# -*- coding: utf-8 -*-
#
# 2010-2015 Nico Schottelius (nico-cdist at schottelius.org)
# 2013-2017 Steven Armstrong (steven-cdist at armstrong.cc)
# 2016-2017 Darko Poljak (darko.poljak at gmail.com)
#
# This file is part of cdist.
#
# cdist is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# cdist is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with cdist. If not, see <http://www.gnu.org/licenses/>.
#
#
import logging
import os
import sys
import time
import itertools
import tempfile
import multiprocessing
import atexit
import shutil
import socket
from cdist.mputil import mp_pool_run, mp_sig_handler
from cdist import core, inventory
from cdist.util.remoteutil import inspect_ssh_mux_opts
import cdist
import cdist.hostsource
import cdist.exec.local
import cdist.exec.remote
import cdist.util.ipaddr as ipaddr
import cdist.configuration
def graph_check_cycle(graph):
# Start from each node in the graph and check for cycle starting from it.
for node in graph:
# Cycle path.
path = [node]
has_cycle = _graph_dfs_cycle(graph, node, path)
if has_cycle:
return has_cycle, path
return False, None
def _graph_dfs_cycle(graph, node, path):
for neighbour in graph.get(node, ()):
# If node is already in path then this is cycle.
if neighbour in path:
path.append(neighbour)
return True
path.append(neighbour)
rv = _graph_dfs_cycle(graph, neighbour, path)
if rv:
return True
# Remove last item from list - neighbour whose DFS path we have have
# just checked.
del path[-1]
return False
class Config:
"""Cdist main class to hold arbitrary data"""
# list of paths (files and/or directories) that will be removed on finish
_paths_for_removal = []
@classmethod
def _register_path_for_removal(cls, path):
cls._paths_for_removal.append(path)
@classmethod
def _remove_paths(cls):
while cls._paths_for_removal:
path = cls._paths_for_removal.pop()
if os.path.isfile(path):
os.remove(path)
else:
shutil.rmtree(path)
def __init__(self, local, remote, dry_run=False, jobs=None,
cleanup_cmds=None, remove_remote_files_dirs=False):
self.local = local
self.remote = remote
self._open_logger()
self.dry_run = dry_run
self.jobs = jobs
if cleanup_cmds:
self.cleanup_cmds = cleanup_cmds
else:
self.cleanup_cmds = []
self.remove_remote_files_dirs = remove_remote_files_dirs
self.explorer = core.Explorer(self.local.target_host, self.local,
self.remote, jobs=self.jobs,
dry_run=self.dry_run)
self.manifest = core.Manifest(self.local.target_host, self.local,
dry_run=self.dry_run)
self.code = core.Code(self.local.target_host, self.local, self.remote,
dry_run=self.dry_run)
def _init_files_dirs(self):
"""Prepare files and directories for the run"""
self.local.create_files_dirs()
self.remote.create_files_dirs()
def _remove_remote_files_dirs(self):
"""Remove remote files and directories for the run"""
self.remote.remove_files_dirs()
def _remove_files_dirs(self):
"""Remove files and directories for the run"""
if self.remove_remote_files_dirs:
self._remove_remote_files_dirs()
self.manifest.cleanup()
@staticmethod
def hosts(source):
try:
yield from cdist.hostsource.HostSource(source)()
except (IOError, OSError, UnicodeError) as e:
raise cdist.Error(
"Error reading hosts from \'{}\': {}".format(
source, e))
@staticmethod
def construct_remote_exec_copy_patterns(args):
# default remote cmd patterns
args.remote_cmds_cleanup_pattern = ""
args.remote_exec_pattern = None
args.remote_copy_pattern = None
# Determine forcing IPv4/IPv6 options if any, only for
# default remote commands.
if args.force_ipv:
force_addr_opt = " -{}".format(args.force_ipv)
else:
force_addr_opt = ""
args_dict = vars(args)
# if remote-exec and/or remote-copy args are None then user
# didn't specify command line options nor env vars:
# inspect multiplexing options for default cdist.REMOTE_COPY/EXEC
if (args_dict['remote_copy'] is None or
args_dict['remote_exec'] is None):
mux_opts = inspect_ssh_mux_opts()
if args_dict['remote_exec'] is None:
args.remote_exec_pattern = (cdist.REMOTE_EXEC +
force_addr_opt + mux_opts)
if args_dict['remote_copy'] is None:
args.remote_copy_pattern = (cdist.REMOTE_COPY +
force_addr_opt + mux_opts)
if mux_opts:
cleanup_pattern = cdist.REMOTE_CMDS_CLEANUP_PATTERN
else:
cleanup_pattern = ""
args.remote_cmds_cleanup_pattern = cleanup_pattern
@classmethod
def _check_and_prepare_args(cls, args):
if args.manifest == '-' and args.hostfile == '-':
raise cdist.Error(("Cannot read both, manifest and host file, "
"from stdin"))
if not (args.hostfile or args.host):
if args.tag or args.all_tagged_hosts:
raise cdist.Error(("Target host tag(s) missing"))
else:
raise cdist.Error(("Target host(s) missing"))
if args.manifest == '-':
# read initial manifest from stdin
try:
handle, initial_manifest_temp_path = tempfile.mkstemp(
prefix='cdist.stdin.')
with os.fdopen(handle, 'w') as fd:
fd.write(sys.stdin.read())
except (IOError, OSError) as e:
raise cdist.Error(("Creating tempfile for stdin data "
"failed: {}").format(e))
args.manifest = initial_manifest_temp_path
atexit.register(lambda: os.remove(initial_manifest_temp_path))
@classmethod
def commandline(cls, args):
"""Configure remote system"""
if (args.parallel and args.parallel != 1) or args.jobs:
if args.timestamp:
cdist.log.setupTimestampingParallelLogging()
else:
cdist.log.setupParallelLogging()
elif args.timestamp:
cdist.log.setupTimestampingLogging()
log = logging.getLogger("config")
# No new child process if only one host at a time.
if args.parallel == 1:
log.debug("Only 1 parallel process, doing it sequentially")
args.parallel = 0
if args.parallel:
import signal
signal.signal(signal.SIGTERM, mp_sig_handler)
signal.signal(signal.SIGHUP, mp_sig_handler)
cls._check_and_prepare_args(args)
failed_hosts = []
time_start = time.time()
cls.construct_remote_exec_copy_patterns(args)
base_root_path = cls.create_base_root_path(args.out_path)
hostcnt = 0
cfg = cdist.configuration.Configuration(args)
configuration = cfg.get_config(section='GLOBAL')
if args.tag or args.all_tagged_hosts:
inventory.determine_default_inventory_dir(args, configuration)
if args.all_tagged_hosts:
inv_list = inventory.InventoryList(
hosts=None, istag=True, hostfile=None,
db_basedir=args.inventory_dir)
else:
inv_list = inventory.InventoryList(
hosts=args.host, istag=True, hostfile=args.hostfile,
db_basedir=args.inventory_dir,
has_all_tags=args.has_all_tags)
it = inv_list.entries()
else:
it = itertools.chain(cls.hosts(args.host),
cls.hosts(args.hostfile))
process_args = []
if args.parallel:
log.trace("Processing hosts in parallel")
else:
log.trace("Processing hosts sequentially")
for entry in it:
if isinstance(entry, tuple):
# if configuring by specified tags
host = entry[0]
host_tags = entry[1]
else:
# if configuring by host then check inventory for tags
host = entry
inventory.determine_default_inventory_dir(args, configuration)
inv_list = inventory.InventoryList(
hosts=(host,), db_basedir=args.inventory_dir)
inv = tuple(inv_list.entries())
if inv:
# host is present in inventory and has tags
host_tags = inv[0][1]
else:
# host is not present in inventory or has no tags
host_tags = None
host_base_path, hostdir = cls.create_host_base_dirs(
host, base_root_path)
log.debug("Base root path for target host \"%s\" is \"%s\"",
host, host_base_path)
hostcnt += 1
if args.parallel:
pargs = (host, host_tags, host_base_path, hostdir, args, True,
configuration)
log.trace("Args for multiprocessing operation for host %s: %s",
host, pargs)
process_args.append(pargs)
else:
try:
cls.onehost(host, host_tags, host_base_path, hostdir,
args, parallel=False,
configuration=configuration)
except cdist.Error:
failed_hosts.append(host)
if args.parallel and len(process_args) == 1:
log.debug("Only 1 host for parallel processing, doing it "
"sequentially")
try:
cls.onehost(*process_args[0])
except cdist.Error:
failed_hosts.append(host)
elif args.parallel:
log.trace("Multiprocessing start method is %s",
multiprocessing.get_start_method())
log.trace("Starting multiprocessing Pool for %d parallel host"
" operation", args.parallel)
results = mp_pool_run(cls.onehost,
process_args,
jobs=args.parallel)
log.trace(("Multiprocessing for parallel host operation "
"finished"))
log.trace("Multiprocessing for parallel host operation "
"results: %s", results)
failed_hosts = [host for host, result in results if not result]
time_end = time.time()
log.verbose("Total processing time for %s host(s): %s", hostcnt,
(time_end - time_start))
if len(failed_hosts) > 0:
raise cdist.Error("Failed to configure the following hosts: " +
" ".join(failed_hosts))
elif not args.out_path:
# If tmp out path created then remove it, but only if no failed
# hosts.
shutil.rmtree(base_root_path)
@classmethod
def _resolve_ssh_control_path(cls):
base_path = tempfile.mkdtemp()
cls._register_path_for_removal(base_path)
control_path = os.path.join(base_path, "s")
return control_path
@classmethod
def _resolve_remote_cmds(cls, args):
if (args.remote_exec_pattern or
args.remote_copy_pattern or
args.remote_cmds_cleanup_pattern):
control_path = cls._resolve_ssh_control_path()
# If we constructed patterns for remote commands then there is
# placeholder for ssh ControlPath, format it and we have unique
# ControlPath for each host.
#
# If not then use args.remote_exec/copy that user specified.
if args.remote_exec_pattern:
remote_exec = args.remote_exec_pattern.format(control_path)
else:
remote_exec = args.remote_exec
if args.remote_copy_pattern:
remote_copy = args.remote_copy_pattern.format(control_path)
else:
remote_copy = args.remote_copy
if args.remote_cmds_cleanup_pattern:
remote_cmds_cleanup = args.remote_cmds_cleanup_pattern.format(
control_path)
else:
remote_cmds_cleanup = ""
return (remote_exec, remote_copy, remote_cmds_cleanup, )
@staticmethod
def _address_family(args):
if args.force_ipv == 4:
family = socket.AF_INET
elif args.force_ipv == 6:
family = socket.AF_INET6
else:
family = 0
return family
@staticmethod
def resolve_target_addresses(host, family):
try:
return ipaddr.resolve_target_addresses(host, family)
except: # noqa
e = sys.exc_info()[1]
raise cdist.Error(("Error resolving target addresses for host '{}'"
": {}").format(host, e))
@classmethod
def onehost(cls, host, host_tags, host_base_path, host_dir_name, args,
parallel, configuration, remove_remote_files_dirs=False):
"""Configure ONE system.
If operating in parallel then return tuple (host, True|False, )
so that main process knows for which host function was successful.
"""
log = logging.getLogger(host)
try:
if args.log_server:
# Start a log server so that nested `cdist config` runs
# have a place to send their logs to.
log_server_socket_dir = tempfile.mkdtemp()
cls._register_path_for_removal(log_server_socket_dir)
cdist.log.setupLogServer(log_server_socket_dir, log)
remote_exec, remote_copy, cleanup_cmd = cls._resolve_remote_cmds(
args)
log.debug("remote_exec for host \"%s\": %s", host, remote_exec)
log.debug("remote_copy for host \"%s\": %s", host, remote_copy)
family = cls._address_family(args)
log.debug("address family: %s", family)
target_host = cls.resolve_target_addresses(host, family)
log.debug("target_host for host \"%s\": %s", host, target_host)
local = cdist.exec.local.Local(
target_host=target_host,
target_host_tags=host_tags,
base_root_path=host_base_path,
host_dir_name=host_dir_name,
initial_manifest=args.manifest,
add_conf_dirs=args.conf_dir,
cache_path_pattern=args.cache_path_pattern,
quiet_mode=args.quiet,
configuration=configuration,
exec_path=sys.argv[0],
save_output_streams=args.save_output_streams)
# Make __global state dir available to custom remote scripts.
os.environ['__global'] = local.base_path
remote = cdist.exec.remote.Remote(
target_host=target_host,
remote_exec=remote_exec,
remote_copy=remote_copy,
base_path=args.remote_out_path,
quiet_mode=args.quiet,
archiving_mode=args.use_archiving,
configuration=configuration,
stdout_base_path=local.stdout_base_path,
stderr_base_path=local.stderr_base_path,
save_output_streams=args.save_output_streams)
cleanup_cmds = []
if cleanup_cmd:
cleanup_cmds.append(cleanup_cmd)
c = cls(local, remote, dry_run=args.dry_run, jobs=args.jobs,
cleanup_cmds=cleanup_cmds,
remove_remote_files_dirs=remove_remote_files_dirs)
c.run()
cls._remove_paths()
except cdist.Error as e:
log.error(e)
if parallel:
return (host, False, )
else:
raise
if parallel:
return (host, True, )
@staticmethod
def create_base_root_path(out_path=None):
if out_path:
base_root_path = out_path
else:
base_root_path = tempfile.mkdtemp()
return base_root_path
@staticmethod
def create_host_base_dirs(host, base_root_path):
hostdir = cdist.str_hash(host)
host_base_path = os.path.join(base_root_path, hostdir)
return (host_base_path, hostdir)
def run(self):
"""Do what is most often done: deploy & cleanup"""
start_time = time.time()
self.log.info("Starting %s run",
'dry' if self.dry_run else 'configuration')
self._init_files_dirs()
self.explorer.run_global_explorers(self.local.global_explorer_out_path)
try:
self.manifest.run_initial_manifest(self.local.initial_manifest)
except cdist.Error as e:
which = "init"
stdout_path = os.path.join(self.local.stdout_base_path, which)
stderr_path = os.path.join(self.local.stderr_base_path, which)
raise cdist.InitialManifestError(self.local.initial_manifest,
stdout_path, stderr_path, e)
self.iterate_until_finished()
self.cleanup()
self._remove_files_dirs()
self.local.save_cache(start_time)
self.log.info("Finished %s run in %.2f seconds",
'dry' if self.dry_run else 'successful',
time.time() - start_time)
def cleanup(self):
self.log.debug("Running cleanup commands")
for cleanup_cmd in self.cleanup_cmds:
cmd = cleanup_cmd.split()
cmd.append(self.local.target_host[0])
try:
if self.log.getEffectiveLevel() <= logging.DEBUG:
quiet_mode = False
else:
quiet_mode = True
self.local.run(cmd, return_output=False, save_output=False,
quiet_mode=quiet_mode)
except cdist.Error as e:
# Log warning but continue.
self.log.warning("Cleanup command failed: %s", e)
def object_list(self):
"""Short name for object list retrieval"""
for cdist_object in core.CdistObject.list_objects(
self.local.object_path, self.local.type_path,
self.local.object_marker_name):
if cdist_object.cdist_type.is_install:
self.log.debug("Running in config mode, ignoring install "
"object: %s", cdist_object)
else:
yield cdist_object
def iterate_once(self):
"""
Iterate over the objects once - helper method for
iterate_until_finished
"""
if self.jobs:
objects_changed = self._iterate_once_parallel()
else:
objects_changed = self._iterate_once_sequential()
return objects_changed
def _iterate_once_sequential(self):
self.log.debug("Iteration in sequential mode")
objects_changed = False
for cdist_object in self.object_list():
if cdist_object.has_requirements_unfinished(
cdist_object.requirements):
"""We cannot do anything for this poor object"""
continue
if cdist_object.state == core.CdistObject.STATE_UNDEF:
"""Prepare the virgin object"""
self.object_prepare(cdist_object)
objects_changed = True
if cdist_object.has_requirements_unfinished(
cdist_object.autorequire):
"""The previous step created objects we depend on -
wait for them
"""
continue
if cdist_object.state == core.CdistObject.STATE_PREPARED:
self.object_run(cdist_object)
objects_changed = True
return objects_changed
def _iterate_once_parallel(self):
self.log.debug("Iteration in parallel mode in %d jobs", self.jobs)
objects_changed = False
cargo = []
for cdist_object in self.object_list():
if cdist_object.has_requirements_unfinished(
cdist_object.requirements):
"""We cannot do anything for this poor object"""
continue
if cdist_object.state == core.CdistObject.STATE_UNDEF:
"""Prepare the virgin object"""
# self.object_prepare(cdist_object)
# objects_changed = True
cargo.append(cdist_object)
n = len(cargo)
if n == 1:
self.log.debug("Only one object, preparing sequentially")
self.object_prepare(cargo[0])
objects_changed = True
elif cargo:
self.log.trace("Multiprocessing start method is %s",
multiprocessing.get_start_method())
self.log.trace("Multiprocessing cargo: %s", cargo)
cargo_types = set()
for c in cargo:
cargo_types.add(c.cdist_type)
self.log.trace("Multiprocessing cargo_types: %s", cargo_types)
nt = len(cargo_types)
if nt == 1:
self.log.debug(("Only one type, transferring explorers "
"sequentially"))
self.explorer.transfer_type_explorers(cargo_types.pop())
else:
self.log.trace("Starting multiprocessing Pool for %d "
"parallel types explorers transferring", nt)
args = [
(ct, ) for ct in cargo_types
]
mp_pool_run(self.explorer.transfer_type_explorers, args,
jobs=self.jobs)
self.log.trace(("Multiprocessing for parallel transferring "
"types' explorers finished"))
self.log.trace("Starting multiprocessing Pool for %d parallel "
"objects preparation", n)
args = [
(c, False, ) for c in cargo
]
mp_pool_run(self.object_prepare, args, jobs=self.jobs)
self.log.trace(("Multiprocessing for parallel object "
"preparation finished"))
objects_changed = True
del cargo[:]
for cdist_object in self.object_list():
if cdist_object.has_requirements_unfinished(
cdist_object.requirements):
"""We cannot do anything for this poor object"""
continue
if cdist_object.state == core.CdistObject.STATE_PREPARED:
if cdist_object.has_requirements_unfinished(
cdist_object.autorequire):
"""The previous step created objects we depend on -
wait for them
"""
continue
# self.object_run(cdist_object)
# objects_changed = True
# put objects in chuncks of distinct types
# so that there is no more than one object
# of the same type in one chunk because there is a
# possibility of object's process locking which
# prevents parallel execution at remote
# and do this only for nonparallel marked types
for chunk in cargo:
for obj in chunk:
if (obj.cdist_type == cdist_object.cdist_type and
cdist_object.cdist_type.is_nonparallel):
break
else:
chunk.append(cdist_object)
break
else:
chunk = [cdist_object, ]
cargo.append(chunk)
for chunk in cargo:
self.log.trace("Running chunk: %s", chunk)
n = len(chunk)
if n == 1:
self.log.debug("Only one object, running sequentially")
self.object_run(chunk[0])
objects_changed = True
elif chunk:
self.log.trace("Multiprocessing start method is %s",
multiprocessing.get_start_method())
self.log.trace("Starting multiprocessing Pool for %d "
"parallel object run", n)
args = [
(c, ) for c in chunk
]
mp_pool_run(self.object_run, args, jobs=self.jobs)
self.log.trace(("Multiprocessing for parallel object "
"run finished"))
objects_changed = True
return objects_changed
def _open_logger(self):
self.log = logging.getLogger(self.local.target_host[0])
# logger is not pickable, so remove it when we pickle
def __getstate__(self):
state = self.__dict__.copy()
if 'log' in state:
del state['log']
return state
# recreate logger when we unpickle
def __setstate__(self, state):
self.__dict__.update(state)
self._open_logger()
def _validate_dependencies(self):
'''
Build dependency graph for unfinished objects and
check for cycles.
'''
graph = {}
def _add_requirements(cdist_object, requirements):
obj_name = cdist_object.name
if obj_name not in graph:
graph[obj_name] = []
for requirement in cdist_object.requirements_unfinished(
requirements):
graph[obj_name].append(requirement.name)
for cdist_object in self.object_list():
if cdist_object.state == cdist_object.STATE_DONE:
continue
_add_requirements(cdist_object, cdist_object.requirements)
_add_requirements(cdist_object, cdist_object.autorequire)
return graph_check_cycle(graph)
def iterate_until_finished(self):
"""
Go through all objects and solve them
one after another
"""
objects_changed = True
while objects_changed:
# Check for cycles as early as possible.
has_cycle, path = self._validate_dependencies()
if has_cycle:
raise cdist.UnresolvableRequirementsError(
"Cycle detected in object dependencies:\n{}!".format(
" -> ".join(path)))
objects_changed = self.iterate_once()
# Check whether all objects have been finished
unfinished_objects = []
for cdist_object in self.object_list():
if not cdist_object.state == cdist_object.STATE_DONE:
unfinished_objects.append(cdist_object)
if unfinished_objects:
info_string = []
for cdist_object in unfinished_objects:
requirement_names = []
autorequire_names = []
for requirement in cdist_object.requirements_unfinished(
cdist_object.requirements):
requirement_names.append(requirement.name)
for requirement in cdist_object.requirements_unfinished(
cdist_object.autorequire):
autorequire_names.append(requirement.name)
requirements = "\n ".join(requirement_names)
autorequire = "\n ".join(autorequire_names)
info_string.append(("%s requires:\n"
" %s\n"
"%s ""autorequires:\n"
" %s" % (
cdist_object.name,
requirements, cdist_object.name,
autorequire)))
raise cdist.UnresolvableRequirementsError(
("The requirements of the following objects could not be "
"resolved:\n{}").format("\n".join(info_string)))
def _handle_deprecation(self, cdist_object):
cdist_type = cdist_object.cdist_type
deprecated = cdist_type.deprecated
if deprecated is not None:
if deprecated:
self.log.warning("Type %s is deprecated: %s", cdist_type.name,
deprecated)
else:
self.log.warning("Type %s is deprecated.", cdist_type.name)
for param in cdist_object.parameters:
if param in cdist_type.deprecated_parameters:
msg = cdist_type.deprecated_parameters[param]
if msg:
format = "%s parameter of type %s is deprecated: %s"
args = [param, cdist_type.name, msg]
else:
format = "%s parameter of type %s is deprecated."
args = [param, cdist_type.name]
self.log.warning(format, *args)
def object_prepare(self, cdist_object, transfer_type_explorers=True):
"""Prepare object: Run type explorer + manifest"""
self._handle_deprecation(cdist_object)
self.log.verbose("Preparing object %s", cdist_object.name)
self.log.verbose("Running manifest and explorers for %s",
cdist_object.name)
self.explorer.run_type_explorers(cdist_object, transfer_type_explorers)
try:
self.manifest.run_type_manifest(cdist_object)
self.log.trace("[ORDER_DEP] Removing order dep files for %s",
cdist_object)
cdist_object.cleanup()
cdist_object.state = core.CdistObject.STATE_PREPARED
except cdist.Error as e:
raise cdist.CdistObjectError(cdist_object, e)
def object_run(self, cdist_object):
"""Run gencode and code for an object"""
try:
self.log.verbose("Running object %s", cdist_object.name)
if cdist_object.state == core.CdistObject.STATE_DONE:
raise cdist.Error(("Attempting to run an already finished "
"object: {}").format(cdist_object))
# Generate
self.log.debug("Generating code for %s", cdist_object.name)
cdist_object.code_local = self.code.run_gencode_local(cdist_object)
cdist_object.code_remote = self.code.run_gencode_remote(
cdist_object)
if cdist_object.code_local or cdist_object.code_remote:
cdist_object.changed = True
# Execute
if cdist_object.code_local or cdist_object.code_remote:
self.log.info("Processing %s", cdist_object.name)
if not self.dry_run:
if cdist_object.code_local:
self.log.trace("Executing local code for %s",
cdist_object.name)
self.code.run_code_local(cdist_object)
if cdist_object.code_remote:
self.log.trace("Executing remote code for %s",
cdist_object.name)
self.code.transfer_code_remote(cdist_object)
self.code.run_code_remote(cdist_object)
# Mark this object as done
self.log.trace("Finishing run of %s", cdist_object.name)
cdist_object.state = core.CdistObject.STATE_DONE
except cdist.Error as e:
raise cdist.CdistObjectError(cdist_object, e)