# Copyright (c) 2010-2012 OpenStack Foundation # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or # implied. # See the License for the specific language governing permissions and # limitations under the License. # NOTE: swift_conn # You'll see swift_conn passed around a few places in this file. This is the # source bufferedhttp connection of whatever it is attached to. # It is used when early termination of reading from the connection should # happen, such as when a range request is satisfied but there's still more the # source connection would like to send. To prevent having to read all the data # that could be left, the source connection can be .close() and then reads # commence to empty out any buffers. # These shenanigans are to ensure all related objects can be garbage # collected. We've seen objects hang around forever otherwise. from six.moves.urllib.parse import quote import os import time import functools import inspect import itertools import operator from sys import exc_info from swift import gettext_ as _ from eventlet import sleep from eventlet.timeout import Timeout import six from swift.common.wsgi import make_pre_authed_env from swift.common.utils import Timestamp, config_true_value, \ public, split_path, list_from_csv, GreenthreadSafeIterator, \ GreenAsyncPile, quorum_size, parse_content_type, \ http_response_to_document_iters, document_iters_to_http_response_body from swift.common.bufferedhttp import http_connect from swift.common.exceptions import ChunkReadTimeout, ChunkWriteTimeout, \ ConnectionTimeout, RangeAlreadyComplete from swift.common.http import is_informational, is_success, is_redirection, \ is_server_error, HTTP_OK, HTTP_PARTIAL_CONTENT, HTTP_MULTIPLE_CHOICES, \ HTTP_BAD_REQUEST, HTTP_NOT_FOUND, HTTP_SERVICE_UNAVAILABLE, \ HTTP_INSUFFICIENT_STORAGE, HTTP_UNAUTHORIZED, HTTP_CONTINUE from swift.common.swob import Request, Response, HeaderKeyDict, Range, \ HTTPException, HTTPRequestedRangeNotSatisfiable, HTTPServiceUnavailable, \ status_map from swift.common.request_helpers import strip_sys_meta_prefix, \ strip_user_meta_prefix, is_user_meta, is_sys_meta, is_sys_or_user_meta from swift.common.storage_policy import POLICIES def update_headers(response, headers): """ Helper function to update headers in the response. :param response: swob.Response object :param headers: dictionary headers """ if hasattr(headers, 'items'): headers = headers.items() for name, value in headers: if name == 'etag': response.headers[name] = value.replace('"', '') elif name not in ('date', 'content-length', 'content-type', 'connection', 'x-put-timestamp', 'x-delete-after'): response.headers[name] = value def source_key(resp): """ Provide the timestamp of the swift http response as a floating point value. Used as a sort key. :param resp: bufferedhttp response object """ return Timestamp(resp.getheader('x-backend-timestamp') or resp.getheader('x-put-timestamp') or resp.getheader('x-timestamp') or 0) def delay_denial(func): """ Decorator to declare which methods should have any swift.authorize call delayed. This is so the method can load the Request object up with additional information that may be needed by the authorization system. :param func: function for which authorization will be delayed """ func.delay_denial = True @functools.wraps(func) def wrapped(*a, **kw): return func(*a, **kw) return wrapped def get_account_memcache_key(account): cache_key, env_key = _get_cache_key(account, None) return cache_key def get_container_memcache_key(account, container): if not container: raise ValueError("container not provided") cache_key, env_key = _get_cache_key(account, container) return cache_key def _prep_headers_to_info(headers, server_type): """ Helper method that iterates once over a dict of headers, converting all keys to lower case and separating into subsets containing user metadata, system metadata and other headers. """ meta = {} sysmeta = {} other = {} for key, val in dict(headers).items(): lkey = key.lower() if is_user_meta(server_type, lkey): meta[strip_user_meta_prefix(server_type, lkey)] = val elif is_sys_meta(server_type, lkey): sysmeta[strip_sys_meta_prefix(server_type, lkey)] = val else: other[lkey] = val return other, meta, sysmeta def headers_to_account_info(headers, status_int=HTTP_OK): """ Construct a cacheable dict of account info based on response headers. """ headers, meta, sysmeta = _prep_headers_to_info(headers, 'account') return { 'status': status_int, # 'container_count' anomaly: # Previous code sometimes expects an int sometimes a string # Current code aligns to str and None, yet translates to int in # deprecated functions as needed 'container_count': headers.get('x-account-container-count'), 'total_object_count': headers.get('x-account-object-count'), 'bytes': headers.get('x-account-bytes-used'), 'meta': meta, 'sysmeta': sysmeta } def headers_to_container_info(headers, status_int=HTTP_OK): """ Construct a cacheable dict of container info based on response headers. """ headers, meta, sysmeta = _prep_headers_to_info(headers, 'container') return { 'status': status_int, 'read_acl': headers.get('x-container-read'), 'write_acl': headers.get('x-container-write'), 'sync_key': headers.get('x-container-sync-key'), 'object_count': headers.get('x-container-object-count'), 'bytes': headers.get('x-container-bytes-used'), 'versions': headers.get('x-versions-location'), 'storage_policy': headers.get('X-Backend-Storage-Policy-Index'.lower(), '0'), 'cors': { 'allow_origin': meta.get('access-control-allow-origin'), 'expose_headers': meta.get('access-control-expose-headers'), 'max_age': meta.get('access-control-max-age') }, 'meta': meta, 'sysmeta': sysmeta } def headers_to_object_info(headers, status_int=HTTP_OK): """ Construct a cacheable dict of object info based on response headers. """ headers, meta, sysmeta = _prep_headers_to_info(headers, 'object') info = {'status': status_int, 'length': headers.get('content-length'), 'type': headers.get('content-type'), 'etag': headers.get('etag'), 'meta': meta, 'sysmeta': sysmeta } return info def cors_validation(func): """ Decorator to check if the request is a CORS request and if so, if it's valid. :param func: function to check """ @functools.wraps(func) def wrapped(*a, **kw): controller = a[0] req = a[1] # The logic here was interpreted from # http://www.w3.org/TR/cors/#resource-requests # Is this a CORS request? req_origin = req.headers.get('Origin', None) if req_origin: # Yes, this is a CORS request so test if the origin is allowed container_info = \ controller.container_info(controller.account_name, controller.container_name, req) cors_info = container_info.get('cors', {}) # Call through to the decorated method resp = func(*a, **kw) if controller.app.strict_cors_mode and \ not controller.is_origin_allowed(cors_info, req_origin): return resp # Expose, # - simple response headers, # http://www.w3.org/TR/cors/#simple-response-header # - swift specific: etag, x-timestamp, x-trans-id # - user metadata headers # - headers provided by the user in # x-container-meta-access-control-expose-headers if 'Access-Control-Expose-Headers' not in resp.headers: expose_headers = [ 'cache-control', 'content-language', 'content-type', 'expires', 'last-modified', 'pragma', 'etag', 'x-timestamp', 'x-trans-id'] for header in resp.headers: if header.startswith('X-Container-Meta') or \ header.startswith('X-Object-Meta'): expose_headers.append(header.lower()) if cors_info.get('expose_headers'): expose_headers.extend( [header_line.strip() for header_line in cors_info['expose_headers'].split(' ') if header_line.strip()]) resp.headers['Access-Control-Expose-Headers'] = \ ', '.join(expose_headers) # The user agent won't process the response if the Allow-Origin # header isn't included if 'Access-Control-Allow-Origin' not in resp.headers: if cors_info['allow_origin'] and \ cors_info['allow_origin'].strip() == '*': resp.headers['Access-Control-Allow-Origin'] = '*' else: resp.headers['Access-Control-Allow-Origin'] = req_origin return resp else: # Not a CORS request so make the call as normal return func(*a, **kw) return wrapped def get_object_info(env, app, path=None, swift_source=None): """ Get the info structure for an object, based on env and app. This is useful to middlewares. .. note:: This call bypasses auth. Success does not imply that the request has authorization to the object. """ (version, account, container, obj) = \ split_path(path or env['PATH_INFO'], 4, 4, True) info = _get_object_info(app, env, account, container, obj, swift_source=swift_source) if not info: info = headers_to_object_info({}, 0) return info def get_container_info(env, app, swift_source=None): """ Get the info structure for a container, based on env and app. This is useful to middlewares. .. note:: This call bypasses auth. Success does not imply that the request has authorization to the container. """ (version, account, container, unused) = \ split_path(env['PATH_INFO'], 3, 4, True) info = get_info(app, env, account, container, ret_not_found=True, swift_source=swift_source) if not info: info = headers_to_container_info({}, 0) info.setdefault('storage_policy', '0') return info def get_account_info(env, app, swift_source=None): """ Get the info structure for an account, based on env and app. This is useful to middlewares. .. note:: This call bypasses auth. Success does not imply that the request has authorization to the account. :raises ValueError: when path can't be split(path, 2, 4) """ (version, account, _junk, _junk) = \ split_path(env['PATH_INFO'], 2, 4, True) info = get_info(app, env, account, ret_not_found=True, swift_source=swift_source) if not info: info = headers_to_account_info({}, 0) if info.get('container_count') is None: info['container_count'] = 0 else: info['container_count'] = int(info['container_count']) return info def _get_cache_key(account, container): """ Get the keys for both memcache (cache_key) and env (env_key) where info about accounts and containers is cached :param account: The name of the account :param container: The name of the container (or None if account) :returns a tuple of (cache_key, env_key) """ if container: cache_key = 'container/%s/%s' % (account, container) else: cache_key = 'account/%s' % account # Use a unique environment cache key per account and one container. # This allows caching both account and container and ensures that when we # copy this env to form a new request, it won't accidentally reuse the # old container or account info env_key = 'swift.%s' % cache_key return cache_key, env_key def get_object_env_key(account, container, obj): """ Get the keys for env (env_key) where info about object is cached :param account: The name of the account :param container: The name of the container :param obj: The name of the object :returns a string env_key """ env_key = 'swift.object/%s/%s/%s' % (account, container, obj) return env_key def _set_info_cache(app, env, account, container, resp): """ Cache info in both memcache and env. Caching is used to avoid unnecessary calls to account & container servers. This is a private function that is being called by GETorHEAD_base and by clear_info_cache. Any attempt to GET or HEAD from the container/account server should use the GETorHEAD_base interface which would than set the cache. :param app: the application object :param account: the unquoted account name :param container: the unquoted container name or None :param resp: the response received or None if info cache should be cleared """ if container: cache_time = app.recheck_container_existence else: cache_time = app.recheck_account_existence cache_key, env_key = _get_cache_key(account, container) if resp: if resp.status_int == HTTP_NOT_FOUND: cache_time *= 0.1 elif not is_success(resp.status_int): cache_time = None else: cache_time = None # Next actually set both memcache and the env cache memcache = getattr(app, 'memcache', None) or env.get('swift.cache') if not cache_time: env.pop(env_key, None) if memcache: memcache.delete(cache_key) return if container: info = headers_to_container_info(resp.headers, resp.status_int) else: info = headers_to_account_info(resp.headers, resp.status_int) if memcache: memcache.set(cache_key, info, time=cache_time) env[env_key] = info def _set_object_info_cache(app, env, account, container, obj, resp): """ Cache object info env. Do not cache object informations in memcache. This is an intentional omission as it would lead to cache pressure. This is a per-request cache. Caching is used to avoid unnecessary calls to object servers. This is a private function that is being called by GETorHEAD_base. Any attempt to GET or HEAD from the object server should use the GETorHEAD_base interface which would then set the cache. :param app: the application object :param account: the unquoted account name :param container: the unquoted container name or None :param object: the unquoted object name or None :param resp: the response received or None if info cache should be cleared """ env_key = get_object_env_key(account, container, obj) if not resp: env.pop(env_key, None) return info = headers_to_object_info(resp.headers, resp.status_int) env[env_key] = info def clear_info_cache(app, env, account, container=None): """ Clear the cached info in both memcache and env :param app: the application object :param account: the account name :param container: the containr name or None if setting info for containers """ _set_info_cache(app, env, account, container, None) def _get_info_cache(app, env, account, container=None): """ Get the cached info from env or memcache (if used) in that order Used for both account and container info A private function used by get_info :param app: the application object :param env: the environment used by the current request :returns the cached info or None if not cached """ cache_key, env_key = _get_cache_key(account, container) if env_key in env: return env[env_key] memcache = getattr(app, 'memcache', None) or env.get('swift.cache') if memcache: info = memcache.get(cache_key) if info: for key in info: if isinstance(info[key], six.text_type): info[key] = info[key].encode("utf-8") env[env_key] = info return info return None def _prepare_pre_auth_info_request(env, path, swift_source): """ Prepares a pre authed request to obtain info using a HEAD. :param env: the environment used by the current request :param path: The unquoted request path :param swift_source: value for swift.source in WSGI environment :returns: the pre authed request """ # Set the env for the pre_authed call without a query string newenv = make_pre_authed_env(env, 'HEAD', path, agent='Swift', query_string='', swift_source=swift_source) # This is a sub request for container metadata- drop the Origin header from # the request so the it is not treated as a CORS request. newenv.pop('HTTP_ORIGIN', None) # Note that Request.blank expects quoted path return Request.blank(quote(path), environ=newenv) def get_info(app, env, account, container=None, ret_not_found=False, swift_source=None): """ Get the info about accounts or containers Note: This call bypasses auth. Success does not imply that the request has authorization to the info. :param app: the application object :param env: the environment used by the current request :param account: The unquoted name of the account :param container: The unquoted name of the container (or None if account) :returns: the cached info or None if cannot be retrieved """ info = _get_info_cache(app, env, account, container) if info: if ret_not_found or is_success(info['status']): return info return None # Not in cache, let's try the account servers path = '/v1/%s' % account if container: # Stop and check if we have an account? if not get_info(app, env, account) and not account.startswith( getattr(app, 'auto_create_account_prefix', '.')): return None path += '/' + container req = _prepare_pre_auth_info_request( env, path, (swift_source or 'GET_INFO')) # Whenever we do a GET/HEAD, the GETorHEAD_base will set the info in # the environment under environ[env_key] and in memcache. We will # pick the one from environ[env_key] and use it to set the caller env resp = req.get_response(app) cache_key, env_key = _get_cache_key(account, container) try: info = resp.environ[env_key] env[env_key] = info if ret_not_found or is_success(info['status']): return info except (KeyError, AttributeError): pass return None def _get_object_info(app, env, account, container, obj, swift_source=None): """ Get the info about object Note: This call bypasses auth. Success does not imply that the request has authorization to the info. :param app: the application object :param env: the environment used by the current request :param account: The unquoted name of the account :param container: The unquoted name of the container :param obj: The unquoted name of the object :returns: the cached info or None if cannot be retrieved """ env_key = get_object_env_key(account, container, obj) info = env.get(env_key) if info: return info # Not in cached, let's try the object servers path = '/v1/%s/%s/%s' % (account, container, obj) req = _prepare_pre_auth_info_request(env, path, swift_source) # Whenever we do a GET/HEAD, the GETorHEAD_base will set the info in # the environment under environ[env_key]. We will # pick the one from environ[env_key] and use it to set the caller env resp = req.get_response(app) try: info = resp.environ[env_key] env[env_key] = info return info except (KeyError, AttributeError): pass return None def close_swift_conn(src): """ Force close the http connection to the backend. :param src: the response from the backend """ try: # Since the backends set "Connection: close" in their response # headers, the response object (src) is solely responsible for the # socket. The connection object (src.swift_conn) has no references # to the socket, so calling its close() method does nothing, and # therefore we don't do it. # # Also, since calling the response's close() method might not # close the underlying socket but only decrement some # reference-counter, we have a special method here that really, # really kills the underlying socket with a close() syscall. src.nuke_from_orbit() # it's the only way to be sure except Exception: pass def bytes_to_skip(record_size, range_start): """ Assume an object is composed of N records, where the first N-1 are all the same size and the last is at most that large, but may be smaller. When a range request is made, it might start with a partial record. This must be discarded, lest the consumer get bad data. This is particularly true of suffix-byte-range requests, e.g. "Range: bytes=-12345" where the size of the object is unknown at the time the request is made. This function computes the number of bytes that must be discarded to ensure only whole records are yielded. Erasure-code decoding needs this. This function could have been inlined, but it took enough tries to get right that some targeted unit tests were desirable, hence its extraction. """ return (record_size - (range_start % record_size)) % record_size class ResumingGetter(object): def __init__(self, app, req, server_type, node_iter, partition, path, backend_headers, client_chunk_size=None, newest=None): self.app = app self.node_iter = node_iter self.server_type = server_type self.partition = partition self.path = path self.backend_headers = backend_headers self.client_chunk_size = client_chunk_size self.skip_bytes = 0 self.used_nodes = [] self.used_source_etag = '' # stuff from request self.req_method = req.method self.req_path = req.path self.req_query_string = req.query_string if newest is None: self.newest = config_true_value(req.headers.get('x-newest', 'f')) else: self.newest = newest # populated when finding source self.statuses = [] self.reasons = [] self.bodies = [] self.source_headers = [] # populated from response headers self.start_byte = self.end_byte = self.length = None def fast_forward(self, num_bytes): """ Will skip num_bytes into the current ranges. :params num_bytes: the number of bytes that have already been read on this request. This will change the Range header so that the next req will start where it left off. :raises ValueError: if invalid range header :raises HTTPRequestedRangeNotSatisfiable: if begin + num_bytes > end of range + 1 :raises RangeAlreadyComplete: if begin + num_bytes == end of range + 1 """ if 'Range' in self.backend_headers: req_range = Range(self.backend_headers['Range']) begin, end = req_range.ranges[0] if begin is None: # this is a -50 range req (last 50 bytes of file) end -= num_bytes else: begin += num_bytes if end and begin == end + 1: # we sent out exactly the first range's worth of bytes, so # we're done with it raise RangeAlreadyComplete() elif end and begin > end: raise HTTPRequestedRangeNotSatisfiable() elif end and begin: req_range.ranges = [(begin, end)] + req_range.ranges[1:] elif end: req_range.ranges = [(None, end)] + req_range.ranges[1:] else: req_range.ranges = [(begin, None)] + req_range.ranges[1:] self.backend_headers['Range'] = str(req_range) else: self.backend_headers['Range'] = 'bytes=%d-' % num_bytes def pop_range(self): """ Remove the first byterange from our Range header. This is used after a byterange has been completely sent to the client; this way, should we need to resume the download from another object server, we do not re-fetch byteranges that the client already has. If we have no Range header, this is a no-op. """ if 'Range' in self.backend_headers: try: req_range = Range(self.backend_headers['Range']) except ValueError: # there's a Range header, but it's garbage, so get rid of it self.backend_headers.pop('Range') return begin, end = req_range.ranges.pop(0) if len(req_range.ranges) > 0: self.backend_headers['Range'] = str(req_range) else: self.backend_headers.pop('Range') def learn_size_from_content_range(self, start, end, length): """ If client_chunk_size is set, makes sure we yield things starting on chunk boundaries based on the Content-Range header in the response. Sets our Range header's first byterange to the value learned from the Content-Range header in the response; if we were given a fully-specified range (e.g. "bytes=123-456"), this is a no-op. If we were given a half-specified range (e.g. "bytes=123-" or "bytes=-456"), then this changes the Range header to a semantically-equivalent one *and* it lets us resume on a proper boundary instead of just in the middle of a piece somewhere. """ if length == 0: return if self.client_chunk_size: self.skip_bytes = bytes_to_skip(self.client_chunk_size, start) if 'Range' in self.backend_headers: try: req_range = Range(self.backend_headers['Range']) new_ranges = [(start, end)] + req_range.ranges[1:] except ValueError: new_ranges = [(start, end)] else: new_ranges = [(start, end)] self.backend_headers['Range'] = ( "bytes=" + (",".join("%s-%s" % (s if s is not None else '', e if e is not None else '') for s, e in new_ranges))) def is_good_source(self, src): """ Indicates whether or not the request made to the backend found what it was looking for. :param src: the response from the backend :returns: True if found, False if not """ if self.server_type == 'Object' and src.status == 416: return True return is_success(src.status) or is_redirection(src.status) def response_parts_iter(self, req): source, node = self._get_source_and_node() it = None if source: it = self._get_response_parts_iter(req, node, source) return it def _get_response_parts_iter(self, req, node, source): # Someday we can replace this [mess] with python 3's "nonlocal" source = [source] node = [node] try: client_chunk_size = self.client_chunk_size node_timeout = self.app.node_timeout if self.server_type == 'Object': node_timeout = self.app.recoverable_node_timeout # This is safe; it sets up a generator but does not call next() # on it, so no IO is performed. parts_iter = [ http_response_to_document_iters( source[0], read_chunk_size=self.app.object_chunk_size)] def get_next_doc_part(): while True: try: # This call to next() performs IO when we have a # multipart/byteranges response; it reads the MIME # boundary and part headers. # # If we don't have a multipart/byteranges response, # but just a 200 or a single-range 206, then this # performs no IO, and either just returns source or # raises StopIteration. with ChunkReadTimeout(node_timeout): # if StopIteration is raised, it escapes and is # handled elsewhere start_byte, end_byte, length, headers, part = next( parts_iter[0]) return (start_byte, end_byte, length, headers, part) except ChunkReadTimeout: new_source, new_node = self._get_source_and_node() if new_source: self.app.exception_occurred( node[0], _('Object'), _('Trying to read during GET (retrying)')) # Close-out the connection as best as possible. if getattr(source[0], 'swift_conn', None): close_swift_conn(source[0]) source[0] = new_source node[0] = new_node # This is safe; it sets up a generator but does # not call next() on it, so no IO is performed. parts_iter[0] = http_response_to_document_iters( new_source, read_chunk_size=self.app.object_chunk_size) else: raise StopIteration() def iter_bytes_from_response_part(part_file): nchunks = 0 buf = '' bytes_used_from_backend = 0 while True: try: with ChunkReadTimeout(node_timeout): chunk = part_file.read(self.app.object_chunk_size) nchunks += 1 buf += chunk except ChunkReadTimeout: exc_type, exc_value, exc_traceback = exc_info() if self.newest or self.server_type != 'Object': six.reraise(exc_type, exc_value, exc_traceback) try: self.fast_forward(bytes_used_from_backend) except (HTTPException, ValueError): six.reraise(exc_type, exc_value, exc_traceback) except RangeAlreadyComplete: break buf = '' new_source, new_node = self._get_source_and_node() if new_source: self.app.exception_occurred( node[0], _('Object'), _('Trying to read during GET (retrying)')) # Close-out the connection as best as possible. if getattr(source[0], 'swift_conn', None): close_swift_conn(source[0]) source[0] = new_source node[0] = new_node # This is safe; it just sets up a generator but # does not call next() on it, so no IO is # performed. parts_iter[0] = http_response_to_document_iters( new_source, read_chunk_size=self.app.object_chunk_size) try: _junk, _junk, _junk, _junk, part_file = \ get_next_doc_part() except StopIteration: # Tried to find a new node from which to # finish the GET, but failed. There's # nothing more to do here. return else: six.reraise(exc_type, exc_value, exc_traceback) else: if buf and self.skip_bytes: if self.skip_bytes < len(buf): buf = buf[self.skip_bytes:] bytes_used_from_backend += self.skip_bytes self.skip_bytes = 0 else: self.skip_bytes -= len(buf) bytes_used_from_backend += len(buf) buf = '' if not chunk: if buf: with ChunkWriteTimeout( self.app.client_timeout): bytes_used_from_backend += len(buf) yield buf buf = '' break if client_chunk_size is not None: while len(buf) >= client_chunk_size: client_chunk = buf[:client_chunk_size] buf = buf[client_chunk_size:] with ChunkWriteTimeout( self.app.client_timeout): yield client_chunk bytes_used_from_backend += len(client_chunk) else: with ChunkWriteTimeout(self.app.client_timeout): yield buf bytes_used_from_backend += len(buf) buf = '' # This is for fairness; if the network is outpacing # the CPU, we'll always be able to read and write # data without encountering an EWOULDBLOCK, and so # eventlet will not switch greenthreads on its own. # We do it manually so that clients don't starve. # # The number 5 here was chosen by making stuff up. # It's not every single chunk, but it's not too big # either, so it seemed like it would probably be an # okay choice. # # Note that we may trampoline to other greenthreads # more often than once every 5 chunks, depending on # how blocking our network IO is; the explicit sleep # here simply provides a lower bound on the rate of # trampolining. if nchunks % 5 == 0: sleep() part_iter = None try: while True: start_byte, end_byte, length, headers, part = \ get_next_doc_part() self.learn_size_from_content_range( start_byte, end_byte, length) part_iter = iter_bytes_from_response_part(part) yield {'start_byte': start_byte, 'end_byte': end_byte, 'entity_length': length, 'headers': headers, 'part_iter': part_iter} self.pop_range() except GeneratorExit: if part_iter: part_iter.close() raise except StopIteration: req.environ['swift.non_client_disconnect'] = True except ChunkReadTimeout: self.app.exception_occurred(node[0], _('Object'), _('Trying to read during GET')) raise except ChunkWriteTimeout: self.app.logger.warn( _('Client did not read from proxy within %ss') % self.app.client_timeout) self.app.logger.increment('client_timeouts') except GeneratorExit: if not req.environ.get('swift.non_client_disconnect'): self.app.logger.warn(_('Client disconnected on read')) except Exception: self.app.logger.exception(_('Trying to send to client')) raise finally: # Close-out the connection as best as possible. if getattr(source[0], 'swift_conn', None): close_swift_conn(source[0]) @property def last_status(self): if self.statuses: return self.statuses[-1] else: return None @property def last_headers(self): if self.source_headers: return self.source_headers[-1] else: return None def _get_source_and_node(self): self.statuses = [] self.reasons = [] self.bodies = [] self.source_headers = [] sources = [] node_timeout = self.app.node_timeout if self.server_type == 'Object' and not self.newest: node_timeout = self.app.recoverable_node_timeout for node in self.node_iter: if node in self.used_nodes: continue start_node_timing = time.time() try: with ConnectionTimeout(self.app.conn_timeout): conn = http_connect( node['ip'], node['port'], node['device'], self.partition, self.req_method, self.path, headers=self.backend_headers, query_string=self.req_query_string) self.app.set_node_timing(node, time.time() - start_node_timing) with Timeout(node_timeout): possible_source = conn.getresponse() # See NOTE: swift_conn at top of file about this. possible_source.swift_conn = conn except (Exception, Timeout): self.app.exception_occurred( node, self.server_type, _('Trying to %(method)s %(path)s') % {'method': self.req_method, 'path': self.req_path}) continue if self.is_good_source(possible_source): # 404 if we know we don't have a synced copy if not float(possible_source.getheader('X-PUT-Timestamp', 1)): self.statuses.append(HTTP_NOT_FOUND) self.reasons.append('') self.bodies.append('') self.source_headers.append([]) close_swift_conn(possible_source) else: if self.used_source_etag: src_headers = dict( (k.lower(), v) for k, v in possible_source.getheaders()) if self.used_source_etag != src_headers.get( 'x-object-sysmeta-ec-etag', src_headers.get('etag', '')).strip('"'): self.statuses.append(HTTP_NOT_FOUND) self.reasons.append('') self.bodies.append('') self.source_headers.append([]) continue self.statuses.append(possible_source.status) self.reasons.append(possible_source.reason) self.bodies.append(None) self.source_headers.append(possible_source.getheaders()) sources.append((possible_source, node)) if not self.newest: # one good source is enough break else: self.statuses.append(possible_source.status) self.reasons.append(possible_source.reason) self.bodies.append(possible_source.read()) self.source_headers.append(possible_source.getheaders()) if possible_source.status == HTTP_INSUFFICIENT_STORAGE: self.app.error_limit(node, _('ERROR Insufficient Storage')) elif is_server_error(possible_source.status): self.app.error_occurred( node, _('ERROR %(status)d %(body)s ' 'From %(type)s Server') % {'status': possible_source.status, 'body': self.bodies[-1][:1024], 'type': self.server_type}) if sources: sources.sort(key=lambda s: source_key(s[0])) source, node = sources.pop() for src, _junk in sources: close_swift_conn(src) self.used_nodes.append(node) src_headers = dict( (k.lower(), v) for k, v in possible_source.getheaders()) # Save off the source etag so that, if we lose the connection # and have to resume from a different node, we can be sure that # we have the same object (replication) or a fragment archive # from the same object (EC). Otherwise, if the cluster has two # versions of the same object, we might end up switching between # old and new mid-stream and giving garbage to the client. self.used_source_etag = src_headers.get( 'x-object-sysmeta-ec-etag', src_headers.get('etag', '')).strip('"') return source, node return None, None class GetOrHeadHandler(ResumingGetter): def _make_app_iter(self, req, node, source): """ Returns an iterator over the contents of the source (via its read func). There is also quite a bit of cleanup to ensure garbage collection works and the underlying socket of the source is closed. :param req: incoming request object :param source: The httplib.Response object this iterator should read from. :param node: The node the source is reading from, for logging purposes. """ ct = source.getheader('Content-Type') if ct: content_type, content_type_attrs = parse_content_type(ct) is_multipart = content_type == 'multipart/byteranges' else: is_multipart = False boundary = "dontcare" if is_multipart: # we need some MIME boundary; fortunately, the object server has # furnished one for us, so we'll just re-use it boundary = dict(content_type_attrs)["boundary"] parts_iter = self._get_response_parts_iter(req, node, source) def add_content_type(response_part): response_part["content_type"] = \ HeaderKeyDict(response_part["headers"]).get("Content-Type") return response_part return document_iters_to_http_response_body( (add_content_type(pi) for pi in parts_iter), boundary, is_multipart, self.app.logger) def get_working_response(self, req): source, node = self._get_source_and_node() res = None if source: res = Response(request=req) res.status = source.status update_headers(res, source.getheaders()) if req.method == 'GET' and \ source.status in (HTTP_OK, HTTP_PARTIAL_CONTENT): res.app_iter = self._make_app_iter(req, node, source) # See NOTE: swift_conn at top of file about this. res.swift_conn = source.swift_conn if not res.environ: res.environ = {} res.environ['swift_x_timestamp'] = \ source.getheader('x-timestamp') res.accept_ranges = 'bytes' res.content_length = source.getheader('Content-Length') if source.getheader('Content-Type'): res.charset = None res.content_type = source.getheader('Content-Type') return res class NodeIter(object): """ Yields nodes for a ring partition, skipping over error limited nodes and stopping at the configurable number of nodes. If a node yielded subsequently gets error limited, an extra node will be yielded to take its place. Note that if you're going to iterate over this concurrently from multiple greenthreads, you'll want to use a swift.common.utils.GreenthreadSafeIterator to serialize access. Otherwise, you may get ValueErrors from concurrent access. (You also may not, depending on how logging is configured, the vagaries of socket IO and eventlet, and the phase of the moon.) :param app: a proxy app :param ring: ring to get yield nodes from :param partition: ring partition to yield nodes for :param node_iter: optional iterable of nodes to try. Useful if you want to filter or reorder the nodes. """ def __init__(self, app, ring, partition, node_iter=None): self.app = app self.ring = ring self.partition = partition part_nodes = ring.get_part_nodes(partition) if node_iter is None: node_iter = itertools.chain( part_nodes, ring.get_more_nodes(partition)) num_primary_nodes = len(part_nodes) self.nodes_left = self.app.request_node_count(num_primary_nodes) self.expected_handoffs = self.nodes_left - num_primary_nodes # Use of list() here forcibly yanks the first N nodes (the primary # nodes) from node_iter, so the rest of its values are handoffs. self.primary_nodes = self.app.sort_nodes( list(itertools.islice(node_iter, num_primary_nodes))) self.handoff_iter = node_iter def __iter__(self): self._node_iter = self._node_gen() return self def log_handoffs(self, handoffs): """ Log handoff requests if handoff logging is enabled and the handoff was not expected. We only log handoffs when we've pushed the handoff count further than we would normally have expected under normal circumstances, that is (request_node_count - num_primaries), when handoffs goes higher than that it means one of the primaries must have been skipped because of error limiting before we consumed all of our nodes_left. """ if not self.app.log_handoffs: return extra_handoffs = handoffs - self.expected_handoffs if extra_handoffs > 0: self.app.logger.increment('handoff_count') self.app.logger.warning( 'Handoff requested (%d)' % handoffs) if (extra_handoffs == len(self.primary_nodes)): # all the primaries were skipped, and handoffs didn't help self.app.logger.increment('handoff_all_count') def _node_gen(self): for node in self.primary_nodes: if not self.app.error_limited(node): yield node if not self.app.error_limited(node): self.nodes_left -= 1 if self.nodes_left <= 0: return handoffs = 0 for node in self.handoff_iter: if not self.app.error_limited(node): handoffs += 1 self.log_handoffs(handoffs) yield node if not self.app.error_limited(node): self.nodes_left -= 1 if self.nodes_left <= 0: return def next(self): return next(self._node_iter) def __next__(self): return self.next() class Controller(object): """Base WSGI controller class for the proxy""" server_type = 'Base' # Ensure these are all lowercase pass_through_headers = [] def __init__(self, app): """ Creates a controller attached to an application instance :param app: the application instance """ self.account_name = None self.app = app self.trans_id = '-' self._allowed_methods = None @property def allowed_methods(self): if self._allowed_methods is None: self._allowed_methods = set() all_methods = inspect.getmembers(self, predicate=inspect.ismethod) for name, m in all_methods: if getattr(m, 'publicly_accessible', False): self._allowed_methods.add(name) return self._allowed_methods def _x_remove_headers(self): """ Returns a list of headers that must not be sent to the backend :returns: a list of header """ return [] def transfer_headers(self, src_headers, dst_headers): """ Transfer legal headers from an original client request to dictionary that will be used as headers by the backend request :param src_headers: A dictionary of the original client request headers :param dst_headers: A dictionary of the backend request headers """ st = self.server_type.lower() x_remove = 'x-remove-%s-meta-' % st dst_headers.update((k.lower().replace('-remove', '', 1), '') for k in src_headers if k.lower().startswith(x_remove) or k.lower() in self._x_remove_headers()) dst_headers.update((k.lower(), v) for k, v in src_headers.items() if k.lower() in self.pass_through_headers or is_sys_or_user_meta(st, k)) def generate_request_headers(self, orig_req=None, additional=None, transfer=False): """ Create a list of headers to be used in backend requests :param orig_req: the original request sent by the client to the proxy :param additional: additional headers to send to the backend :param transfer: If True, transfer headers from original client request :returns: a dictionary of headers """ # Use the additional headers first so they don't overwrite the headers # we require. headers = HeaderKeyDict(additional) if additional else HeaderKeyDict() if transfer: self.transfer_headers(orig_req.headers, headers) headers.setdefault('x-timestamp', Timestamp(time.time()).internal) if orig_req: referer = orig_req.as_referer() else: referer = '' headers['x-trans-id'] = self.trans_id headers['connection'] = 'close' headers['user-agent'] = 'proxy-server %s' % os.getpid() headers['referer'] = referer return headers def account_info(self, account, req=None): """ Get account information, and also verify that the account exists. :param account: name of the account to get the info for :param req: caller's HTTP request context object (optional) :returns: tuple of (account partition, account nodes, container_count) or (None, None, None) if it does not exist """ partition, nodes = self.app.account_ring.get_nodes(account) if req: env = getattr(req, 'environ', {}) else: env = {} info = get_info(self.app, env, account) if not info: return None, None, None if info.get('container_count') is None: container_count = 0 else: container_count = int(info['container_count']) return partition, nodes, container_count def container_info(self, account, container, req=None): """ Get container information and thusly verify container existence. This will also verify account existence. :param account: account name for the container :param container: container name to look up :param req: caller's HTTP request context object (optional) :returns: dict containing at least container partition ('partition'), container nodes ('containers'), container read acl ('read_acl'), container write acl ('write_acl'), and container sync key ('sync_key'). Values are set to None if the container does not exist. """ part, nodes = self.app.container_ring.get_nodes(account, container) if req: env = getattr(req, 'environ', {}) else: env = {} info = get_info(self.app, env, account, container) if not info: info = headers_to_container_info({}, 0) info['partition'] = None info['nodes'] = None else: info['partition'] = part info['nodes'] = nodes if info.get('storage_policy') is None: info['storage_policy'] = 0 return info def _make_request(self, nodes, part, method, path, headers, query, logger_thread_locals): """ Iterates over the given node iterator, sending an HTTP request to one node at a time. The first non-informational, non-server-error response is returned. If no non-informational, non-server-error response is received from any of the nodes, returns None. :param nodes: an iterator of the backend server and handoff servers :param part: the partition number :param method: the method to send to the backend :param path: the path to send to the backend (full path ends up being /<$device>/<$part>/<$path>) :param headers: dictionary of headers :param query: query string to send to the backend. :param logger_thread_locals: The thread local values to be set on the self.app.logger to retain transaction logging information. :returns: a swob.Response object, or None if no responses were received """ self.app.logger.thread_locals = logger_thread_locals for node in nodes: try: start_node_timing = time.time() with ConnectionTimeout(self.app.conn_timeout): conn = http_connect(node['ip'], node['port'], node['device'], part, method, path, headers=headers, query_string=query) conn.node = node self.app.set_node_timing(node, time.time() - start_node_timing) with Timeout(self.app.node_timeout): resp = conn.getresponse() if not is_informational(resp.status) and \ not is_server_error(resp.status): return resp.status, resp.reason, resp.getheaders(), \ resp.read() elif resp.status == HTTP_INSUFFICIENT_STORAGE: self.app.error_limit(node, _('ERROR Insufficient Storage')) elif is_server_error(resp.status): self.app.error_occurred( node, _('ERROR %(status)d ' 'Trying to %(method)s %(path)s' 'From Container Server') % { 'status': resp.status, 'method': method, 'path': path}) except (Exception, Timeout): self.app.exception_occurred( node, self.server_type, _('Trying to %(method)s %(path)s') % {'method': method, 'path': path}) def make_requests(self, req, ring, part, method, path, headers, query_string='', overrides=None): """ Sends an HTTP request to multiple nodes and aggregates the results. It attempts the primary nodes concurrently, then iterates over the handoff nodes as needed. :param req: a request sent by the client :param ring: the ring used for finding backend servers :param part: the partition number :param method: the method to send to the backend :param path: the path to send to the backend (full path ends up being /<$device>/<$part>/<$path>) :param headers: a list of dicts, where each dict represents one backend request that should be made. :param query_string: optional query string to send to the backend :param overrides: optional return status override map used to override the returned status of a request. :returns: a swob.Response object """ start_nodes = ring.get_part_nodes(part) nodes = GreenthreadSafeIterator(self.app.iter_nodes(ring, part)) pile = GreenAsyncPile(len(start_nodes)) for head in headers: pile.spawn(self._make_request, nodes, part, method, path, head, query_string, self.app.logger.thread_locals) response = [] statuses = [] for resp in pile: if not resp: continue response.append(resp) statuses.append(resp[0]) if self.have_quorum(statuses, len(start_nodes)): break # give any pending requests *some* chance to finish finished_quickly = pile.waitall(self.app.post_quorum_timeout) for resp in finished_quickly: if not resp: continue response.append(resp) statuses.append(resp[0]) while len(response) < len(start_nodes): response.append((HTTP_SERVICE_UNAVAILABLE, '', '', '')) statuses, reasons, resp_headers, bodies = zip(*response) return self.best_response(req, statuses, reasons, bodies, '%s %s' % (self.server_type, req.method), overrides=overrides, headers=resp_headers) def _quorum_size(self, n): """ Number of successful backend responses needed for the proxy to consider the client request successful. """ return quorum_size(n) def have_quorum(self, statuses, node_count, quorum=None): """ Given a list of statuses from several requests, determine if a quorum response can already be decided. :param statuses: list of statuses returned :param node_count: number of nodes being queried (basically ring count) :param quorum: number of statuses required for quorum :returns: True or False, depending on if quorum is established """ if quorum is None: quorum = self._quorum_size(node_count) if len(statuses) >= quorum: for hundred in (HTTP_CONTINUE, HTTP_OK, HTTP_MULTIPLE_CHOICES, HTTP_BAD_REQUEST): if sum(1 for s in statuses if hundred <= s < hundred + 100) >= quorum: return True return False def best_response(self, req, statuses, reasons, bodies, server_type, etag=None, headers=None, overrides=None, quorum_size=None): """ Given a list of responses from several servers, choose the best to return to the API. :param req: swob.Request object :param statuses: list of statuses returned :param reasons: list of reasons for each status :param bodies: bodies of each response :param server_type: type of server the responses came from :param etag: etag :param headers: headers of each response :param overrides: overrides to apply when lacking quorum :param quorum_size: quorum size to use :returns: swob.Response object with the correct status, body, etc. set """ if quorum_size is None: quorum_size = self._quorum_size(len(statuses)) resp = self._compute_quorum_response( req, statuses, reasons, bodies, etag, headers, quorum_size=quorum_size) if overrides and not resp: faked_up_status_indices = set() transformed = [] for (i, (status, reason, hdrs, body)) in enumerate(zip( statuses, reasons, headers, bodies)): if status in overrides: faked_up_status_indices.add(i) transformed.append((overrides[status], '', '', '')) else: transformed.append((status, reason, hdrs, body)) statuses, reasons, headers, bodies = zip(*transformed) resp = self._compute_quorum_response( req, statuses, reasons, bodies, etag, headers, indices_to_avoid=faked_up_status_indices, quorum_size=quorum_size) if not resp: resp = HTTPServiceUnavailable(request=req) self.app.logger.error(_('%(type)s returning 503 for %(statuses)s'), {'type': server_type, 'statuses': statuses}) return resp def _compute_quorum_response(self, req, statuses, reasons, bodies, etag, headers, quorum_size, indices_to_avoid=()): if not statuses: return None for hundred in (HTTP_OK, HTTP_MULTIPLE_CHOICES, HTTP_BAD_REQUEST): hstatuses = \ [(i, s) for i, s in enumerate(statuses) if hundred <= s < hundred + 100] if len(hstatuses) >= quorum_size: try: status_index, status = max( ((i, stat) for i, stat in hstatuses if i not in indices_to_avoid), key=operator.itemgetter(1)) except ValueError: # All statuses were indices to avoid continue resp = status_map[status](request=req) resp.status = '%s %s' % (status, reasons[status_index]) resp.body = bodies[status_index] if headers: update_headers(resp, headers[status_index]) if etag: resp.headers['etag'] = etag.strip('"') return resp return None @public def GET(self, req): """ Handler for HTTP GET requests. :param req: The client request :returns: the response to the client """ return self.GETorHEAD(req) @public def HEAD(self, req): """ Handler for HTTP HEAD requests. :param req: The client request :returns: the response to the client """ return self.GETorHEAD(req) def autocreate_account(self, req, account): """ Autocreate an account :param env: the environment of the request leading to this autocreate :param account: the unquoted account name """ partition, nodes = self.app.account_ring.get_nodes(account) path = '/%s' % account headers = {'X-Timestamp': Timestamp(time.time()).internal, 'X-Trans-Id': self.trans_id, 'Connection': 'close'} # transfer any x-account-sysmeta headers from original request # to the autocreate PUT headers.update((k, v) for k, v in req.headers.items() if is_sys_meta('account', k)) resp = self.make_requests(Request.blank('/v1' + path), self.app.account_ring, partition, 'PUT', path, [headers] * len(nodes)) if is_success(resp.status_int): self.app.logger.info('autocreate account %r' % path) clear_info_cache(self.app, req.environ, account) else: self.app.logger.warning('Could not autocreate account %r' % path) def GETorHEAD_base(self, req, server_type, node_iter, partition, path, client_chunk_size=None): """ Base handler for HTTP GET or HEAD requests. :param req: swob.Request object :param server_type: server type used in logging :param node_iter: an iterator to obtain nodes from :param partition: partition :param path: path for the request :param client_chunk_size: chunk size for response body iterator :returns: swob.Response object """ backend_headers = self.generate_request_headers( req, additional=req.headers) handler = GetOrHeadHandler(self.app, req, self.server_type, node_iter, partition, path, backend_headers, client_chunk_size=client_chunk_size) res = handler.get_working_response(req) if not res: res = self.best_response( req, handler.statuses, handler.reasons, handler.bodies, '%s %s' % (server_type, req.method), headers=handler.source_headers) try: (vrs, account, container) = req.split_path(2, 3) _set_info_cache(self.app, req.environ, account, container, res) except ValueError: pass try: (vrs, account, container, obj) = req.split_path(4, 4, True) _set_object_info_cache(self.app, req.environ, account, container, obj, res) except ValueError: pass # if a backend policy index is present in resp headers, translate it # here with the friendly policy name if 'X-Backend-Storage-Policy-Index' in res.headers and \ is_success(res.status_int): policy = \ POLICIES.get_by_index( res.headers['X-Backend-Storage-Policy-Index']) if policy: res.headers['X-Storage-Policy'] = policy.name else: self.app.logger.error( 'Could not translate %s (%r) from %r to policy', 'X-Backend-Storage-Policy-Index', res.headers['X-Backend-Storage-Policy-Index'], path) return res def is_origin_allowed(self, cors_info, origin): """ Is the given Origin allowed to make requests to this resource :param cors_info: the resource's CORS related metadata headers :param origin: the origin making the request :return: True or False """ allowed_origins = set() if cors_info.get('allow_origin'): allowed_origins.update( [a.strip() for a in cors_info['allow_origin'].split(' ') if a.strip()]) if self.app.cors_allow_origin: allowed_origins.update(self.app.cors_allow_origin) return origin in allowed_origins or '*' in allowed_origins @public def OPTIONS(self, req): """ Base handler for OPTIONS requests :param req: swob.Request object :returns: swob.Response object """ # Prepare the default response headers = {'Allow': ', '.join(self.allowed_methods)} resp = Response(status=200, request=req, headers=headers) # If this isn't a CORS pre-flight request then return now req_origin_value = req.headers.get('Origin', None) if not req_origin_value: return resp # This is a CORS preflight request so check it's allowed try: container_info = \ self.container_info(self.account_name, self.container_name, req) except AttributeError: # This should only happen for requests to the Account. A future # change could allow CORS requests to the Account level as well. return resp cors = container_info.get('cors', {}) # If the CORS origin isn't allowed return a 401 if not self.is_origin_allowed(cors, req_origin_value) or ( req.headers.get('Access-Control-Request-Method') not in self.allowed_methods): resp.status = HTTP_UNAUTHORIZED return resp # Allow all headers requested in the request. The CORS # specification does leave the door open for this, as mentioned in # http://www.w3.org/TR/cors/#resource-preflight-requests # Note: Since the list of headers can be unbounded # simply returning headers can be enough. allow_headers = set() if req.headers.get('Access-Control-Request-Headers'): allow_headers.update( list_from_csv(req.headers['Access-Control-Request-Headers'])) # Populate the response with the CORS preflight headers if cors.get('allow_origin') and \ cors.get('allow_origin').strip() == '*': headers['access-control-allow-origin'] = '*' else: headers['access-control-allow-origin'] = req_origin_value if cors.get('max_age') is not None: headers['access-control-max-age'] = cors.get('max_age') headers['access-control-allow-methods'] = \ ', '.join(self.allowed_methods) if allow_headers: headers['access-control-allow-headers'] = ', '.join(allow_headers) resp.headers = headers return resp