amal/uncloud
1
0
Fork 0
Code Issues Pull requests Projects Releases Wiki Activity

ucloud-{api,scheduler,host,filescanner,imagescanner,metadata} combined

Browse source
This commit is contained in:
ahmadbilalkhalid 2019-10-25 11:42:40 +05:00
commit da77ac65eb
29 changed files with 3941 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

32
host/config.py Executable file
View file

@ -0,0 +1,32 @@
import logging
from etcd3_wrapper import Etcd3Wrapper
from ucloud_common.vm import VmPool
from ucloud_common.host import HostPool
from ucloud_common.request import RequestPool
from decouple import config
WITHOUT_CEPH = config("WITHOUT_CEPH", False, cast=bool)
logging.basicConfig(
level=logging.DEBUG,
filename="log.txt",
filemode="a",
format="%(asctime)s: %(levelname)s - %(message)s",
datefmt="%d-%b-%y %H:%M:%S",
)
etcd_client = Etcd3Wrapper(host=config("ETCD_URL"))
HOST_PREFIX = config("HOST_PREFIX")
VM_PREFIX = config("VM_PREFIX")
REQUEST_PREFIX = config("REQUEST_PREFIX")
VM_DIR = config("VM_DIR")
IMAGE_DIR = config("IMAGE_DIR")
host_pool = HostPool(etcd_client, HOST_PREFIX)
vm_pool = VmPool(etcd_client, VM_PREFIX)
request_pool = RequestPool(etcd_client, REQUEST_PREFIX)
running_vms = []

137
host/main.py Executable file
View file

@ -0,0 +1,137 @@
import argparse
import threading
import time
import os
import sys
import virtualmachine
from ucloud_common.host import HostEntry
from ucloud_common.request import RequestEntry, RequestType
from config import (vm_pool, host_pool, request_pool,
etcd_client, logging, running_vms,
REQUEST_PREFIX, WITHOUT_CEPH)
def update_heartbeat(host: HostEntry):
while True:
host.update_heartbeat()
host_pool.put(host)
time.sleep(10)
logging.info("Updated last heartbeat time %s", host.last_heartbeat)
def maintenance(host):
# To capture vm running according to running_vms list
# This is to capture successful migration of a VM.
# Suppose, this host is running "vm1" and user initiated
# request to migrate this "vm1" to some other host. On,
# successful migration the destination host would set
# the vm hostname to itself. Thus, we are checking
# whether this host vm is successfully migrated. If yes
# then we shutdown "vm1" on this host.
for running_vm in running_vms:
with vm_pool.get_put(running_vm.key) as vm_entry:
if vm_entry.hostname != host.key and not vm_entry.in_migration:
running_vm.handle.shutdown()
vm_entry.add_log("VM on source host shutdown.")
# To check vm running according to etcd entries
alleged_running_vms = vm_pool.by_status("RUNNING", vm_pool.by_host(host.key))
for vm_entry in alleged_running_vms:
_vm = virtualmachine.get_vm(running_vms, vm_entry.key)
# Whether, the allegedly running vm is in our
# running_vms list or not if it is said to be
# running on this host but it is not then we
# need to shut it down
# This is to capture poweroff/shutdown of a VM
# initiated by user inside VM. OR crash of VM by some
# user running process
if (_vm and not _vm.handle.is_running()) or not _vm:
vm_entry.add_log("""{} is not running but is said to be running.
So, shutting it down and declare it killed""".format(vm_entry.key))
vm_entry.declare_killed()
vm_pool.put(vm_entry)
if _vm:
running_vms.remove(_vm)
def main():
argparser = argparse.ArgumentParser()
argparser.add_argument("hostname", help="Name of this host. e.g /v1/host/1")
args = argparser.parse_args()
host = host_pool.get(args.hostname)
if not host:
print("No Such Host")
exit(1)
if WITHOUT_CEPH and not os.path.isdir("/var/vm"):
print("You have set WITHOUT_CEPH to True. So, the /var/vm must exists. But, it don't")
sys.exit(1)
logging.info("%s Session Started %s", '*' * 5, '*' * 5)
# It is seen that under heavy load, timeout event doesn't come
# in a predictive manner (which is intentional because we give
# higher priority to customer's requests) which delays heart
# beat update which in turn misunderstood by scheduler that the
# host is dead when it is actually alive. So, to ensure that we
# update the heart beat in a predictive manner we start Heart
# beat updating mechanism in separated thread
heartbeat_updating_thread = threading.Thread(target=update_heartbeat, args=(host,))
try:
heartbeat_updating_thread.start()
except Exception as e:
logging.info("No Need To Go Further. Our heartbeat updating mechanism is not working")
logging.exception(e)
exit(-1)
for events_iterator in [
etcd_client.get_prefix(REQUEST_PREFIX, value_in_json=True),
etcd_client.watch_prefix(REQUEST_PREFIX, timeout=10, value_in_json=True),
]:
for request_event in events_iterator:
request_event = RequestEntry(request_event)
if request_event.type == "TIMEOUT":
logging.info("Timeout Event")
maintenance(host)
continue
# If the event is directed toward me OR I am destination of a InitVMMigration
if (request_event.hostname == host.key or request_event.destination == host.key):
logging.debug("EVENT: %s", request_event)
request_pool.client.client.delete(request_event.key)
vm_entry = vm_pool.get(request_event.uuid)
if vm_entry:
if request_event.type == RequestType.StartVM:
virtualmachine.start(vm_entry)
elif request_event.type == RequestType.StopVM:
virtualmachine.stop(vm_entry)
elif request_event.type == RequestType.DeleteVM:
virtualmachine.delete(vm_entry)
elif request_event.type == RequestType.InitVMMigration:
virtualmachine.init_migration(vm_entry, host.key)
elif request_event.type == RequestType.TransferVM:
virtualmachine.transfer(request_event)
else:
logging.info("VM Entry missing")
logging.info("Running VMs %s", running_vms)
main()

534
host/qmp/__init__.py Executable file
View file

@ -0,0 +1,534 @@
# QEMU library
#
# Copyright (C) 2015-2016 Red Hat Inc.
# Copyright (C) 2012 IBM Corp.
#
# Authors:
# Fam Zheng <famz@redhat.com>
#
# This work is licensed under the terms of the GNU GPL, version 2. See
# the COPYING file in the top-level directory.
#
# Based on qmp.py.
#
import errno
import logging
import os
import subprocess
import re
import shutil
import socket
import tempfile
from . import qmp
LOG = logging.getLogger(__name__)
# Mapping host architecture to any additional architectures it can
# support which often includes its 32 bit cousin.
ADDITIONAL_ARCHES = {
"x86_64" : "i386",
"aarch64" : "armhf"
}
def kvm_available(target_arch=None):
host_arch = os.uname()[4]
if target_arch and target_arch != host_arch:
if target_arch != ADDITIONAL_ARCHES.get(host_arch):
return False
return os.access("/dev/kvm", os.R_OK | os.W_OK)
class QEMUMachineError(Exception):
"""
Exception called when an error in QEMUMachine happens.
"""
class QEMUMachineAddDeviceError(QEMUMachineError):
"""
Exception raised when a request to add a device can not be fulfilled
The failures are caused by limitations, lack of information or conflicting
requests on the QEMUMachine methods. This exception does not represent
failures reported by the QEMU binary itself.
"""
class MonitorResponseError(qmp.QMPError):
"""
Represents erroneous QMP monitor reply
"""
def __init__(self, reply):
try:
desc = reply["error"]["desc"]
except KeyError:
desc = reply
super(MonitorResponseError, self).__init__(desc)
self.reply = reply
class QEMUMachine(object):
"""
A QEMU VM
Use this object as a context manager to ensure the QEMU process terminates::
with VM(binary) as vm:
...
# vm is guaranteed to be shut down here
"""
def __init__(self, binary, args=None, wrapper=None, name=None,
test_dir="/var/tmp", monitor_address=None,
socket_scm_helper=None):
'''
Initialize a QEMUMachine
@param binary: path to the qemu binary
@param args: list of extra arguments
@param wrapper: list of arguments used as prefix to qemu binary
@param name: prefix for socket and log file names (default: qemu-PID)
@param test_dir: where to create socket and log file
@param monitor_address: address for QMP monitor
@param socket_scm_helper: helper program, required for send_fd_scm()
@note: Qemu process is not started until launch() is used.
'''
if args is None:
args = []
if wrapper is None:
wrapper = []
if name is None:
name = "qemu-%d" % os.getpid()
self._name = name
self._monitor_address = monitor_address
self._vm_monitor = None
self._qemu_log_path = None
self._qemu_log_file = None
self._popen = None
self._binary = binary
self._args = list(args) # Force copy args in case we modify them
self._wrapper = wrapper
self._events = []
self._iolog = None
self._socket_scm_helper = socket_scm_helper
self._qmp = None
self._qemu_full_args = None
self._test_dir = test_dir
self._temp_dir = None
self._launched = False
self._machine = None
self._console_set = False
self._console_device_type = None
self._console_address = None
self._console_socket = None
# just in case logging wasn't configured by the main script:
logging.basicConfig(level=logging.DEBUG)
def __enter__(self):
return self
def __exit__(self, exc_type, exc_val, exc_tb):
self.shutdown()
return False
# This can be used to add an unused monitor instance.
def add_monitor_null(self):
self._args.append('-monitor')
self._args.append('null')
def add_fd(self, fd, fdset, opaque, opts=''):
"""
Pass a file descriptor to the VM
"""
options = ['fd=%d' % fd,
'set=%d' % fdset,
'opaque=%s' % opaque]
if opts:
options.append(opts)
# This did not exist before 3.4, but since then it is
# mandatory for our purpose
if hasattr(os, 'set_inheritable'):
os.set_inheritable(fd, True)
self._args.append('-add-fd')
self._args.append(','.join(options))
return self
# Exactly one of fd and file_path must be given.
# (If it is file_path, the helper will open that file and pass its
# own fd)
def send_fd_scm(self, fd=None, file_path=None):
# In iotest.py, the qmp should always use unix socket.
assert self._qmp.is_scm_available()
if self._socket_scm_helper is None:
raise QEMUMachineError("No path to socket_scm_helper set")
if not os.path.exists(self._socket_scm_helper):
raise QEMUMachineError("%s does not exist" %
self._socket_scm_helper)
# This did not exist before 3.4, but since then it is
# mandatory for our purpose
if hasattr(os, 'set_inheritable'):
os.set_inheritable(self._qmp.get_sock_fd(), True)
if fd is not None:
os.set_inheritable(fd, True)
fd_param = ["%s" % self._socket_scm_helper,
"%d" % self._qmp.get_sock_fd()]
if file_path is not None:
assert fd is None
fd_param.append(file_path)
else:
assert fd is not None
fd_param.append(str(fd))
devnull = open(os.path.devnull, 'rb')
proc = subprocess.Popen(fd_param, stdin=devnull, stdout=subprocess.PIPE,
stderr=subprocess.STDOUT, close_fds=False)
output = proc.communicate()[0]
if output:
LOG.debug(output)
return proc.returncode
@staticmethod
def _remove_if_exists(path):
"""
Remove file object at path if it exists
"""
try:
os.remove(path)
except OSError as exception:
if exception.errno == errno.ENOENT:
return
raise
def is_running(self):
return self._popen is not None and self._popen.poll() is None
def exitcode(self):
if self._popen is None:
return None
return self._popen.poll()
def get_pid(self):
if not self.is_running():
return None
return self._popen.pid
def _load_io_log(self):
if self._qemu_log_path is not None:
with open(self._qemu_log_path, "r") as iolog:
self._iolog = iolog.read()
def _base_args(self):
if isinstance(self._monitor_address, tuple):
moncdev = "socket,id=mon,host=%s,port=%s" % (
self._monitor_address[0],
self._monitor_address[1])
else:
moncdev = 'socket,id=mon,path=%s' % self._vm_monitor
args = ['-chardev', moncdev,
'-mon', 'chardev=mon,mode=control']
if self._machine is not None:
args.extend(['-machine', self._machine])
if self._console_set:
self._console_address = os.path.join(self._temp_dir,
self._name + "-console.sock")
chardev = ('socket,id=console,path=%s,server,nowait' %
self._console_address)
args.extend(['-chardev', chardev])
if self._console_device_type is None:
args.extend(['-serial', 'chardev:console'])
else:
device = '%s,chardev=console' % self._console_device_type
args.extend(['-device', device])
return args
def _pre_launch(self):
self._temp_dir = tempfile.mkdtemp(dir=self._test_dir)
if self._monitor_address is not None:
self._vm_monitor = self._monitor_address
else:
self._vm_monitor = os.path.join(self._temp_dir,
self._name + "-monitor.sock")
self._qemu_log_path = os.path.join(self._temp_dir, self._name + ".log")
self._qemu_log_file = open(self._qemu_log_path, 'wb')
self._qmp = qmp.QEMUMonitorProtocol(self._vm_monitor,
server=True)
def _post_launch(self):
self._qmp.accept()
def _post_shutdown(self):
if self._qemu_log_file is not None:
self._qemu_log_file.close()
self._qemu_log_file = None
self._qemu_log_path = None
if self._console_socket is not None:
self._console_socket.close()
self._console_socket = None
if self._temp_dir is not None:
shutil.rmtree(self._temp_dir)
self._temp_dir = None
def launch(self):
"""
Launch the VM and make sure we cleanup and expose the
command line/output in case of exception
"""
if self._launched:
raise QEMUMachineError('VM already launched')
self._iolog = None
self._qemu_full_args = None
try:
self._launch()
self._launched = True
except:
self.shutdown()
LOG.debug('Error launching VM')
if self._qemu_full_args:
LOG.debug('Command: %r', ' '.join(self._qemu_full_args))
if self._iolog:
LOG.debug('Output: %r', self._iolog)
raise
def _launch(self):
"""
Launch the VM and establish a QMP connection
"""
devnull = open(os.path.devnull, 'rb')
self._pre_launch()
self._qemu_full_args = (self._wrapper + [self._binary] +
self._base_args() + self._args)
LOG.debug('VM launch command: %r', ' '.join(self._qemu_full_args))
self._popen = subprocess.Popen(self._qemu_full_args,
stdin=devnull,
stdout=self._qemu_log_file,
stderr=subprocess.STDOUT,
shell=False,
close_fds=False)
self._post_launch()
def wait(self):
"""
Wait for the VM to power off
"""
self._popen.wait()
self._qmp.close()
self._load_io_log()
self._post_shutdown()
def shutdown(self):
"""
Terminate the VM and clean up
"""
if self.is_running():
try:
self._qmp.cmd('quit')
self._qmp.close()
except:
self._popen.kill()
self._popen.wait()
self._load_io_log()
self._post_shutdown()
exitcode = self.exitcode()
if exitcode is not None and exitcode < 0:
msg = 'qemu received signal %i: %s'
if self._qemu_full_args:
command = ' '.join(self._qemu_full_args)
else:
command = ''
LOG.warn(msg, -exitcode, command)
self._launched = False
def qmp(self, cmd, conv_keys=True, **args):
"""
Invoke a QMP command and return the response dict
"""
qmp_args = dict()
for key, value in args.items():
if conv_keys:
qmp_args[key.replace('_', '-')] = value
else:
qmp_args[key] = value
return self._qmp.cmd(cmd, args=qmp_args)
def command(self, cmd, conv_keys=True, **args):
"""
Invoke a QMP command.
On success return the response dict.
On failure raise an exception.
"""
reply = self.qmp(cmd, conv_keys, **args)
if reply is None:
raise qmp.QMPError("Monitor is closed")
if "error" in reply:
raise MonitorResponseError(reply)
return reply["return"]
def get_qmp_event(self, wait=False):
"""
Poll for one queued QMP events and return it
"""
if len(self._events) > 0:
return self._events.pop(0)
return self._qmp.pull_event(wait=wait)
def get_qmp_events(self, wait=False):
"""
Poll for queued QMP events and return a list of dicts
"""
events = self._qmp.get_events(wait=wait)
events.extend(self._events)
del self._events[:]
self._qmp.clear_events()
return events
@staticmethod
def event_match(event, match=None):
"""
Check if an event matches optional match criteria.
The match criteria takes the form of a matching subdict. The event is
checked to be a superset of the subdict, recursively, with matching
values whenever the subdict values are not None.
This has a limitation that you cannot explicitly check for None values.
Examples, with the subdict queries on the left:
- None matches any object.
- {"foo": None} matches {"foo": {"bar": 1}}
- {"foo": None} matches {"foo": 5}
- {"foo": {"abc": None}} does not match {"foo": {"bar": 1}}
- {"foo": {"rab": 2}} matches {"foo": {"bar": 1, "rab": 2}}
"""
if match is None:
return True
try:
for key in match:
if key in event:
if not QEMUMachine.event_match(event[key], match[key]):
return False
else:
return False
return True
except TypeError:
# either match or event wasn't iterable (not a dict)
return match == event
def event_wait(self, name, timeout=60.0, match=None):
"""
event_wait waits for and returns a named event from QMP with a timeout.
name: The event to wait for.
timeout: QEMUMonitorProtocol.pull_event timeout parameter.
match: Optional match criteria. See event_match for details.
"""
return self.events_wait([(name, match)], timeout)
def events_wait(self, events, timeout=60.0):
"""
events_wait waits for and returns a named event from QMP with a timeout.
events: a sequence of (name, match_criteria) tuples.
The match criteria are optional and may be None.
See event_match for details.
timeout: QEMUMonitorProtocol.pull_event timeout parameter.
"""
def _match(event):
for name, match in events:
if (event['event'] == name and
self.event_match(event, match)):
return True
return False
# Search cached events
for event in self._events:
if _match(event):
self._events.remove(event)
return event
# Poll for new events
while True:
event = self._qmp.pull_event(wait=timeout)
if _match(event):
return event
self._events.append(event)
return None
def get_log(self):
"""
After self.shutdown or failed qemu execution, this returns the output
of the qemu process.
"""
return self._iolog
def add_args(self, *args):
"""
Adds to the list of extra arguments to be given to the QEMU binary
"""
self._args.extend(args)
def set_machine(self, machine_type):
"""
Sets the machine type
If set, the machine type will be added to the base arguments
of the resulting QEMU command line.
"""
self._machine = machine_type
def set_console(self, device_type=None):
"""
Sets the device type for a console device
If set, the console device and a backing character device will
be added to the base arguments of the resulting QEMU command
line.
This is a convenience method that will either use the provided
device type, or default to a "-serial chardev:console" command
line argument.
The actual setting of command line arguments will be be done at
machine launch time, as it depends on the temporary directory
to be created.
@param device_type: the device type, such as "isa-serial". If
None is given (the default value) a "-serial
chardev:console" command line argument will
be used instead, resorting to the machine's
default device type.
"""
self._console_set = True
self._console_device_type = device_type
@property
def console_socket(self):
"""
Returns a socket connected to the console
"""
if self._console_socket is None:
self._console_socket = socket.socket(socket.AF_UNIX,
socket.SOCK_STREAM)
self._console_socket.connect(self._console_address)
return self._console_socket

BIN
host/qmp/__pycache__/__init__.cpython-37.pyc Executable file
View file

Binary file not shown.

BIN
host/qmp/__pycache__/qmp.cpython-37.pyc Executable file
View file

Binary file not shown.

256
host/qmp/qmp.py Executable file
View file

@ -0,0 +1,256 @@
# QEMU Monitor Protocol Python class
#
# Copyright (C) 2009, 2010 Red Hat Inc.
#
# Authors:
# Luiz Capitulino <lcapitulino@redhat.com>
#
# This work is licensed under the terms of the GNU GPL, version 2. See
# the COPYING file in the top-level directory.
import json
import errno
import socket
import logging
class QMPError(Exception):
pass
class QMPConnectError(QMPError):
pass
class QMPCapabilitiesError(QMPError):
pass
class QMPTimeoutError(QMPError):
pass
class QEMUMonitorProtocol(object):
#: Logger object for debugging messages
logger = logging.getLogger('QMP')
#: Socket's error class
error = socket.error
#: Socket's timeout
timeout = socket.timeout
def __init__(self, address, server=False):
"""
Create a QEMUMonitorProtocol class.
@param address: QEMU address, can be either a unix socket path (string)
or a tuple in the form ( address, port ) for a TCP
connection
@param server: server mode listens on the socket (bool)
@raise socket.error on socket connection errors
@note No connection is established, this is done by the connect() or
accept() methods
"""
self.__events = []
self.__address = address
self.__sock = self.__get_sock()
self.__sockfile = None
if server:
self.__sock.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
self.__sock.bind(self.__address)
self.__sock.listen(1)
def __get_sock(self):
if isinstance(self.__address, tuple):
family = socket.AF_INET
else:
family = socket.AF_UNIX
return socket.socket(family, socket.SOCK_STREAM)
def __negotiate_capabilities(self):
greeting = self.__json_read()
if greeting is None or "QMP" not in greeting:
raise QMPConnectError
# Greeting seems ok, negotiate capabilities
resp = self.cmd('qmp_capabilities')
if "return" in resp:
return greeting
raise QMPCapabilitiesError
def __json_read(self, only_event=False):
while True:
data = self.__sockfile.readline()
if not data:
return
resp = json.loads(data)
if 'event' in resp:
self.logger.debug("<<< %s", resp)
self.__events.append(resp)
if not only_event:
continue
return resp
def __get_events(self, wait=False):
"""
Check for new events in the stream and cache them in __events.
@param wait (bool): block until an event is available.
@param wait (float): If wait is a float, treat it as a timeout value.
@raise QMPTimeoutError: If a timeout float is provided and the timeout
period elapses.
@raise QMPConnectError: If wait is True but no events could be
retrieved or if some other error occurred.
"""
# Check for new events regardless and pull them into the cache:
self.__sock.setblocking(0)
try:
self.__json_read()
except socket.error as err:
if err[0] == errno.EAGAIN:
# No data available
pass
self.__sock.setblocking(1)
# Wait for new events, if needed.
# if wait is 0.0, this means "no wait" and is also implicitly false.
if not self.__events and wait:
if isinstance(wait, float):
self.__sock.settimeout(wait)
try:
ret = self.__json_read(only_event=True)
except socket.timeout:
raise QMPTimeoutError("Timeout waiting for event")
except:
raise QMPConnectError("Error while reading from socket")
if ret is None:
raise QMPConnectError("Error while reading from socket")
self.__sock.settimeout(None)
def connect(self, negotiate=True):
"""
Connect to the QMP Monitor and perform capabilities negotiation.
@return QMP greeting dict
@raise socket.error on socket connection errors
@raise QMPConnectError if the greeting is not received
@raise QMPCapabilitiesError if fails to negotiate capabilities
"""
self.__sock.connect(self.__address)
self.__sockfile = self.__sock.makefile()
if negotiate:
return self.__negotiate_capabilities()
def accept(self):
"""
Await connection from QMP Monitor and perform capabilities negotiation.
@return QMP greeting dict
@raise socket.error on socket connection errors
@raise QMPConnectError if the greeting is not received
@raise QMPCapabilitiesError if fails to negotiate capabilities
"""
self.__sock.settimeout(15)
self.__sock, _ = self.__sock.accept()
self.__sockfile = self.__sock.makefile()
return self.__negotiate_capabilities()
def cmd_obj(self, qmp_cmd):
"""
Send a QMP command to the QMP Monitor.
@param qmp_cmd: QMP command to be sent as a Python dict
@return QMP response as a Python dict or None if the connection has
been closed
"""
self.logger.debug(">>> %s", qmp_cmd)
try:
self.__sock.sendall(json.dumps(qmp_cmd).encode('utf-8'))
except socket.error as err:
if err[0] == errno.EPIPE:
return
raise socket.error(err)
resp = self.__json_read()
self.logger.debug("<<< %s", resp)
return resp
def cmd(self, name, args=None, cmd_id=None):
"""
Build a QMP command and send it to the QMP Monitor.
@param name: command name (string)
@param args: command arguments (dict)
@param cmd_id: command id (dict, list, string or int)
"""
qmp_cmd = {'execute': name}
if args:
qmp_cmd['arguments'] = args
if cmd_id:
qmp_cmd['id'] = cmd_id
return self.cmd_obj(qmp_cmd)
def command(self, cmd, **kwds):
"""
Build and send a QMP command to the monitor, report errors if any
"""
ret = self.cmd(cmd, kwds)
if "error" in ret:
raise Exception(ret['error']['desc'])
return ret['return']
def pull_event(self, wait=False):
"""
Pulls a single event.
@param wait (bool): block until an event is available.
@param wait (float): If wait is a float, treat it as a timeout value.
@raise QMPTimeoutError: If a timeout float is provided and the timeout
period elapses.
@raise QMPConnectError: If wait is True but no events could be
retrieved or if some other error occurred.
@return The first available QMP event, or None.
"""
self.__get_events(wait)
if self.__events:
return self.__events.pop(0)
return None
def get_events(self, wait=False):
"""
Get a list of available QMP events.
@param wait (bool): block until an event is available.
@param wait (float): If wait is a float, treat it as a timeout value.
@raise QMPTimeoutError: If a timeout float is provided and the timeout
period elapses.
@raise QMPConnectError: If wait is True but no events could be
retrieved or if some other error occurred.
@return The list of available QMP events.
"""
self.__get_events(wait)
return self.__events
def clear_events(self):
"""
Clear current list of pending events.
"""
self.__events = []
def close(self):
self.__sock.close()
self.__sockfile.close()
def settimeout(self, timeout):
self.__sock.settimeout(timeout)
def get_sock_fd(self):
return self.__sock.fileno()
def is_scm_available(self):
return self.__sock.family == socket.AF_UNIX

310
host/virtualmachine.py Executable file
View file

@ -0,0 +1,310 @@
# QEMU Manual
# https://qemu.weilnetz.de/doc/qemu-doc.html
# For QEMU Monitor Protocol Commands Information, See
# https://qemu.weilnetz.de/doc/qemu-doc.html#pcsys_005fmonitor
import errno
import os
import subprocess
import tempfile
import time
from functools import wraps
from os.path import join
from typing import Union
import bitmath
import sshtunnel
from decouple import config
import qmp
from config import (WITHOUT_CEPH, VM_PREFIX, VM_DIR, IMAGE_DIR,
etcd_client, logging, request_pool,
running_vms, vm_pool)
from ucloud_common.helpers import get_ipv4_address
from ucloud_common.request import RequestEntry, RequestType
from ucloud_common.vm import VMEntry, VMStatus
class VM:
def __init__(self, key, handle, vnc_socket_file):
self.key = key # type: str
self.handle = handle # type: qmp.QEMUMachine
self.vnc_socket_file = vnc_socket_file # type: tempfile.NamedTemporaryFile
def __repr__(self):
return "VM({})".format(self.key)
def get_start_command_args(
vm_entry, vnc_sock_filename: str, migration=False, migration_port=4444
):
threads_per_core = 1
vm_memory = int(bitmath.parse_string(vm_entry.specs["ram"]).to_MB())
vm_cpus = int(vm_entry.specs["cpu"])
vm_uuid = vm_entry.uuid
if WITHOUT_CEPH:
command = "-drive file={},format=raw,if=virtio,cache=none".format(
os.path.join(VM_DIR, vm_uuid)
)
else:
command = "-drive file=rbd:uservms/{},format=raw,if=virtio,cache=none".format(
vm_uuid
)
command += " -device virtio-rng-pci -vnc unix:{}".format(vnc_sock_filename)
command += " -m {} -smp cores={},threads={}".format(
vm_memory, vm_cpus, threads_per_core
)
command += " -name {}".format(vm_uuid)
if migration:
command += " -incoming tcp:0:{}".format(migration_port)
command += " -nic tap,model=virtio,mac={}".format(vm_entry.mac)
return command.split(" ")
def create_vm_object(vm_entry, migration=False, migration_port=4444):
# NOTE: If migration suddenly stop working, having different
# VNC unix filename on source and destination host can
# be a possible cause of it.
# REQUIREMENT: Use Unix Socket instead of TCP Port for VNC
vnc_sock_file = tempfile.NamedTemporaryFile()
qemu_args = get_start_command_args(
vm_entry=vm_entry,
vnc_sock_filename=vnc_sock_file.name,
migration=migration,
migration_port=migration_port,
)
qemu_machine = qmp.QEMUMachine("/usr/bin/qemu-system-x86_64", args=qemu_args)
return VM(vm_entry.key, qemu_machine, vnc_sock_file)
def get_vm(vm_list: list, vm_key) -> Union[VM, None]:
return next((vm for vm in vm_list if vm.key == vm_key), None)
def need_running_vm(func):
@wraps(func)
def wrapper(e):
vm = get_vm(running_vms, e.key)
if vm:
try:
status = vm.handle.command("query-status")
logging.debug("VM Status Check - %s", status)
except Exception as exception:
logging.info("%s failed - VM %s %s", func.__name__, e, exception)
else:
return func(e)
return None
else:
logging.info("%s failed because VM %s is not running", func.__name__, e.key)
return None
return wrapper
def create(vm_entry: VMEntry):
vm_hdd = int(bitmath.parse_string(vm_entry.specs["os-ssd"]).to_MB())
if WITHOUT_CEPH:
_command_to_create = [
"cp",
os.path.join(IMAGE_DIR, vm_entry.image_uuid),
os.path.join(VM_DIR, vm_entry.uuid),
]
_command_to_extend = [
"qemu-img",
"resize",
"-f", "raw",
os.path.join(VM_DIR, vm_entry.uuid),
"{}M".format(vm_hdd),
]
else:
_command_to_create = [
"rbd",
"clone",
"images/{}@protected".format(vm_entry.image_uuid),
"uservms/{}".format(vm_entry.uuid),
]
_command_to_extend = [
"rbd",
"resize",
"uservms/{}".format(vm_entry.uuid),
"--size",
vm_hdd,
]
try:
subprocess.check_output(_command_to_create)
except subprocess.CalledProcessError as e:
if e.returncode == errno.EEXIST:
logging.debug("Image for vm %s exists", vm_entry.uuid)
# File Already exists. No Problem Continue
return
# This exception catches all other exceptions
# i.e FileNotFound (BaseImage), pool Does Not Exists etc.
logging.exception(e)
vm_entry.status = "ERROR"
else:
try:
subprocess.check_output(_command_to_extend)
except Exception as e:
logging.exception(e)
else:
logging.info("New VM Created")
def start(vm_entry: VMEntry):
_vm = get_vm(running_vms, vm_entry.key)
# VM already running. No need to proceed further.
if _vm:
logging.info("VM %s already running", vm_entry.uuid)
return
else:
create(vm_entry)
launch_vm(vm_entry)
@need_running_vm
def stop(vm_entry):
vm = get_vm(running_vms, vm_entry.key)
vm.handle.shutdown()
if not vm.handle.is_running():
vm_entry.add_log("Shutdown successfully")
vm_entry.declare_stopped()
vm_pool.put(vm_entry)
running_vms.remove(vm)
def delete(vm_entry):
logging.info("Deleting VM | %s", vm_entry)
stop(vm_entry)
path_without_protocol = vm_entry.path[vm_entry.path.find(":") + 1 :]
if WITHOUT_CEPH:
vm_deletion_command = ["rm", os.path.join(VM_DIR, vm_entry.uuid)]
else:
vm_deletion_command = ["rbd", "rm", path_without_protocol]
try:
subprocess.check_output(vm_deletion_command)
except Exception as e:
logging.exception(e)
else:
etcd_client.client.delete(vm_entry.key)
def transfer(request_event):
# This function would run on source host i.e host on which the vm
# is running initially. This host would be responsible for transferring
# vm state to destination host.
_host, _port = request_event.parameters["host"], request_event.parameters["port"]
_uuid = request_event.uuid
_destination = request_event.destination_host_key
vm = get_vm(running_vms, join(VM_PREFIX, _uuid))
if vm:
tunnel = sshtunnel.SSHTunnelForwarder(
(_host, 22),
ssh_username=config("ssh_username"),
ssh_pkey=config("ssh_pkey"),
ssh_private_key_password=config("ssh_private_key_password"),
remote_bind_address=("127.0.0.1", _port),
)
try:
tunnel.start()
except sshtunnel.BaseSSHTunnelForwarderError:
logging.exception("Couldn't establish connection to (%s, 22)", _host)
else:
vm.handle.command(
"migrate", uri="tcp:{}:{}".format(_host, tunnel.local_bind_port)
)
status = vm.handle.command("query-migrate")["status"]
while status not in ["failed", "completed"]:
time.sleep(2)
status = vm.handle.command("query-migrate")["status"]
with vm_pool.get_put(request_event.uuid) as source_vm:
if status == "failed":
source_vm.add_log("Migration Failed")
elif status == "completed":
# If VM is successfully migrated then shutdown the VM
# on this host and update hostname to destination host key
source_vm.add_log("Successfully migrated")
source_vm.hostname = _destination
running_vms.remove(vm)
vm.handle.shutdown()
source_vm.in_migration = False # VM transfer finished
finally:
tunnel.close()
def init_migration(vm_entry, destination_host_key):
# This function would run on destination host i.e host on which the vm
# would be transferred after migration.
# This host would be responsible for starting VM that would receive
# state of VM running on source host.
_vm = get_vm(running_vms, vm_entry.key)
if _vm:
# VM already running. No need to proceed further.
logging.info("%s Already running", _vm.key)
return
launch_vm(vm_entry, migration=True, migration_port=4444,
destination_host_key=destination_host_key)
def launch_vm(vm_entry, migration=False, migration_port=None, destination_host_key=None):
logging.info("Starting %s", vm_entry.key)
vm = create_vm_object(vm_entry, migration=migration, migration_port=migration_port)
try:
vm.handle.launch()
except Exception as e:
logging.exception(e)
if migration:
# We don't care whether MachineError or any other error occurred
vm.handle.shutdown()
else:
# Error during typical launch of a vm
vm_entry.add_log("Error Occurred while starting VM")
vm_entry.declare_killed()
vm_pool.put(vm_entry)
else:
vm_entry.vnc_socket = vm.vnc_socket_file.name
running_vms.append(vm)
if migration:
vm_entry.in_migration = True
r = RequestEntry.from_scratch(
type=RequestType.TransferVM,
hostname=vm_entry.hostname,
parameters={"host": get_ipv4_address(), "port": 4444},
uuid=vm_entry.uuid,
destination_host_key=destination_host_key,
)
request_pool.put(r)
else:
# Typical launching of a vm
vm_entry.status = VMStatus.running
vm_entry.add_log("Started successfully")
vm_pool.put(vm_entry)