path: root/bin/cloud-init

#!/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 <scott.moser@canonical.com>
#    Author: Juerg Haefliger <juerg.haefliger@hp.com>
#    Author: Joshua Harlow <harlowja@yahoo-inc.com>
#
#    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 <http://www.gnu.org/licenses/>.

import argparse
import os
import sys
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 log as logging
from cloudinit import netinfo
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 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):
    tpl_params = {
        'version': version.version_string(),
        'uptime': util.uptime(),
        'timestamp': util.time_rfc2822(),
        'action': action,
    }
    tpl_msg = templater.render_string(WELCOME_MSG_TPL, tpl_params)
    util.multi_log("%s\n" % (tpl_msg),
                   console=False, stderr=True)


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 0
    else:
        LOG.debug("Ran %s modules with %s failures",
                  len(which_ran), len(failures))
        return len(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 configuration
    # 10. Run the modules for the 'init' stage
    # 11. Done!
    welcome(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)
    # 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 0
    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:
        util.logexc(LOG, ("No instance datasource found!"
                          " Likely bad things to come!"))
        # In the case of cloud-init (net mode) 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 not args.force:
            if args.local:
                return 0
            else:
                return 1
    # 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_userdata',
                                             init.consume_userdata,
                                             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_userdata(PER_ALWAYS)
    except Exception:
        util.logexc(LOG, "Consuming user data failed!")
        return 1
    # Stage 8 - TODO - do we really need to re-extract our configs?
    mods = stages.Modules(init, extract_fns(args))
    # Stage 9 - TODO is this really needed??
    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!")
    # Stage 10
    return 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!
    welcome("%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
        util.logexc(LOG, ('Can not apply stage %s, '
                          'no datasource found!'
                          " Likely bad things to come!"), name)
        print_exc(('Can not apply stage %s, '
                   'no datasource found!'
                   " Likely bad things to come!") % (name))
        if not args.force:
            return 1
    # 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)
    # 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
    welcome("%s:%s" % (name, mod_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)
    # 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 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()

    (name, functor) = args.action
    return functor(name, args)


if __name__ == '__main__':
    sys.exit(main())