服务启动后，用户就可以像proxy发送请求了，我们都知道proxy会转发请求到相应的object-server上，进行处理。一个请求来了以后主要的流程会通过鉴权，然后server.py中的handle_request()方法会判断鉴权结果，如果通过了鉴权，执行相应的请求，下面我们就通过一个PUT请求的例子（上传文件）来分析服务转发的过程。
服务转发模型
对于一个PUT请求，proxy判断是什么controlller(object,container,account)，之后调用相应controller的PUT ,比如你PUT一个object，就会执行obj.py下的PUT方法。
PUT请求主要的功能大概分三个部分，第一部分，会解析请求，进行相应的设置，第二部分转发请求，第三部分，根据转发请求的到的响应，制作最终的响应，然后返回。
如图根据你要上传的object的account container object ，通过ring文件会返回三个node,
然后生成这个三个node的conn，然后每个conn初始化一个Queue和一个线程函数，
然后程序会想管道中put数据，然后线程函数会get 数据，然后转发。
数据全部转发后，跳出put操作，接受响应，然后进入第三步。
源码：
Application.__call__():
def __call__(self, env, start_response):
"""
WSGI entry point.
Wraps env in webob.Request object and passes it down.
:param env: WSGI environment dictionary
:param start_response: WSGI callable
"""
try:
if self.memcache is None:
self.memcache = cache_from_env(env)#如果缓存为空，缓存env
req = self.update_request(Request(env))
return self.handle_request(req)(env, start_response)#请求处理函数
except UnicodeError:
err = HTTPPreconditionFailed(request=req, body='Invalid UTF8')
return err(env, start_response)
except (Exception, Timeout):
start_response('500 Server Error',
[('Content-Type', 'text/plain')])
return ['Internal server error.\n']重点来了 handle_request()
def handle_request(self, req):
"""
Entry point for proxy server.
Should return a WSGI-style callable (such as webob.Response).
:param req: webob.Request object
"""
try:
self.logger.set_statsd_prefix('proxy-server')
if req.content_length and req.content_length < 0:#content_length存在倒是小于0 返回错误。
self.logger.increment('errors')
return HTTPBadRequest(request=req,
body='Invalid Content-Length')
try:
if not check_utf8(req.path_info):#是utf8返回错误
self.logger.increment('errors')
return HTTPPreconditionFailed(request=req,
body='Invalid UTF8')
except UnicodeError:
self.logger.increment('errors')
return HTTPPreconditionFailed(request=req, body='Invalid UTF8')
try:
controller, path_parts = self.get_controller(req.path)#根据path获取相应的controller,path_parts
p = req.path_info
if isinstance(p, unicode):
p = p.encode('utf-8')
except ValueError:
self.logger.increment('errors')
return HTTPNotFound(request=req)
if not controller:
self.logger.increment('errors')
return HTTPPreconditionFailed(request=req, body='Bad URL')
if self.deny_host_headers and \
req.host.split(':')[0] in self.deny_host_headers:
return HTTPForbidden(request=req, body='Invalid host header')
self.logger.set_statsd_prefix('proxy-server.' +
controller.server_type.lower())
controller = controller(self, **path_parts)#实例化controller
if 'swift.trans_id' not in req.environ:
# if this wasn't set by an earlier middleware, set it now
trans_id = 'tx' + uuid.uuid4().hex
req.environ['swift.trans_id'] = trans_id
self.logger.txn_id = trans_id
req.headers['x-trans-id'] = req.environ['swift.trans_id']
controller.trans_id = req.environ['swift.trans_id']
self.logger.client_ip = get_remote_client(req)
try:
handler = getattr(controller, req.method)#获取处理方法句柄
getattr(handler, 'publicly_accessible')
except AttributeError:
return HTTPMethodNotAllowed(request=req)
if path_parts['version']:
req.path_info_pop()
if 'swift.authorize' in req.environ:#swift.authrize是swift_auth提供的句柄，
# We call authorize before the handler, always. If authorized,
# we remove the swift.authorize hook so isn't ever called
# again. If not authorized, we return the denial unless the
# controller's method indicates it'd like to gather more
# information and try again later.
resp = req.environ['swift.authorize'](req)#如果存在这个句柄调用这个方法
if not resp:
# No resp means authorized, no delayed recheck required.
del req.environ['swift.authorize']#如果鉴权成功，删除环境句柄
else:
# Response indicates denial, but we might delay the denial
# and recheck later. If not delayed, return the error now.
if not getattr(handler, 'delay_denial', None):
return resp
# Save off original request method (GET, POST, etc.) in case it
# gets mutated during handling. This way logging can display the
# method the client actually sent.
req.environ['swift.orig_req_method'] = req.method
return handler(req)#执行方法
except (Exception, Timeout):
self.logger.exception(_('ERROR Unhandled exception in request'))
return HTTPServerError(request=req)最终执行obj.py中的PUT方法。
其中_send_file是转发线程函数，_connect_put_node()方法建立连接。
def _send_file(self, conn, path):
"""Method for a file PUT coro"""
while True:
chunk = conn.queue.get()#从队列里取
if not conn.failed:
try:
with ChunkWriteTimeout(self.app.node_timeout):
conn.send(chunk)#发送
except (Exception, ChunkWriteTimeout):
conn.failed = True
self.exception_occurred(conn.node, _('Object'),
_('Trying to write to %s') % path)
conn.queue.task_done()#一次取任务的结束
def _connect_put_node(self, nodes, part, path, headers,
logger_thread_locals):
"""Method for a file PUT connect"""
self.app.logger.thread_locals = logger_thread_locals
for node in nodes:
try:
with ConnectionTimeout(self.app.conn_timeout):
conn = http_connect(node['ip'], node['port'],
node['device'], part, 'PUT', path, headers)#与相应的node建立连接
with Timeout(self.app.node_timeout):
resp = conn.getexpect()#？
if resp.status == HTTP_CONTINUE:
conn.node = node
return conn#返回连接
elif resp.status == HTTP_INSUFFICIENT_STORAGE:
self.error_limit(node)
except:
self.exception_occurred(node, _('Object'),
_('Expect: 100-continue on %s') % path)
@public
@delay_denial
def PUT(self, req):@PUT方法
"""HTTP PUT request handler."""
(container_partition, containers, _junk, req.acl,
req.environ['swift_sync_key'], object_versions) = \
self.container_info(self.account_name, self.container_name,
account_autocreate=self.app.account_autocreate)#获取相关的container，account信息。
if 'swift.authorize' in req.environ:#????????????????
aresp = req.environ['swift.authorize'](req)
if aresp:
return aresp
if not containers:#没有相应的container返回NOTFOUND
return HTTPNotFound(request=req)
if 'x-delete-after' in req.headers:#设置object的有效期限
try:
x_delete_after = int(req.headers['x-delete-after'])
except ValueError:
return HTTPBadRequest(request=req,
content_type='text/plain',
body='Non-integer X-Delete-After')
req.headers['x-delete-at'] = '%d' % (time.time() + x_delete_after)
if 'x-delete-at' in req.headers:
try:
x_delete_at = int(req.headers['x-delete-at'])
if x_delete_at < time.time():
return HTTPBadRequest(body='X-Delete-At in past',
request=req, content_type='text/plain')
except ValueError:
return HTTPBadRequest(request=req, content_type='text/plain',
body='Non-integer X-Delete-At')
delete_at_container = str(x_delete_at /
self.app.expiring_objects_container_divisor *
self.app.expiring_objects_container_divisor)
delete_at_part, delete_at_nodes = \
self.app.container_ring.get_nodes(
self.app.expiring_objects_account, delete_at_container)
else:
delete_at_part = delete_at_nodes = None
partition, nodes = self.app.object_ring.get_nodes(#从ring文件中找出相应的partition,node。
self.account_name, self.container_name, self.object_name)
# do a HEAD request for container sync and checking object versions#时间戳
if 'x-timestamp' in req.headers or (object_versions and not
req.environ.get('swift_versioned_copy')):
hreq = Request.blank(req.path_info, headers={'X-Newest': 'True'},
environ={'REQUEST_METHOD': 'HEAD'})
hresp = self.GETorHEAD_base(hreq, _('Object'), partition, nodes,
hreq.path_info, len(nodes))
# Used by container sync feature
if 'x-timestamp' in req.headers:
try:
req.headers['X-Timestamp'] = \
normalize_timestamp(float(req.headers['x-timestamp']))
if hresp.environ and 'swift_x_timestamp' in hresp.environ and \
float(hresp.environ['swift_x_timestamp']) >= \
float(req.headers['x-timestamp']):
return HTTPAccepted(request=req)
except ValueError:
return HTTPBadRequest(request=req, content_type='text/plain',
body='X-Timestamp should be a UNIX timestamp float value; '
'was %r' % req.headers['x-timestamp'])
else:
req.headers['X-Timestamp'] = normalize_timestamp(time.time())
# Sometimes the 'content-type' header exists, but is set to None.
content_type_manually_set = True
if not req.headers.get('content-type'):
guessed_type, _junk = mimetypes.guess_type(req.path_info)
req.headers['Content-Type'] = guessed_type or \
'application/octet-stream'
content_type_manually_set = False
error_response = check_object_creation(req, self.object_name)
if error_response:
return error_response
if object_versions and not req.environ.get('swift_versioned_copy'):
is_manifest = 'x-object-manifest' in req.headers or \
'x-object-manifest' in hresp.headers
if hresp.status_int != HTTP_NOT_FOUND and not is_manifest:#判断是否需要调用COPY
# This is a version manifest and needs to be handled
# differently. First copy the existing data to a new object,
# then write the data from this request to the version manifest
# object.
lcontainer = object_versions.split('/')[0]
prefix_len = '%03x' % len(self.object_name)
lprefix = prefix_len + self.object_name + '/'
ts_source = hresp.environ.get('swift_x_timestamp')
if ts_source is None:
ts_source = time.mktime(time.strptime(
hresp.headers['last-modified'],
'%a, %d %b %Y %H:%M:%S GMT'))
new_ts = normalize_timestamp(ts_source)
vers_obj_name = lprefix + new_ts
copy_headers = {
'Destination': '%s/%s' % (lcontainer, vers_obj_name)}
copy_environ = {'REQUEST_METHOD': 'COPY',
'swift_versioned_copy': True
}
copy_req = Request.blank(req.path_info, headers=copy_headers,
environ=copy_environ)
copy_resp = self.COPY(copy_req)
if is_client_error(copy_resp.status_int):
# missing container or bad permissions
return HTTPPreconditionFailed(request=req)
elif not is_success(copy_resp.status_int):
# could not copy the data, bail
return HTTPServiceUnavailable(request=req)
reader = req.environ['wsgi.input'].read
data_source = iter(lambda: reader(self.app.client_chunk_size), '')
source_header = req.headers.get('X-Copy-From')
source_resp = None
if source_header:#如果请求为从COPY-From
source_header = unquote(source_header)
acct = req.path_info.split('/', 2)[1]
if isinstance(acct, unicode):
acct = acct.encode('utf-8')
if not source_header.startswith('/'):
source_header = '/' + source_header
source_header = '/' + acct + source_header
try:
src_container_name, src_obj_name = \
source_header.split('/', 3)[2:]
except ValueError:
return HTTPPreconditionFailed(request=req,
body='X-Copy-From header must be of the form'
'<container name>/<object name>')
source_req = req.copy_get()
source_req.path_info = source_header
source_req.headers['X-Newest'] = 'true'
orig_obj_name = self.object_name
orig_container_name = self.container_name
self.object_name = src_obj_name
self.container_name = src_container_name
source_resp = self.GET(source_req)
if source_resp.status_int >= HTTP_MULTIPLE_CHOICES:
return source_resp
self.object_name = orig_obj_name
self.container_name = orig_container_name
new_req = Request.blank(req.path_info,
environ=req.environ, headers=req.headers)
data_source = source_resp.app_iter
new_req.content_length = source_resp.content_length
if new_req.content_length is None:
# This indicates a transfer-encoding: chunked source object,
# which currently only happens because there are more than
# CONTAINER_LISTING_LIMIT segments in a segmented object. In
# this case, we're going to refuse to do the server-side copy.
return HTTPRequestEntityTooLarge(request=req)
new_req.etag = source_resp.etag
# we no longer need the X-Copy-From header
del new_req.headers['X-Copy-From']
if not content_type_manually_set:
new_req.headers['Content-Type'] = \
source_resp.headers['Content-Type']
if new_req.headers.get('x-fresh-metadata', 'false').lower() \
not in TRUE_VALUES:
for k, v in source_resp.headers.items():
if k.lower().startswith('x-object-meta-'):
new_req.headers[k] = v
for k, v in req.headers.items():
if k.lower().startswith('x-object-meta-'):
new_req.headers[k] = v
req = new_req
node_iter = self.iter_nodes(partition, nodes, self.app.object_ring)
pile = GreenPile(len(nodes))
for container in containers:
nheaders = dict(req.headers.iteritems())
nheaders['Connection'] = 'close'
nheaders['X-Container-Host'] = '%(ip)s:%(port)s' % container
nheaders['X-Container-Partition'] = container_partition
nheaders['X-Container-Device'] = container['device']
nheaders['Expect'] = '100-continue'
if delete_at_nodes:
node = delete_at_nodes.pop(0)
nheaders['X-Delete-At-Host'] = '%(ip)s:%(port)s' % node
nheaders['X-Delete-At-Partition'] = delete_at_part
nheaders['X-Delete-At-Device'] = node['device']
pile.spawn(self._connect_put_node, node_iter, partition,
req.path_info, nheaders, self.app.logger.thread_locals)
conns = [conn for conn in pile if conn]
if len(conns) <= len(nodes) / 2:
self.app.logger.error(
_('Object PUT returning 503, %(conns)s/%(nodes)s '
'required connections'),
{'conns': len(conns), 'nodes': len(nodes) // 2 + 1})
return HTTPServiceUnavailable(request=req)
chunked = req.headers.get('transfer-encoding')
bytes_transferred = 0
try:
with ContextPool(len(nodes)) as pool:#如果就是上传一个新的object，最前的初始化，和逻辑操作做好后，下面是真正的转发请求，使用queue
for conn in conns:
conn.failed = False
conn.queue = Queue(self.app.put_queue_depth)
pool.spawn(self._send_file, conn, req.path)
while True:
with ChunkReadTimeout(self.app.client_timeout):
try:
chunk = next(data_source)
except StopIteration:
if chunked:
[conn.queue.put('0\r\n\r\n') for conn in conns]
break
bytes_transferred += len(chunk)
if bytes_transferred > MAX_FILE_SIZE:
return HTTPRequestEntityTooLarge(request=req)
for conn in list(conns):
if not conn.failed:
conn.queue.put('%x\r\n%s\r\n' % (len(chunk), chunk)
if chunked else chunk)
else:
conns.remove(conn)
if len(conns) <= len(nodes) / 2:
self.app.logger.error(_('Object PUT exceptions during'
' send, %(conns)s/%(nodes)s required connections'),
{'conns': len(conns), 'nodes': len(nodes) / 2 + 1})
return HTTPServiceUnavailable(request=req)
for conn in conns:
if conn.queue.unfinished_tasks:
conn.queue.join()
conns = [conn for conn in conns if not conn.failed]
except ChunkReadTimeout, err:
self.app.logger.warn(
_('ERROR Client read timeout (%ss)'), err.seconds)
self.app.logger.increment('client_timeouts')
return HTTPRequestTimeout(request=req)
except (Exception, Timeout):
self.app.logger.exception(
_('ERROR Exception causing client disconnect'))
return HTTPClientDisconnect(request=req)
if req.content_length and bytes_transferred < req.content_length:
req.client_disconnect = True
self.app.logger.warn(
_('Client disconnected without sending enough data'))
self.app.logger.increment('client_disconnects')
return HTTPClientDisconnect(request=req)
statuses = []#后面的操作，生成响应，然后返回。
reasons = []
bodies = []
etags = set()
for conn in conns:
try:
with Timeout(self.app.node_timeout):
response = conn.getresponse()
statuses.append(response.status)
reasons.append(response.reason)
bodies.append(response.read())
if response.status >= HTTP_INTERNAL_SERVER_ERROR:
self.error_occurred(conn.node,
_('ERROR %(status)d %(body)s From Object Server ' \
're: %(path)s') % {'status': response.status,
'body': bodies[-1][:1024], 'path': req.path})
elif is_success(response.status):
etags.add(response.getheader('etag').strip('"'))
except (Exception, Timeout):
self.exception_occurred(conn.node, _('Object'),
_('Trying to get final status of PUT to %s') % req.path)
if len(etags) > 1:
self.app.logger.error(
_('Object servers returned %s mismatched etags'), len(etags))
return HTTPServerError(request=req)
etag = len(etags) and etags.pop() or None
while len(statuses) < len(nodes):
statuses.append(HTTP_SERVICE_UNAVAILABLE)
reasons.append('')
bodies.append('')
resp = self.best_response(req, statuses, reasons, bodies,
_('Object PUT'), etag=etag)
if source_header:
resp.headers['X-Copied-From'] = quote(
source_header.split('/', 2)[2])
if 'last-modified' in source_resp.headers:
resp.headers['X-Copied-From-Last-Modified'] = \
source_resp.headers['last-modified']
for k, v in req.headers.items():
if k.lower().startswith('x-object-meta-'):
resp.headers[k] = v
resp.last_modified = float(req.headers['X-Timestamp'])
return resp
1847
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
