|
|
if result.get('skipped', False) or result.get('failed', False):
|
|
return result
|
|
|
|
# Get current boot time
|
|
try:
|
|
self._previous_boot_time = self.get_system_boot_time()
|
|
except Exception as e:
|
|
result['failed'] = True
|
|
result['reboot'] = False
|
|
result['msg'] = to_text(e)
|
|
return result
|
|
|
|
# Initiate reboot
|
|
reboot_result = self.perform_reboot()
|
|
|
|
if reboot_result['failed']:
|
|
result = reboot_result
|
|
elapsed = datetime.utcnow() - reboot_result['start']
|
|
result['elapsed'] = elapsed.seconds
|
|
return result
|
|
|
|
# Make sure reboot was successful
|
|
result = self.validate_reboot()
|
|
|
|
elapsed = datetime.utcnow() - reboot_result['start']
|
|
result['elapsed'] = elapsed.seconds
|
|
|
|
return result
|