16 import arvados.config as config
18 import arvados.retry as retry
21 _logger = logging.getLogger('arvados.keep')
22 global_client_object = None
25 class KeepLocator(object):
26 EPOCH_DATETIME = datetime.datetime.utcfromtimestamp(0)
27 HINT_RE = re.compile(r'^[A-Z][A-Za-z0-9@_-]+$')
29 def __init__(self, locator_str):
32 self._perm_expiry = None
33 pieces = iter(locator_str.split('+'))
34 self.md5sum = next(pieces)
36 self.size = int(next(pieces))
40 if self.HINT_RE.match(hint) is None:
41 raise ValueError("invalid hint format: {}".format(hint))
42 elif hint.startswith('A'):
43 self.parse_permission_hint(hint)
45 self.hints.append(hint)
49 str(s) for s in [self.md5sum, self.size,
50 self.permission_hint()] + self.hints
54 if self.size is not None:
55 return "%s+%i" % (self.md5sum, self.size)
59 def _make_hex_prop(name, length):
60 # Build and return a new property with the given name that
61 # must be a hex string of the given length.
62 data_name = '_{}'.format(name)
64 return getattr(self, data_name)
65 def setter(self, hex_str):
66 if not arvados.util.is_hex(hex_str, length):
67 raise ValueError("{} is not a {}-digit hex string: {}".
68 format(name, length, hex_str))
69 setattr(self, data_name, hex_str)
70 return property(getter, setter)
72 md5sum = _make_hex_prop('md5sum', 32)
73 perm_sig = _make_hex_prop('perm_sig', 40)
76 def perm_expiry(self):
77 return self._perm_expiry
80 def perm_expiry(self, value):
81 if not arvados.util.is_hex(value, 1, 8):
83 "permission timestamp must be a hex Unix timestamp: {}".
85 self._perm_expiry = datetime.datetime.utcfromtimestamp(int(value, 16))
87 def permission_hint(self):
88 data = [self.perm_sig, self.perm_expiry]
91 data[1] = int((data[1] - self.EPOCH_DATETIME).total_seconds())
92 return "A{}@{:08x}".format(*data)
94 def parse_permission_hint(self, s):
96 self.perm_sig, self.perm_expiry = s[1:].split('@', 1)
98 raise ValueError("bad permission hint {}".format(s))
100 def permission_expired(self, as_of_dt=None):
101 if self.perm_expiry is None:
103 elif as_of_dt is None:
104 as_of_dt = datetime.datetime.now()
105 return self.perm_expiry <= as_of_dt
109 """Simple interface to a global KeepClient object.
111 THIS CLASS IS DEPRECATED. Please instantiate your own KeepClient with your
112 own API client. The global KeepClient will build an API client from the
113 current Arvados configuration, which may not match the one you built.
118 def global_client_object(cls):
119 global global_client_object
120 # Previously, KeepClient would change its behavior at runtime based
121 # on these configuration settings. We simulate that behavior here
122 # by checking the values and returning a new KeepClient if any of
124 key = (config.get('ARVADOS_API_HOST'),
125 config.get('ARVADOS_API_TOKEN'),
126 config.flag_is_true('ARVADOS_API_HOST_INSECURE'),
127 config.get('ARVADOS_KEEP_PROXY'),
128 config.get('ARVADOS_EXTERNAL_CLIENT') == 'true',
129 os.environ.get('KEEP_LOCAL_STORE'))
130 if (global_client_object is None) or (cls._last_key != key):
131 global_client_object = KeepClient()
133 return global_client_object
136 def get(locator, **kwargs):
137 return Keep.global_client_object().get(locator, **kwargs)
140 def put(data, **kwargs):
141 return Keep.global_client_object().put(data, **kwargs)
143 class KeepBlockCache(object):
144 # Default RAM cache is 256MiB
145 def __init__(self, cache_max=(256 * 1024 * 1024)):
146 self.cache_max = cache_max
148 self._cache_lock = threading.Lock()
150 class CacheSlot(object):
151 __slots__ = ("locator", "ready", "content")
153 def __init__(self, locator):
154 self.locator = locator
155 self.ready = threading.Event()
162 def set(self, value):
167 if self.content is None:
170 return len(self.content)
173 '''Cap the cache size to self.cache_max'''
174 with self._cache_lock:
175 # Select all slots except those where ready.is_set() and content is
176 # None (that means there was an error reading the block).
177 self._cache = [c for c in self._cache if not (c.ready.is_set() and c.content is None)]
178 sm = sum([slot.size() for slot in self._cache])
179 while len(self._cache) > 0 and sm > self.cache_max:
180 for i in xrange(len(self._cache)-1, -1, -1):
181 if self._cache[i].ready.is_set():
184 sm = sum([slot.size() for slot in self._cache])
186 def _get(self, locator):
187 # Test if the locator is already in the cache
188 for i in xrange(0, len(self._cache)):
189 if self._cache[i].locator == locator:
192 # move it to the front
194 self._cache.insert(0, n)
198 def get(self, locator):
199 with self._cache_lock:
200 return self._get(locator)
202 def reserve_cache(self, locator):
203 '''Reserve a cache slot for the specified locator,
204 or return the existing slot.'''
205 with self._cache_lock:
206 n = self._get(locator)
210 # Add a new cache slot for the locator
211 n = KeepBlockCache.CacheSlot(locator)
212 self._cache.insert(0, n)
215 class Counter(object):
216 def __init__(self, v=0):
217 self._lk = threading.Lock()
229 class KeepClient(object):
231 # Default Keep server connection timeout: 2 seconds
232 # Default Keep server read timeout: 256 seconds
233 # Default Keep server bandwidth minimum: 32768 bytes per second
234 # Default Keep proxy connection timeout: 20 seconds
235 # Default Keep proxy read timeout: 256 seconds
236 # Default Keep proxy bandwidth minimum: 32768 bytes per second
237 DEFAULT_TIMEOUT = (2, 256, 32768)
238 DEFAULT_PROXY_TIMEOUT = (20, 256, 32768)
240 class ThreadLimiter(object):
241 """Limit the number of threads writing to Keep at once.
243 This ensures that only a number of writer threads that could
244 potentially achieve the desired replication level run at once.
245 Once the desired replication level is achieved, queued threads
246 are instructed not to run.
248 Should be used in a "with" block.
250 def __init__(self, want_copies, max_service_replicas):
252 self._want_copies = want_copies
254 self._thread_failures = 0
255 self._response = None
256 self._start_lock = threading.Condition()
257 if (not max_service_replicas) or (max_service_replicas >= want_copies):
260 max_threads = math.ceil(float(want_copies) / max_service_replicas)
261 _logger.debug("Limiter max threads is %d", max_threads)
262 self._todo_lock = threading.Semaphore(max_threads)
263 self._done_lock = threading.Lock()
264 self._thread_failures_lock = threading.Lock()
265 self._local = threading.local()
268 self._start_lock.acquire()
269 if getattr(self._local, 'sequence', None) is not None:
270 # If the calling thread has used set_sequence(N), then
271 # we wait here until N other threads have started.
272 while self._started < self._local.sequence:
273 self._start_lock.wait()
274 self._todo_lock.acquire()
276 self._start_lock.notifyAll()
277 self._start_lock.release()
280 def __exit__(self, type, value, traceback):
281 with self._thread_failures_lock:
282 if self._thread_failures > 0:
283 self._thread_failures -= 1
284 self._todo_lock.release()
286 # If work is finished, release al pending threads
287 if not self.shall_i_proceed():
288 self._todo_lock.release()
290 def set_sequence(self, sequence):
291 self._local.sequence = sequence
293 def shall_i_proceed(self):
295 Return true if the current thread should write to Keep.
296 Return false otherwise.
298 with self._done_lock:
299 return (self._done < self._want_copies)
301 def save_response(self, response_body, replicas_stored):
303 Records a response body (a locator, possibly signed) returned by
304 the Keep server, and the number of replicas it stored.
306 if replicas_stored == 0:
307 # Failure notification, should start a new thread to try to reach full replication
308 with self._thread_failures_lock:
309 self._thread_failures += 1
311 with self._done_lock:
312 self._done += replicas_stored
313 self._response = response_body
316 """Return the body from the response to a PUT request."""
317 with self._done_lock:
318 return self._response
321 """Return the total number of replicas successfully stored."""
322 with self._done_lock:
325 class KeepService(object):
326 """Make requests to a single Keep service, and track results.
328 A KeepService is intended to last long enough to perform one
329 transaction (GET or PUT) against one Keep service. This can
330 involve calling either get() or put() multiple times in order
331 to retry after transient failures. However, calling both get()
332 and put() on a single instance -- or using the same instance
333 to access two different Keep services -- will not produce
340 arvados.errors.HttpError,
343 def __init__(self, root, user_agent_pool=Queue.LifoQueue(),
345 download_counter=None, **headers):
347 self._user_agent_pool = user_agent_pool
348 self._result = {'error': None}
351 self.get_headers = {'Accept': 'application/octet-stream'}
352 self.get_headers.update(headers)
353 self.put_headers = headers
354 self.upload_counter = upload_counter
355 self.download_counter = download_counter
358 """Is it worth attempting a request?"""
362 """Did the request succeed or encounter permanent failure?"""
363 return self._result['error'] == False or not self._usable
365 def last_result(self):
368 def _get_user_agent(self):
370 return self._user_agent_pool.get(False)
374 def _put_user_agent(self, ua):
377 self._user_agent_pool.put(ua, False)
382 def _socket_open(family, socktype, protocol, address=None):
383 """Because pycurl doesn't have CURLOPT_TCP_KEEPALIVE"""
384 s = socket.socket(family, socktype, protocol)
385 s.setsockopt(socket.SOL_SOCKET, socket.SO_KEEPALIVE, 1)
386 s.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPIDLE, 75)
387 s.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPINTVL, 75)
390 def get(self, locator, method="GET", timeout=None):
391 # locator is a KeepLocator object.
392 url = self.root + str(locator)
393 _logger.debug("Request: %s %s", method, url)
394 curl = self._get_user_agent()
397 with timer.Timer() as t:
399 response_body = cStringIO.StringIO()
400 curl.setopt(pycurl.NOSIGNAL, 1)
401 curl.setopt(pycurl.OPENSOCKETFUNCTION, self._socket_open)
402 curl.setopt(pycurl.URL, url.encode('utf-8'))
403 curl.setopt(pycurl.HTTPHEADER, [
404 '{}: {}'.format(k,v) for k,v in self.get_headers.iteritems()])
405 curl.setopt(pycurl.WRITEFUNCTION, response_body.write)
406 curl.setopt(pycurl.HEADERFUNCTION, self._headerfunction)
408 curl.setopt(pycurl.NOBODY, True)
409 self._setcurltimeouts(curl, timeout)
413 except Exception as e:
414 raise arvados.errors.HttpError(0, str(e))
416 'status_code': curl.getinfo(pycurl.RESPONSE_CODE),
417 'body': response_body.getvalue(),
418 'headers': self._headers,
422 ok = retry.check_http_response_success(self._result['status_code'])
424 self._result['error'] = arvados.errors.HttpError(
425 self._result['status_code'],
426 self._headers.get('x-status-line', 'Error'))
427 except self.HTTP_ERRORS as e:
431 self._usable = ok != False
432 if self._result.get('status_code', None):
433 # The client worked well enough to get an HTTP status
434 # code, so presumably any problems are just on the
435 # server side and it's OK to reuse the client.
436 self._put_user_agent(curl)
438 # Don't return this client to the pool, in case it's
442 _logger.debug("Request fail: GET %s => %s: %s",
443 url, type(self._result['error']), str(self._result['error']))
446 _logger.info("HEAD %s: %s bytes",
447 self._result['status_code'],
448 self._result.get('content-length'))
451 _logger.info("GET %s: %s bytes in %s msec (%.3f MiB/sec)",
452 self._result['status_code'],
453 len(self._result['body']),
455 (len(self._result['body'])/(1024.0*1024))/t.secs if t.secs > 0 else 0)
457 if self.download_counter:
458 self.download_counter.add(len(self._result['body']))
459 resp_md5 = hashlib.md5(self._result['body']).hexdigest()
460 if resp_md5 != locator.md5sum:
461 _logger.warning("Checksum fail: md5(%s) = %s",
463 self._result['error'] = arvados.errors.HttpError(
466 return self._result['body']
468 def put(self, hash_s, body, timeout=None):
469 url = self.root + hash_s
470 _logger.debug("Request: PUT %s", url)
471 curl = self._get_user_agent()
474 with timer.Timer() as t:
476 body_reader = cStringIO.StringIO(body)
477 response_body = cStringIO.StringIO()
478 curl.setopt(pycurl.NOSIGNAL, 1)
479 curl.setopt(pycurl.OPENSOCKETFUNCTION, self._socket_open)
480 curl.setopt(pycurl.URL, url.encode('utf-8'))
481 # Using UPLOAD tells cURL to wait for a "go ahead" from the
482 # Keep server (in the form of a HTTP/1.1 "100 Continue"
483 # response) instead of sending the request body immediately.
484 # This allows the server to reject the request if the request
485 # is invalid or the server is read-only, without waiting for
486 # the client to send the entire block.
487 curl.setopt(pycurl.UPLOAD, True)
488 curl.setopt(pycurl.INFILESIZE, len(body))
489 curl.setopt(pycurl.READFUNCTION, body_reader.read)
490 curl.setopt(pycurl.HTTPHEADER, [
491 '{}: {}'.format(k,v) for k,v in self.put_headers.iteritems()])
492 curl.setopt(pycurl.WRITEFUNCTION, response_body.write)
493 curl.setopt(pycurl.HEADERFUNCTION, self._headerfunction)
494 self._setcurltimeouts(curl, timeout)
497 except Exception as e:
498 raise arvados.errors.HttpError(0, str(e))
500 'status_code': curl.getinfo(pycurl.RESPONSE_CODE),
501 'body': response_body.getvalue(),
502 'headers': self._headers,
505 ok = retry.check_http_response_success(self._result['status_code'])
507 self._result['error'] = arvados.errors.HttpError(
508 self._result['status_code'],
509 self._headers.get('x-status-line', 'Error'))
510 except self.HTTP_ERRORS as e:
514 self._usable = ok != False # still usable if ok is True or None
515 if self._result.get('status_code', None):
516 # Client is functional. See comment in get().
517 self._put_user_agent(curl)
521 _logger.debug("Request fail: PUT %s => %s: %s",
522 url, type(self._result['error']), str(self._result['error']))
524 _logger.info("PUT %s: %s bytes in %s msec (%.3f MiB/sec)",
525 self._result['status_code'],
528 (len(body)/(1024.0*1024))/t.secs if t.secs > 0 else 0)
529 if self.upload_counter:
530 self.upload_counter.add(len(body))
533 def _setcurltimeouts(self, curl, timeouts):
536 elif isinstance(timeouts, tuple):
537 if len(timeouts) == 2:
538 conn_t, xfer_t = timeouts
539 bandwidth_bps = KeepClient.DEFAULT_TIMEOUT[2]
541 conn_t, xfer_t, bandwidth_bps = timeouts
543 conn_t, xfer_t = (timeouts, timeouts)
544 bandwidth_bps = KeepClient.DEFAULT_TIMEOUT[2]
545 curl.setopt(pycurl.CONNECTTIMEOUT_MS, int(conn_t*1000))
546 curl.setopt(pycurl.LOW_SPEED_TIME, int(math.ceil(xfer_t)))
547 curl.setopt(pycurl.LOW_SPEED_LIMIT, int(math.ceil(bandwidth_bps)))
549 def _headerfunction(self, header_line):
550 header_line = header_line.decode('iso-8859-1')
551 if ':' in header_line:
552 name, value = header_line.split(':', 1)
553 name = name.strip().lower()
554 value = value.strip()
556 name = self._lastheadername
557 value = self._headers[name] + ' ' + header_line.strip()
558 elif header_line.startswith('HTTP/'):
559 name = 'x-status-line'
562 _logger.error("Unexpected header line: %s", header_line)
564 self._lastheadername = name
565 self._headers[name] = value
566 # Returning None implies all bytes were written
569 class KeepWriterThread(threading.Thread):
571 Write a blob of data to the given Keep server. On success, call
572 save_response() of the given ThreadLimiter to save the returned
575 def __init__(self, keep_service, **kwargs):
576 super(KeepClient.KeepWriterThread, self).__init__()
577 self.service = keep_service
579 self._success = False
585 limiter = self.args['thread_limiter']
586 sequence = self.args['thread_sequence']
587 if sequence is not None:
588 limiter.set_sequence(sequence)
590 if not limiter.shall_i_proceed():
591 # My turn arrived, but the job has been done without
594 self.run_with_limiter(limiter)
596 def run_with_limiter(self, limiter):
597 if self.service.finished():
599 _logger.debug("KeepWriterThread %s proceeding %s+%i %s",
600 str(threading.current_thread()),
601 self.args['data_hash'],
602 len(self.args['data']),
603 self.args['service_root'])
604 self._success = bool(self.service.put(
605 self.args['data_hash'],
607 timeout=self.args.get('timeout', None)))
608 result = self.service.last_result()
610 _logger.debug("KeepWriterThread %s succeeded %s+%i %s",
611 str(threading.current_thread()),
612 self.args['data_hash'],
613 len(self.args['data']),
614 self.args['service_root'])
615 # Tick the 'done' counter for the number of replica
616 # reported stored by the server, for the case that
617 # we're talking to a proxy or other backend that
618 # stores to multiple copies for us.
620 replicas_stored = int(result['headers']['x-keep-replicas-stored'])
621 except (KeyError, ValueError):
623 limiter.save_response(result['body'].strip(), replicas_stored)
624 elif result.get('status_code', None):
625 _logger.debug("Request fail: PUT %s => %s %s",
626 self.args['data_hash'],
627 result['status_code'],
629 if not self._success:
630 # Notify the failure so that the Thread limiter allows
632 limiter.save_response(None, 0)
635 def __init__(self, api_client=None, proxy=None,
636 timeout=DEFAULT_TIMEOUT, proxy_timeout=DEFAULT_PROXY_TIMEOUT,
637 api_token=None, local_store=None, block_cache=None,
638 num_retries=0, session=None):
639 """Initialize a new KeepClient.
643 The API client to use to find Keep services. If not
644 provided, KeepClient will build one from available Arvados
648 If specified, this KeepClient will send requests to this Keep
649 proxy. Otherwise, KeepClient will fall back to the setting of the
650 ARVADOS_KEEP_PROXY configuration setting. If you want to ensure
651 KeepClient does not use a proxy, pass in an empty string.
654 The initial timeout (in seconds) for HTTP requests to Keep
655 non-proxy servers. A tuple of three floats is interpreted as
656 (connection_timeout, read_timeout, minimum_bandwidth). A connection
657 will be aborted if the average traffic rate falls below
658 minimum_bandwidth bytes per second over an interval of read_timeout
659 seconds. Because timeouts are often a result of transient server
660 load, the actual connection timeout will be increased by a factor
661 of two on each retry.
662 Default: (2, 256, 32768).
665 The initial timeout (in seconds) for HTTP requests to
666 Keep proxies. A tuple of three floats is interpreted as
667 (connection_timeout, read_timeout, minimum_bandwidth). The behavior
668 described above for adjusting connection timeouts on retry also
670 Default: (20, 256, 32768).
673 If you're not using an API client, but only talking
674 directly to a Keep proxy, this parameter specifies an API token
675 to authenticate Keep requests. It is an error to specify both
676 api_client and api_token. If you specify neither, KeepClient
677 will use one available from the Arvados configuration.
680 If specified, this KeepClient will bypass Keep
681 services, and save data to the named directory. If unspecified,
682 KeepClient will fall back to the setting of the $KEEP_LOCAL_STORE
683 environment variable. If you want to ensure KeepClient does not
684 use local storage, pass in an empty string. This is primarily
685 intended to mock a server for testing.
688 The default number of times to retry failed requests.
689 This will be used as the default num_retries value when get() and
690 put() are called. Default 0.
692 self.lock = threading.Lock()
694 proxy = config.get('ARVADOS_KEEP_PROXY')
695 if api_token is None:
696 if api_client is None:
697 api_token = config.get('ARVADOS_API_TOKEN')
699 api_token = api_client.api_token
700 elif api_client is not None:
702 "can't build KeepClient with both API client and token")
703 if local_store is None:
704 local_store = os.environ.get('KEEP_LOCAL_STORE')
706 self.block_cache = block_cache if block_cache else KeepBlockCache()
707 self.timeout = timeout
708 self.proxy_timeout = proxy_timeout
709 self._user_agent_pool = Queue.LifoQueue()
710 self.upload_counter = Counter()
711 self.download_counter = Counter()
712 self.put_counter = Counter()
713 self.get_counter = Counter()
714 self.hits_counter = Counter()
715 self.misses_counter = Counter()
718 self.local_store = local_store
719 self.get = self.local_store_get
720 self.put = self.local_store_put
722 self.num_retries = num_retries
723 self.max_replicas_per_service = None
725 if not proxy.endswith('/'):
727 self.api_token = api_token
728 self._gateway_services = {}
729 self._keep_services = [{
731 'service_type': 'proxy',
732 '_service_root': proxy,
734 self._writable_services = self._keep_services
735 self.using_proxy = True
736 self._static_services_list = True
738 # It's important to avoid instantiating an API client
739 # unless we actually need one, for testing's sake.
740 if api_client is None:
741 api_client = arvados.api('v1')
742 self.api_client = api_client
743 self.api_token = api_client.api_token
744 self._gateway_services = {}
745 self._keep_services = None
746 self._writable_services = None
747 self.using_proxy = None
748 self._static_services_list = False
750 def current_timeout(self, attempt_number):
751 """Return the appropriate timeout to use for this client.
753 The proxy timeout setting if the backend service is currently a proxy,
754 the regular timeout setting otherwise. The `attempt_number` indicates
755 how many times the operation has been tried already (starting from 0
756 for the first try), and scales the connection timeout portion of the
757 return value accordingly.
760 # TODO(twp): the timeout should be a property of a
761 # KeepService, not a KeepClient. See #4488.
762 t = self.proxy_timeout if self.using_proxy else self.timeout
764 return (t[0] * (1 << attempt_number), t[1])
766 return (t[0] * (1 << attempt_number), t[1], t[2])
767 def _any_nondisk_services(self, service_list):
768 return any(ks.get('service_type', 'disk') != 'disk'
769 for ks in service_list)
771 def build_services_list(self, force_rebuild=False):
772 if (self._static_services_list or
773 (self._keep_services and not force_rebuild)):
777 keep_services = self.api_client.keep_services().accessible()
778 except Exception: # API server predates Keep services.
779 keep_services = self.api_client.keep_disks().list()
781 # Gateway services are only used when specified by UUID,
782 # so there's nothing to gain by filtering them by
784 self._gateway_services = {ks['uuid']: ks for ks in
785 keep_services.execute()['items']}
786 if not self._gateway_services:
787 raise arvados.errors.NoKeepServersError()
789 # Precompute the base URI for each service.
790 for r in self._gateway_services.itervalues():
791 host = r['service_host']
792 if not host.startswith('[') and host.find(':') >= 0:
793 # IPv6 URIs must be formatted like http://[::1]:80/...
794 host = '[' + host + ']'
795 r['_service_root'] = "{}://{}:{:d}/".format(
796 'https' if r['service_ssl_flag'] else 'http',
800 _logger.debug(str(self._gateway_services))
801 self._keep_services = [
802 ks for ks in self._gateway_services.itervalues()
803 if not ks.get('service_type', '').startswith('gateway:')]
804 self._writable_services = [ks for ks in self._keep_services
805 if not ks.get('read_only')]
807 # For disk type services, max_replicas_per_service is 1
808 # It is unknown (unlimited) for other service types.
809 if self._any_nondisk_services(self._writable_services):
810 self.max_replicas_per_service = None
812 self.max_replicas_per_service = 1
814 def _service_weight(self, data_hash, service_uuid):
815 """Compute the weight of a Keep service endpoint for a data
816 block with a known hash.
818 The weight is md5(h + u) where u is the last 15 characters of
819 the service endpoint's UUID.
821 return hashlib.md5(data_hash + service_uuid[-15:]).hexdigest()
823 def weighted_service_roots(self, locator, force_rebuild=False, need_writable=False):
824 """Return an array of Keep service endpoints, in the order in
825 which they should be probed when reading or writing data with
826 the given hash+hints.
828 self.build_services_list(force_rebuild)
831 # Use the services indicated by the given +K@... remote
832 # service hints, if any are present and can be resolved to a
834 for hint in locator.hints:
835 if hint.startswith('K@'):
838 "https://keep.{}.arvadosapi.com/".format(hint[2:]))
839 elif len(hint) == 29:
840 svc = self._gateway_services.get(hint[2:])
842 sorted_roots.append(svc['_service_root'])
844 # Sort the available local services by weight (heaviest first)
845 # for this locator, and return their service_roots (base URIs)
847 use_services = self._keep_services
849 use_services = self._writable_services
850 self.using_proxy = self._any_nondisk_services(use_services)
851 sorted_roots.extend([
852 svc['_service_root'] for svc in sorted(
855 key=lambda svc: self._service_weight(locator.md5sum, svc['uuid']))])
856 _logger.debug("{}: {}".format(locator, sorted_roots))
859 def map_new_services(self, roots_map, locator, force_rebuild, need_writable, **headers):
860 # roots_map is a dictionary, mapping Keep service root strings
861 # to KeepService objects. Poll for Keep services, and add any
862 # new ones to roots_map. Return the current list of local
864 headers.setdefault('Authorization', "OAuth2 %s" % (self.api_token,))
865 local_roots = self.weighted_service_roots(locator, force_rebuild, need_writable)
866 for root in local_roots:
867 if root not in roots_map:
868 roots_map[root] = self.KeepService(
869 root, self._user_agent_pool,
870 upload_counter=self.upload_counter,
871 download_counter=self.download_counter,
876 def _check_loop_result(result):
877 # KeepClient RetryLoops should save results as a 2-tuple: the
878 # actual result of the request, and the number of servers available
879 # to receive the request this round.
880 # This method returns True if there's a real result, False if
881 # there are no more servers available, otherwise None.
882 if isinstance(result, Exception):
884 result, tried_server_count = result
885 if (result is not None) and (result is not False):
887 elif tried_server_count < 1:
888 _logger.info("No more Keep services to try; giving up")
893 def get_from_cache(self, loc):
894 """Fetch a block only if is in the cache, otherwise return None."""
895 slot = self.block_cache.get(loc)
896 if slot is not None and slot.ready.is_set():
902 def head(self, loc_s, num_retries=None):
903 return self._get_or_head(loc_s, method="HEAD", num_retries=num_retries)
906 def get(self, loc_s, num_retries=None):
907 return self._get_or_head(loc_s, method="GET", num_retries=num_retries)
909 def _get_or_head(self, loc_s, method="GET", num_retries=None):
910 """Get data from Keep.
912 This method fetches one or more blocks of data from Keep. It
913 sends a request each Keep service registered with the API
914 server (or the proxy provided when this client was
915 instantiated), then each service named in location hints, in
916 sequence. As soon as one service provides the data, it's
920 * loc_s: A string of one or more comma-separated locators to fetch.
921 This method returns the concatenation of these blocks.
922 * num_retries: The number of times to retry GET requests to
923 *each* Keep server if it returns temporary failures, with
924 exponential backoff. Note that, in each loop, the method may try
925 to fetch data from every available Keep service, along with any
926 that are named in location hints in the locator. The default value
927 is set when the KeepClient is initialized.
930 return ''.join(self.get(x) for x in loc_s.split(','))
932 self.get_counter.add(1)
934 locator = KeepLocator(loc_s)
936 slot, first = self.block_cache.reserve_cache(locator.md5sum)
938 self.hits_counter.add(1)
942 self.misses_counter.add(1)
944 # If the locator has hints specifying a prefix (indicating a
945 # remote keepproxy) or the UUID of a local gateway service,
946 # read data from the indicated service(s) instead of the usual
947 # list of local disk services.
948 hint_roots = ['http://keep.{}.arvadosapi.com/'.format(hint[2:])
949 for hint in locator.hints if hint.startswith('K@') and len(hint) == 7]
950 hint_roots.extend([self._gateway_services[hint[2:]]['_service_root']
951 for hint in locator.hints if (
952 hint.startswith('K@') and
954 self._gateway_services.get(hint[2:])
956 # Map root URLs to their KeepService objects.
958 root: self.KeepService(root, self._user_agent_pool,
959 upload_counter=self.upload_counter,
960 download_counter=self.download_counter)
961 for root in hint_roots
964 # See #3147 for a discussion of the loop implementation. Highlights:
965 # * Refresh the list of Keep services after each failure, in case
966 # it's being updated.
967 # * Retry until we succeed, we're out of retries, or every available
968 # service has returned permanent failure.
972 loop = retry.RetryLoop(num_retries, self._check_loop_result,
974 for tries_left in loop:
976 sorted_roots = self.map_new_services(
978 force_rebuild=(tries_left < num_retries),
980 except Exception as error:
981 loop.save_result(error)
984 # Query KeepService objects that haven't returned
985 # permanent failure, in our specified shuffle order.
986 services_to_try = [roots_map[root]
987 for root in sorted_roots
988 if roots_map[root].usable()]
989 for keep_service in services_to_try:
990 blob = keep_service.get(locator, method=method, timeout=self.current_timeout(num_retries-tries_left))
993 loop.save_result((blob, len(services_to_try)))
995 # Always cache the result, then return it if we succeeded.
998 self.block_cache.cap_cache()
1000 if method == "HEAD":
1005 # Q: Including 403 is necessary for the Keep tests to continue
1006 # passing, but maybe they should expect KeepReadError instead?
1007 not_founds = sum(1 for key in sorted_roots
1008 if roots_map[key].last_result().get('status_code', None) in {403, 404, 410})
1009 service_errors = ((key, roots_map[key].last_result()['error'])
1010 for key in sorted_roots)
1012 raise arvados.errors.KeepReadError(
1013 "failed to read {}: no Keep services available ({})".format(
1014 loc_s, loop.last_result()))
1015 elif not_founds == len(sorted_roots):
1016 raise arvados.errors.NotFoundError(
1017 "{} not found".format(loc_s), service_errors)
1019 raise arvados.errors.KeepReadError(
1020 "failed to read {}".format(loc_s), service_errors, label="service")
1023 def put(self, data, copies=2, num_retries=None):
1024 """Save data in Keep.
1026 This method will get a list of Keep services from the API server, and
1027 send the data to each one simultaneously in a new thread. Once the
1028 uploads are finished, if enough copies are saved, this method returns
1029 the most recent HTTP response body. If requests fail to upload
1030 enough copies, this method raises KeepWriteError.
1033 * data: The string of data to upload.
1034 * copies: The number of copies that the user requires be saved.
1036 * num_retries: The number of times to retry PUT requests to
1037 *each* Keep server if it returns temporary failures, with
1038 exponential backoff. The default value is set when the
1039 KeepClient is initialized.
1042 if isinstance(data, unicode):
1043 data = data.encode("ascii")
1044 elif not isinstance(data, str):
1045 raise arvados.errors.ArgumentError("Argument 'data' to KeepClient.put is not type 'str'")
1047 self.put_counter.add(1)
1049 data_hash = hashlib.md5(data).hexdigest()
1050 loc_s = data_hash + '+' + str(len(data))
1053 locator = KeepLocator(loc_s)
1056 # Tell the proxy how many copies we want it to store
1057 headers['X-Keep-Desired-Replicas'] = str(copies)
1059 loop = retry.RetryLoop(num_retries, self._check_loop_result,
1062 for tries_left in loop:
1064 sorted_roots = self.map_new_services(
1066 force_rebuild=(tries_left < num_retries), need_writable=True, **headers)
1067 except Exception as error:
1068 loop.save_result(error)
1071 thread_limiter = KeepClient.ThreadLimiter(
1072 copies - done, self.max_replicas_per_service)
1074 for service_root, ks in [(root, roots_map[root])
1075 for root in sorted_roots]:
1078 t = KeepClient.KeepWriterThread(
1081 data_hash=data_hash,
1082 service_root=service_root,
1083 thread_limiter=thread_limiter,
1084 timeout=self.current_timeout(num_retries-tries_left),
1085 thread_sequence=len(threads))
1090 done += thread_limiter.done()
1091 loop.save_result((done >= copies, len(threads)))
1094 return thread_limiter.response()
1096 raise arvados.errors.KeepWriteError(
1097 "failed to write {}: no Keep services available ({})".format(
1098 data_hash, loop.last_result()))
1100 service_errors = ((key, roots_map[key].last_result()['error'])
1101 for key in sorted_roots
1102 if roots_map[key].last_result()['error'])
1103 raise arvados.errors.KeepWriteError(
1104 "failed to write {} (wanted {} copies but wrote {})".format(
1105 data_hash, copies, thread_limiter.done()), service_errors, label="service")
1107 def local_store_put(self, data, copies=1, num_retries=None):
1108 """A stub for put().
1110 This method is used in place of the real put() method when
1111 using local storage (see constructor's local_store argument).
1113 copies and num_retries arguments are ignored: they are here
1114 only for the sake of offering the same call signature as
1117 Data stored this way can be retrieved via local_store_get().
1119 md5 = hashlib.md5(data).hexdigest()
1120 locator = '%s+%d' % (md5, len(data))
1121 with open(os.path.join(self.local_store, md5 + '.tmp'), 'w') as f:
1123 os.rename(os.path.join(self.local_store, md5 + '.tmp'),
1124 os.path.join(self.local_store, md5))
1127 def local_store_get(self, loc_s, num_retries=None):
1128 """Companion to local_store_put()."""
1130 locator = KeepLocator(loc_s)
1132 raise arvados.errors.NotFoundError(
1133 "Invalid data locator: '%s'" % loc_s)
1134 if locator.md5sum == config.EMPTY_BLOCK_LOCATOR.split('+')[0]:
1136 with open(os.path.join(self.local_store, locator.md5sum), 'r') as f:
1139 def is_cached(self, locator):
1140 return self.block_cache.reserve_cache(expect_hash)