#!/usr/bin/python # vi: ts=4 expandtab # # Copyright (C) 2012 Canonical Ltd. # Copyright (C) 2012 Hewlett-Packard Development Company, L.P. # Copyright (C) 2012 Yahoo! Inc. # # Author: Scott Moser # Author: Juerg Haefliger # Author: Joshua Harlow # # This program is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License version 3, as # published by the Free Software Foundation. # # This program 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 this program. If not, see . import argparse import json import os import sys import time import tempfile import traceback # This is more just for running from the bin folder so that # cloud-init binary can find the cloudinit module possible_topdir = os.path.normpath(os.path.join(os.path.abspath( sys.argv[0]), os.pardir, os.pardir)) if os.path.exists(os.path.join(possible_topdir, "cloudinit", "__init__.py")): sys.path.insert(0, possible_topdir) from cloudinit import patcher patcher.patch() from cloudinit import log as logging from cloudinit import netinfo from cloudinit import signal_handler from cloudinit import sources from cloudinit import stages from cloudinit import templater from cloudinit import util from cloudinit import version from cloudinit.settings import (PER_INSTANCE, PER_ALWAYS, PER_ONCE, CLOUD_CONFIG) # Pretty little cheetah formatted welcome message template WELCOME_MSG_TPL = ("Cloud-init v. ${version} running '${action}' at " "${timestamp}. Up ${uptime} seconds.") # Module section template MOD_SECTION_TPL = "cloud_%s_modules" # Things u can query on QUERY_DATA_TYPES = [ 'data', 'data_raw', 'instance_id', ] # Frequency shortname to full name # (so users don't have to remember the full name...) FREQ_SHORT_NAMES = { 'instance': PER_INSTANCE, 'always': PER_ALWAYS, 'once': PER_ONCE, } LOG = logging.getLogger() # Used for when a logger may not be active # and we still want to print exceptions... def print_exc(msg=''): if msg: sys.stderr.write("%s\n" % (msg)) sys.stderr.write('-' * 60) sys.stderr.write("\n") traceback.print_exc(file=sys.stderr) sys.stderr.write('-' * 60) sys.stderr.write("\n") def welcome(action, msg=None): if not msg: msg = welcome_format(action) util.multi_log("%s\n" % (msg), console=False, stderr=True, log=LOG) return msg def welcome_format(action): tpl_params = { 'version': version.version_string(), 'uptime': util.uptime(), 'timestamp': util.time_rfc2822(), 'action': action, } return templater.render_string(WELCOME_MSG_TPL, tpl_params) def extract_fns(args): # Files are already opened so lets just pass that along # since it would of broke if it couldn't have # read that file already... fn_cfgs = [] if args.files: for fh in args.files: # The realpath is more useful in logging # so lets resolve to that... fn_cfgs.append(os.path.realpath(fh.name)) return fn_cfgs def run_module_section(mods, action_name, section): full_section_name = MOD_SECTION_TPL % (section) (which_ran, failures) = mods.run_section(full_section_name) total_attempted = len(which_ran) + len(failures) if total_attempted == 0: msg = ("No '%s' modules to run" " under section '%s'") % (action_name, full_section_name) sys.stderr.write("%s\n" % (msg)) LOG.debug(msg) return [] else: LOG.debug("Ran %s modules with %s failures", len(which_ran), len(failures)) return failures def main_init(name, args): deps = [sources.DEP_FILESYSTEM, sources.DEP_NETWORK] if args.local: deps = [sources.DEP_FILESYSTEM] if not args.local: # See doc/kernel-cmdline.txt # # This is used in maas datasource, in "ephemeral" (read-only root) # environment where the instance netboots to iscsi ro root. # and the entity that controls the pxe config has to configure # the maas datasource. # # Could be used elsewhere, only works on network based (not local). root_name = "%s.d" % (CLOUD_CONFIG) target_fn = os.path.join(root_name, "91_kernel_cmdline_url.cfg") util.read_write_cmdline_url(target_fn) # Cloud-init 'init' stage is broken up into the following sub-stages # 1. Ensure that the init object fetches its config without errors # 2. Setup logging/output redirections with resultant config (if any) # 3. Initialize the cloud-init filesystem # 4. Check if we can stop early by looking for various files # 5. Fetch the datasource # 6. Connect to the current instance location + update the cache # 7. Consume the userdata (handlers get activated here) # 8. Construct the modules object # 9. Adjust any subsequent logging/output redirections using the modules # objects config as it may be different from init object # 10. Run the modules for the 'init' stage # 11. Done! if not args.local: w_msg = welcome_format(name) else: w_msg = welcome_format("%s-local" % (name)) init = stages.Init(deps) # Stage 1 init.read_cfg(extract_fns(args)) # Stage 2 outfmt = None errfmt = None try: LOG.debug("Closing stdin") util.close_stdin() (outfmt, errfmt) = util.fixup_output(init.cfg, name) except: util.logexc(LOG, "Failed to setup output redirection!") print_exc("Failed to setup output redirection!") if args.debug: # Reset so that all the debug handlers are closed out LOG.debug(("Logging being reset, this logger may no" " longer be active shortly")) logging.resetLogging() logging.setupLogging(init.cfg) # Any log usage prior to setupLogging above did not have local user log # config applied. We send the welcome message now, as stderr/out have # been redirected and log now configured. welcome(name, msg=w_msg) # Stage 3 try: init.initialize() except Exception: util.logexc(LOG, "Failed to initialize, likely bad things to come!") # Stage 4 path_helper = init.paths if not args.local: sys.stderr.write("%s\n" % (netinfo.debug_info())) LOG.debug(("Checking to see if files that we need already" " exist from a previous run that would allow us" " to stop early.")) stop_files = [ os.path.join(path_helper.get_cpath("data"), "no-net"), path_helper.get_ipath_cur("obj_pkl"), ] existing_files = [] for fn in stop_files: try: c = util.load_file(fn) if len(c): existing_files.append((fn, len(c))) except Exception: pass if existing_files: LOG.debug("Exiting early due to the existence of %s files", existing_files) return (None, []) else: LOG.debug("Execution continuing, no previous run detected that" " would allow us to stop early.") else: # The cache is not instance specific, so it has to be purged # but we want 'start' to benefit from a cache if # a previous start-local populated one... manual_clean = util.get_cfg_option_bool(init.cfg, 'manual_cache_clean', False) if manual_clean: LOG.debug("Not purging instance link, manual cleaning enabled") init.purge_cache(False) else: init.purge_cache() # Delete the non-net file as well util.del_file(os.path.join(path_helper.get_cpath("data"), "no-net")) # Stage 5 try: init.fetch() except sources.DataSourceNotFoundException: # In the case of 'cloud-init init' without '--local' it is a bit # more likely that the user would consider it failure if nothing was # found. When using upstart it will also mentions job failure # in console log if exit code is != 0. if args.local: LOG.debug("No local datasource found") else: util.logexc(LOG, ("No instance datasource found!" " Likely bad things to come!")) if not args.force: if args.local: return (None, []) else: return (None, ["No instance datasource found."]) # Stage 6 iid = init.instancify() LOG.debug("%s will now be targeting instance id: %s", name, iid) init.update() # Stage 7 try: # Attempt to consume the data per instance. # This may run user-data handlers and/or perform # url downloads and such as needed. (ran, _results) = init.cloudify().run('consume_data', init.consume_data, args=[PER_INSTANCE], freq=PER_INSTANCE) if not ran: # Just consume anything that is set to run per-always # if nothing ran in the per-instance code # # See: https://bugs.launchpad.net/bugs/819507 for a little # reason behind this... init.consume_data(PER_ALWAYS) except Exception: util.logexc(LOG, "Consuming user data failed!") return (init.datasource, ["Consuming user data failed!"]) # Stage 8 - re-read and apply relevant cloud-config to include user-data mods = stages.Modules(init, extract_fns(args)) # Stage 9 try: outfmt_orig = outfmt errfmt_orig = errfmt (outfmt, errfmt) = util.get_output_cfg(mods.cfg, name) if outfmt_orig != outfmt or errfmt_orig != errfmt: LOG.warn("Stdout, stderr changing to (%s, %s)", outfmt, errfmt) (outfmt, errfmt) = util.fixup_output(mods.cfg, name) except: util.logexc(LOG, "Failed to re-adjust output redirection!") logging.setupLogging(mods.cfg) # Stage 10 return (init.datasource, run_module_section(mods, name, name)) def main_modules(action_name, args): name = args.mode # Cloud-init 'modules' stages are broken up into the following sub-stages # 1. Ensure that the init object fetches its config without errors # 2. Get the datasource from the init object, if it does # not exist then that means the main_init stage never # worked, and thus this stage can not run. # 3. Construct the modules object # 4. Adjust any subsequent logging/output redirections using # the modules objects configuration # 5. Run the modules for the given stage name # 6. Done! w_msg = welcome_format("%s:%s" % (action_name, name)) init = stages.Init(ds_deps=[]) # Stage 1 init.read_cfg(extract_fns(args)) # Stage 2 try: init.fetch() except sources.DataSourceNotFoundException: # There was no datasource found, theres nothing to do msg = ('Can not apply stage %s, no datasource found! Likely bad ' 'things to come!' % name) util.logexc(LOG, msg) print_exc(msg) if not args.force: return [(msg)] # Stage 3 mods = stages.Modules(init, extract_fns(args)) # Stage 4 try: LOG.debug("Closing stdin") util.close_stdin() util.fixup_output(mods.cfg, name) except: util.logexc(LOG, "Failed to setup output redirection!") if args.debug: # Reset so that all the debug handlers are closed out LOG.debug(("Logging being reset, this logger may no" " longer be active shortly")) logging.resetLogging() logging.setupLogging(mods.cfg) # now that logging is setup and stdout redirected, send welcome welcome(name, msg=w_msg) # Stage 5 return run_module_section(mods, name, name) def main_query(name, _args): raise NotImplementedError(("Action '%s' is not" " currently implemented") % (name)) def main_single(name, args): # Cloud-init single stage is broken up into the following sub-stages # 1. Ensure that the init object fetches its config without errors # 2. Attempt to fetch the datasource (warn if it doesn't work) # 3. Construct the modules object # 4. Adjust any subsequent logging/output redirections using # the modules objects configuration # 5. Run the single module # 6. Done! mod_name = args.name w_msg = welcome_format(name) init = stages.Init(ds_deps=[]) # Stage 1 init.read_cfg(extract_fns(args)) # Stage 2 try: init.fetch() except sources.DataSourceNotFoundException: # There was no datasource found, # that might be bad (or ok) depending on # the module being ran (so continue on) util.logexc(LOG, ("Failed to fetch your datasource," " likely bad things to come!")) print_exc(("Failed to fetch your datasource," " likely bad things to come!")) if not args.force: return 1 # Stage 3 mods = stages.Modules(init, extract_fns(args)) mod_args = args.module_args if mod_args: LOG.debug("Using passed in arguments %s", mod_args) mod_freq = args.frequency if mod_freq: LOG.debug("Using passed in frequency %s", mod_freq) mod_freq = FREQ_SHORT_NAMES.get(mod_freq) # Stage 4 try: LOG.debug("Closing stdin") util.close_stdin() util.fixup_output(mods.cfg, None) except: util.logexc(LOG, "Failed to setup output redirection!") if args.debug: # Reset so that all the debug handlers are closed out LOG.debug(("Logging being reset, this logger may no" " longer be active shortly")) logging.resetLogging() logging.setupLogging(mods.cfg) # now that logging is setup and stdout redirected, send welcome welcome(name, msg=w_msg) # Stage 5 (which_ran, failures) = mods.run_single(mod_name, mod_args, mod_freq) if failures: LOG.warn("Ran %s but it failed!", mod_name) return 1 elif not which_ran: LOG.warn("Did not run %s, does it exist?", mod_name) return 1 else: # Guess it worked return 0 def atomic_write_json(path, data): tf = None try: tf = tempfile.NamedTemporaryFile(dir=os.path.dirname(path), delete=False) tf.write(util.encode_text(json.dumps(data, indent=1) + "\n")) tf.close() os.rename(tf.name, path) except Exception as e: if tf is not None: util.del_file(tf.name) raise e def status_wrapper(name, args, data_d=None, link_d=None): if data_d is None: data_d = os.path.normpath("/var/lib/cloud/data") if link_d is None: link_d = os.path.normpath("/run/cloud-init") status_path = os.path.join(data_d, "status.json") status_link = os.path.join(link_d, "status.json") result_path = os.path.join(data_d, "result.json") result_link = os.path.join(link_d, "result.json") util.ensure_dirs((data_d, link_d,)) (_name, functor) = args.action if name == "init": if args.local: mode = "init-local" else: mode = "init" elif name == "modules": mode = "modules-%s" % args.mode else: raise ValueError("unknown name: %s" % name) modes = ('init', 'init-local', 'modules-config', 'modules-final') status = None if mode == 'init-local': for f in (status_link, result_link, status_path, result_path): util.del_file(f) else: try: status = json.loads(util.load_file(status_path)) except: pass if status is None: nullstatus = { 'errors': [], 'start': None, 'finished': None, } status = {'v1': {}} for m in modes: status['v1'][m] = nullstatus.copy() status['v1']['datasource'] = None v1 = status['v1'] v1['stage'] = mode v1[mode]['start'] = time.time() atomic_write_json(status_path, status) util.sym_link(os.path.relpath(status_path, link_d), status_link, force=True) try: ret = functor(name, args) if mode in ('init', 'init-local'): (datasource, errors) = ret if datasource is not None: v1['datasource'] = str(datasource) else: errors = ret v1[mode]['errors'] = [str(e) for e in errors] except Exception as e: util.logexc(LOG, "failed of stage %s", mode) print_exc("failed run of stage %s", mode) v1[mode]['errors'] = [str(e)] v1[mode]['finished'] = time.time() v1['stage'] = None atomic_write_json(status_path, status) if mode == "modules-final": # write the 'finished' file errors = [] for m in modes: if v1[m]['errors']: errors.extend(v1[m].get('errors', [])) atomic_write_json(result_path, {'v1': {'datasource': v1['datasource'], 'errors': errors}}) util.sym_link(os.path.relpath(result_path, link_d), result_link, force=True) return len(v1[mode]['errors']) def main(): parser = argparse.ArgumentParser() # Top level args parser.add_argument('--version', '-v', action='version', version='%(prog)s ' + (version.version_string())) parser.add_argument('--file', '-f', action='append', dest='files', help=('additional yaml configuration' ' files to use'), type=argparse.FileType('rb')) parser.add_argument('--debug', '-d', action='store_true', help=('show additional pre-action' ' logging (default: %(default)s)'), default=False) parser.add_argument('--force', action='store_true', help=('force running even if no datasource is' ' found (use at your own risk)'), dest='force', default=False) subparsers = parser.add_subparsers() # Each action and its sub-options (if any) parser_init = subparsers.add_parser('init', help=('initializes cloud-init and' ' performs initial modules')) parser_init.add_argument("--local", '-l', action='store_true', help="start in local mode (default: %(default)s)", default=False) # This is used so that we can know which action is selected + # the functor to use to run this subcommand parser_init.set_defaults(action=('init', main_init)) # These settings are used for the 'config' and 'final' stages parser_mod = subparsers.add_parser('modules', help=('activates modules ' 'using a given configuration key')) parser_mod.add_argument("--mode", '-m', action='store', help=("module configuration name " "to use (default: %(default)s)"), default='config', choices=('init', 'config', 'final')) parser_mod.set_defaults(action=('modules', main_modules)) # These settings are used when you want to query information # stored in the cloud-init data objects/directories/files parser_query = subparsers.add_parser('query', help=('query information stored ' 'in cloud-init')) parser_query.add_argument("--name", '-n', action="store", help="item name to query on", required=True, choices=QUERY_DATA_TYPES) parser_query.set_defaults(action=('query', main_query)) # This subcommand allows you to run a single module parser_single = subparsers.add_parser('single', help=('run a single module ')) parser_single.set_defaults(action=('single', main_single)) parser_single.add_argument("--name", '-n', action="store", help="module name to run", required=True) parser_single.add_argument("--frequency", action="store", help=("frequency of the module"), required=False, choices=list(FREQ_SHORT_NAMES.keys())) parser_single.add_argument("module_args", nargs="*", metavar='argument', help=('any additional arguments to' ' pass to this module')) parser_single.set_defaults(action=('single', main_single)) args = parser.parse_args() # Setup basic logging to start (until reinitialized) # iff in debug mode... if args.debug: logging.setupBasicLogging() # Setup signal handlers before running signal_handler.attach_handlers() (name, functor) = args.action if name in ("modules", "init"): functor = status_wrapper return util.log_time(logfunc=LOG.debug, msg="cloud-init mode '%s'" % name, get_uptime=True, func=functor, args=(name, args)) if __name__ == '__main__': sys.exit(main())