/usr/lib/python2.7/dist-packages/swift/proxy/controllers/base.py is in python-swift 2.7.0-0ubuntu2.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 | # Copyright (c) 2010-2016 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, \
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.header_key_dict import HeaderKeyDict
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, 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, \
http_response_to_document_iters
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
return func
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', '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 = set([
'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.add(header.lower())
if cors_info.get('expose_headers'):
expose_headers = expose_headers.union(
[header_line.strip().lower()
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 information 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")
if isinstance(info[key], dict):
for subkey, value in info[key].items():
if isinstance(value, six.text_type):
info[key][subkey] = value.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, concurrency=1, 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 = ''
self.concurrency = concurrency
# 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 = []
self.sources = []
# 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()
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 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.warning(
_('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.warning(_('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 _make_node_request(self, node, node_timeout, logger_thread_locals):
self.app.logger.thread_locals = logger_thread_locals
if node in self.used_nodes:
return False
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})
return False
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([])
return False
self.statuses.append(possible_source.status)
self.reasons.append(possible_source.reason)
self.bodies.append(None)
self.source_headers.append(possible_source.getheaders())
self.sources.append((possible_source, node))
if not self.newest: # one good source is enough
return True
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})
return False
def _get_source_and_node(self):
self.statuses = []
self.reasons = []
self.bodies = []
self.source_headers = []
self.sources = []
nodes = GreenthreadSafeIterator(self.node_iter)
node_timeout = self.app.node_timeout
if self.server_type == 'Object' and not self.newest:
node_timeout = self.app.recoverable_node_timeout
pile = GreenAsyncPile(self.concurrency)
for node in nodes:
pile.spawn(self._make_node_request, node, node_timeout,
self.app.logger.thread_locals)
_timeout = self.app.concurrency_timeout \
if pile.inflight < self.concurrency else None
if pile.waitfirst(_timeout):
break
else:
# ran out of nodes, see if any stragglers will finish
any(pile)
if self.sources:
self.sources.sort(key=lambda s: source_key(s[0]))
source, node = self.sources.pop()
for src, _junk in self.sources:
close_swift_conn(src)
self.used_nodes.append(node)
src_headers = dict(
(k.lower(), v) for k, v in
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 req: 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,
concurrency=1, 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 concurrency: number of requests to run concurrently
: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,
concurrency,
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
|