Synnefo

# Copyright 2011 GRNET S.A. All rights reserved.

#

# Redistribution and use in source and binary forms, with or without

# modification, are permitted provided that the following conditions

# are met:

#

#   1. Redistributions of source code must retain the above copyright

#      notice, this list of conditions and the following disclaimer.

#

#  2. Redistributions in binary form must reproduce the above copyright

#     notice, this list of conditions and the following disclaimer in the

#     documentation and/or other materials provided with the distribution.

#

# THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND

# ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE

# IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE

# ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE

# FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL

# DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS

# OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)

# HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT

# LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY

# OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF

# SUCH DAMAGE.

#

# The views and conclusions contained in the software and documentation are

# those of the authors and should not be interpreted as representing official

# policies, either expressed or implied, of GRNET S.A.

# Callback functions used by the dispatcher to process incoming notifications

# from AMQP queues.

import logging

import json

from functools import wraps

from django.db import transaction

from synnefo.db.models import (Backend, VirtualMachine, Network,

                               BackendNetwork, pooled_rapi_client)

from synnefo.logic import utils, backend as backend_mod, rapi

from synnefo.lib.utils import merge_time

log = logging.getLogger(__name__)

def handle_message_delivery(func):

    """ Generic decorator for handling messages.

    This decorator is responsible for converting the message into json format,

    handling of common exceptions and acknowledment of message if needed.

    """

    @wraps(func)

    def wrapper(client, message, *args, **kwargs):

        try:

            msg = None

            msg = json.loads(message['body'])

            func(msg)

            client.basic_ack(message)

        except ValueError as e:

            log.error("Incoming message not in JSON format %s: %s", e, message)

            client.basic_nack(message)

        except KeyError as e:

            log.error("Malformed incoming JSON, missing attribute %s: %s",

                      e, message)

            client.basic_nack(message)

        except Exception as e:

            if msg:

                log.exception("Unexpected error: %s, msg: %s", e, msg)

            else:

                log.exception("Unexpected error: %s", e)

            client.basic_reject(message)

    return wrapper

def instance_from_msg(func):

    """ Decorator for getting the VirtualMachine object of the msg.

    """

    @handle_message_delivery

    @wraps(func)

    def wrapper(msg):

        try:

            vm_id = utils.id_from_instance_name(msg["instance"])

            vm = VirtualMachine.objects.select_for_update().get(id=vm_id)

            func(vm, msg)

        except VirtualMachine.InvalidBackendIdError:

            log.debug("Ignoring msg for unknown instance %s.", msg['instance'])

        except VirtualMachine.DoesNotExist:

            log.error("VM for instance %s with id %d not found in DB.",

                      msg['instance'], vm_id)

        except (Network.InvalidBackendIdError, Network.DoesNotExist) as e:

            log.error("Invalid message, can not find network. msg: %s", msg)

    return wrapper

def network_from_msg(func):

    """ Decorator for getting the BackendNetwork object of the msg.

    """

    @handle_message_delivery

    @wraps(func)

    def wrapper(msg):

        try:

            network_id = utils.id_from_network_name(msg["network"])

            network = Network.objects.select_for_update().get(id=network_id)

            backend = Backend.objects.get(clustername=msg['cluster'])

            bnet, new = BackendNetwork.objects.get_or_create(network=network,

                                                             backend=backend)

            if new:

                log.info("Created missing BackendNetwork %s", bnet)

            func(bnet, msg)

        except Network.InvalidBackendIdError:

            log.debug("Ignoring msg for unknown network %s.", msg['network'])

        except Network.DoesNotExist:

            log.error("Network %s not found in DB.", msg['network'])

        except Backend.DoesNotExist:

            log.error("Backend %s not found in DB.", msg['cluster'])

        except BackendNetwork.DoesNotExist:

            log.error("Network %s on backend %s not found in DB.",

                      msg['network'], msg['cluster'])

    return wrapper

def if_update_required(func):

    """

    Decorator for checking if an incoming message needs to update the db.

    The database will not be updated in the following cases:

    - The message has been redelivered and the action has already been

      completed. In this case the event_time will be equal with the one

      in the database.

    - The message describes a previous state in the ganeti, from the one that

      is described in the db. In this case the event_time will be smaller from

      the one in the database.

    """

    @wraps(func)

    def wrapper(target, msg):

        try:

            event_time = merge_time(msg['event_time'])

        except:

            log.error("Received message with malformed time: %s",

                      msg['event_time'])

            raise KeyError

        db_time = target.backendtime

        if db_time and event_time <= db_time:

            format_ = "%d/%m/%y %H:%M:%S:%f"

            log.debug("Ignoring message %s.\nevent_timestamp: %s"

                      " db_timestamp: %s",

                      msg,

                      event_time.strftime(format_),

                      db_time.strftime(format_))

            return

        # New message. Update the database!

        func(target, msg, event_time)

    return wrapper

@instance_from_msg

@if_update_required

def update_db(vm, msg, event_time):

    """Process a notification of type 'ganeti-op-status'"""

    log.debug("Processing ganeti-op-status msg: %s", msg)

    if msg['type'] != "ganeti-op-status":

        log.error("Message is of unknown type %s.", msg['type'])

        return

    operation = msg["operation"]

    status = msg["status"]

    jobID = msg["jobId"]

    logmsg = msg["logmsg"]

    nics = msg.get("instance_nics", None)

    job_fields = msg.get("job_fields", {})

    result = msg.get("result", [])

    # Special case: OP_INSTANCE_CREATE with opportunistic locking may fail

    # if all Ganeti nodes are already locked. Retry the job without

    # opportunistic locking..

    if (operation == "OP_INSTANCE_CREATE" and status == "error" and

       job_fields.get("opportunistic_locking", False)):

        try:

            error_code = result[1][1]

        except IndexError:

            error_code = None

        if error_code == rapi.ECODE_TEMP_NORES:

            if vm.backendjobid != jobID:  # The job has already been retried!

                return

            # Remove extra fields

            [job_fields.pop(f) for f in ("OP_ID", "reason")]

            # Remove 'pnode' and 'snode' if they were set by Ganeti iallocator.

            # Ganeti will fail if both allocator and nodes are specified.

            allocator = job_fields.pop("iallocator")

            if allocator is not None:

                [job_fields.pop(f) for f in ("pnode", "snode")]

            name = job_fields.pop("name", job_fields.pop("instance_name"))

            # Turn off opportunistic locking before retrying the job

            job_fields["opportunistic_locking"] = False

            with pooled_rapi_client(vm) as c:

                jobID = c.CreateInstance(name=name, **job_fields)

            # Update the VM fields

            vm.backendjobid = jobID

            # Update the task_job_id for commissions

            vm.task_job_id = jobID

            vm.backendjobstatus = None

            vm.save()

            log.info("Retrying failed creation of instance '%s' without"

                     " opportunistic locking. New job ID: '%s'", name, jobID)

            return

    backend_mod.process_op_status(vm, event_time, jobID,

                                  operation, status,

                                  logmsg, nics=nics,

                                  job_fields=job_fields)

    log.debug("Done processing ganeti-op-status msg for vm %s.",

              msg['instance'])

@network_from_msg

@if_update_required

def update_network(network, msg, event_time):

    """Process a notification of type 'ganeti-network-status'"""

    log.debug("Processing ganeti-network-status msg: %s", msg)

    if msg['type'] != "ganeti-network-status":

        log.error("Message is of unknown type %s.", msg['type'])

        return

    opcode = msg['operation']

    status = msg['status']

    jobid = msg['jobId']

    job_fields = msg.get('job_fields', {})

    if opcode == "OP_NETWORK_SET_PARAMS":

        backend_mod.process_network_modify(network, event_time, jobid, opcode,

                                           status, job_fields)

    else:

        backend_mod.process_network_status(network, event_time, jobid, opcode,

                                           status, msg['logmsg'])

    log.debug("Done processing ganeti-network-status msg for network %s.",

              msg['network'])

@instance_from_msg

@if_update_required

def update_build_progress(vm, msg, event_time):

    """

    Process a create progress message. Update build progress, or create

    appropriate diagnostic entries for the virtual machine instance.

    """

    log.debug("Processing ganeti-create-progress msg: %s", msg)

    if msg['type'] not in ('image-copy-progress', 'image-error', 'image-info',

                           'image-warning', 'image-helper'):

        log.error("Message is of unknown type %s", msg['type'])

        return

    if msg['type'] == 'image-copy-progress':

        backend_mod.process_create_progress(vm, event_time, msg['progress'])

        # we do not add diagnostic messages for copy-progress messages

        return

    # default diagnostic fields

    source = msg['type']

    level = 'DEBUG'

    message = msg.get('messages', '')

    if isinstance(message, list):

        message = " ".join(message)

    details = msg.get('stderr', None)

    if msg['type'] == 'image-helper':

        # for helper task events join subtype to diagnostic source and

        # set task name as diagnostic message

        if msg.get('subtype', None):

            if msg.get('subtype') in ['task-start', 'task-end']:

                message = msg.get('task', message)

                source = "%s-%s" % (source, msg.get('subtype'))

        if msg.get('subtype', None) == 'warning':

            level = 'WARNING'

        if msg.get('subtype', None) == 'error':

            level = 'ERROR'

        if msg.get('subtype', None) == 'info':

            level = 'INFO'

    if msg['type'] == 'image-error':

        level = 'ERROR'

    if msg['type'] == 'image-warning':

        level = 'WARNING'

    if not message.strip():

        message = " ".join(source.split("-")).capitalize()

    # create the diagnostic entry

    backend_mod.create_instance_diagnostic(vm, message, source, level,

                                           event_time, details=details)

    log.debug("Done processing ganeti-create-progress msg for vm %s.",

              msg['instance'])

@handle_message_delivery

@transaction.commit_on_success()

def update_cluster(msg):

    clustername = msg.get("cluster")

    if clustername is None:

        return

    backend = Backend.objects.select_for_update().get(clustername=clustername)

    backend_mod.update_backend_disk_templates(backend)

    backend_mod.update_backend_resources(backend)

def dummy_proc(client, message, *args, **kwargs):

    try:

        log.debug("Msg: %s", message['body'])

        client.basic_ack(message)

    except Exception as e:

        log.exception("Could not receive message %s" % e)