-
-
Notifications
You must be signed in to change notification settings - Fork 176
/
utils.py
748 lines (655 loc) · 26.7 KB
/
utils.py
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
import asyncio
import functools
import inspect
import json
import logging
import botocore.awsrequest
from botocore.exceptions import (
InvalidIMDSEndpointError,
MetadataRetrievalError,
)
from botocore.utils import (
DEFAULT_METADATA_SERVICE_TIMEOUT,
METADATA_BASE_URL,
RETRYABLE_HTTP_ERRORS,
ArnParser,
BadIMDSRequestError,
ClientError,
ContainerMetadataFetcher,
HTTPClientError,
IdentityCache,
IMDSFetcher,
IMDSRegionProvider,
InstanceMetadataFetcher,
InstanceMetadataRegionFetcher,
ReadTimeoutError,
S3ExpressIdentityCache,
S3ExpressIdentityResolver,
S3RegionRedirector,
S3RegionRedirectorv2,
get_environ_proxies,
os,
resolve_imds_endpoint_mode,
)
import aiobotocore.httpsession
from aiobotocore._helpers import asynccontextmanager
logger = logging.getLogger(__name__)
class _RefCountedSession(aiobotocore.httpsession.AIOHTTPSession):
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self.__ref_count = 0
self.__lock = None
@asynccontextmanager
async def acquire(self):
if not self.__lock:
self.__lock = asyncio.Lock()
# ensure we have a session
async with self.__lock:
self.__ref_count += 1
try:
if self.__ref_count == 1:
await self.__aenter__()
except BaseException:
self.__ref_count -= 1
raise
try:
yield self
finally:
async with self.__lock:
if self.__ref_count == 1:
await self.__aexit__(None, None, None)
self.__ref_count -= 1
class AioIMDSFetcher(IMDSFetcher):
def __init__(
self,
timeout=DEFAULT_METADATA_SERVICE_TIMEOUT, # noqa: E501, lgtm [py/missing-call-to-init]
num_attempts=1,
base_url=METADATA_BASE_URL,
env=None,
user_agent=None,
config=None,
session=None,
):
self._timeout = timeout
self._num_attempts = num_attempts
if config is None:
config = {}
self._base_url = self._select_base_url(base_url, config)
self._config = config
if env is None:
env = os.environ.copy()
self._disabled = (
env.get('AWS_EC2_METADATA_DISABLED', 'false').lower() == 'true'
)
self._imds_v1_disabled = config.get('ec2_metadata_v1_disabled')
self._user_agent = user_agent
self._session = session or _RefCountedSession(
timeout=self._timeout,
proxies=get_environ_proxies(self._base_url),
)
async def _fetch_metadata_token(self):
self._assert_enabled()
url = self._construct_url(self._TOKEN_PATH)
headers = {
'x-aws-ec2-metadata-token-ttl-seconds': self._TOKEN_TTL,
}
self._add_user_agent(headers)
request = botocore.awsrequest.AWSRequest(
method='PUT', url=url, headers=headers
)
async with self._session.acquire() as session:
for i in range(self._num_attempts):
try:
response = await session.send(request.prepare())
if response.status_code == 200:
return await response.text
elif response.status_code in (404, 403, 405):
return None
elif response.status_code in (400,):
raise BadIMDSRequestError(request)
except ReadTimeoutError:
return None
except RETRYABLE_HTTP_ERRORS as e:
logger.debug(
"Caught retryable HTTP exception while making metadata "
"service request to %s: %s",
url,
e,
exc_info=True,
)
except HTTPClientError as e:
error = e.kwargs.get('error')
if (
error
and getattr(error, 'errno', None) == 8
or str(getattr(error, 'os_error', None))
== 'Domain name not found'
): # threaded vs async resolver
raise InvalidIMDSEndpointError(endpoint=url, error=e)
else:
raise
return None
async def _get_request(self, url_path, retry_func, token=None):
self._assert_enabled()
if not token:
self._assert_v1_enabled()
if retry_func is None:
retry_func = self._default_retry
url = self._construct_url(url_path)
headers = {}
if token is not None:
headers['x-aws-ec2-metadata-token'] = token
self._add_user_agent(headers)
async with self._session.acquire() as session:
for i in range(self._num_attempts):
try:
request = botocore.awsrequest.AWSRequest(
method='GET', url=url, headers=headers
)
response = await session.send(request.prepare())
should_retry = retry_func(response)
if inspect.isawaitable(should_retry):
should_retry = await should_retry
if not should_retry:
return response
except RETRYABLE_HTTP_ERRORS as e:
logger.debug(
"Caught retryable HTTP exception while making metadata "
"service request to %s: %s",
url,
e,
exc_info=True,
)
raise self._RETRIES_EXCEEDED_ERROR_CLS()
async def _default_retry(self, response):
return await self._is_non_ok_response(
response
) or await self._is_empty(response)
async def _is_non_ok_response(self, response):
if response.status_code != 200:
await self._log_imds_response(response, 'non-200', log_body=True)
return True
return False
async def _is_empty(self, response):
if not await response.content:
await self._log_imds_response(response, 'no body', log_body=True)
return True
return False
async def _log_imds_response(
self, response, reason_to_log, log_body=False
):
statement = (
"Metadata service returned %s response "
"with status code of %s for url: %s"
)
logger_args = [reason_to_log, response.status_code, response.url]
if log_body:
statement += ", content body: %s"
logger_args.append(await response.content)
logger.debug(statement, *logger_args)
class AioInstanceMetadataFetcher(AioIMDSFetcher, InstanceMetadataFetcher):
async def retrieve_iam_role_credentials(self):
try:
token = await self._fetch_metadata_token()
role_name = await self._get_iam_role(token)
credentials = await self._get_credentials(role_name, token)
if self._contains_all_credential_fields(credentials):
credentials = {
'role_name': role_name,
'access_key': credentials['AccessKeyId'],
'secret_key': credentials['SecretAccessKey'],
'token': credentials['Token'],
'expiry_time': credentials['Expiration'],
}
self._evaluate_expiration(credentials)
return credentials
else:
if 'Code' in credentials and 'Message' in credentials:
logger.debug(
'Error response received when retrieving'
'credentials: %s.',
credentials,
)
return {}
except self._RETRIES_EXCEEDED_ERROR_CLS:
logger.debug(
"Max number of attempts exceeded (%s) when "
"attempting to retrieve data from metadata service.",
self._num_attempts,
)
except BadIMDSRequestError as e:
logger.debug("Bad IMDS request: %s", e.request)
return {}
async def _get_iam_role(self, token=None):
return await (
await self._get_request(
url_path=self._URL_PATH,
retry_func=self._needs_retry_for_role_name,
token=token,
)
).text
async def _get_credentials(self, role_name, token=None):
r = await self._get_request(
url_path=self._URL_PATH + role_name,
retry_func=self._needs_retry_for_credentials,
token=token,
)
return json.loads(await r.text)
async def _is_invalid_json(self, response):
try:
json.loads(await response.text)
return False
except ValueError:
await self._log_imds_response(response, 'invalid json')
return True
async def _needs_retry_for_role_name(self, response):
return await self._is_non_ok_response(
response
) or await self._is_empty(response)
async def _needs_retry_for_credentials(self, response):
return (
await self._is_non_ok_response(response)
or await self._is_empty(response)
or await self._is_invalid_json(response)
)
class AioIMDSRegionProvider(IMDSRegionProvider):
async def provide(self):
"""Provide the region value from IMDS."""
instance_region = await self._get_instance_metadata_region()
return instance_region
async def _get_instance_metadata_region(self):
fetcher = self._get_fetcher()
region = await fetcher.retrieve_region()
return region
def _create_fetcher(self):
metadata_timeout = self._session.get_config_variable(
'metadata_service_timeout'
)
metadata_num_attempts = self._session.get_config_variable(
'metadata_service_num_attempts'
)
imds_config = {
'ec2_metadata_service_endpoint': self._session.get_config_variable(
'ec2_metadata_service_endpoint'
),
'ec2_metadata_service_endpoint_mode': resolve_imds_endpoint_mode(
self._session
),
'ec2_metadata_v1_disabled': self._session.get_config_variable(
'ec2_metadata_v1_disabled'
),
}
fetcher = AioInstanceMetadataRegionFetcher(
timeout=metadata_timeout,
num_attempts=metadata_num_attempts,
env=self._environ,
user_agent=self._session.user_agent(),
config=imds_config,
)
return fetcher
class AioInstanceMetadataRegionFetcher(
AioIMDSFetcher, InstanceMetadataRegionFetcher
):
async def retrieve_region(self):
try:
region = await self._get_region()
return region
except self._RETRIES_EXCEEDED_ERROR_CLS:
logger.debug(
"Max number of attempts exceeded (%s) when "
"attempting to retrieve data from metadata service.",
self._num_attempts,
)
return None
async def _get_region(self):
token = await self._fetch_metadata_token()
response = await self._get_request(
url_path=self._URL_PATH,
retry_func=self._default_retry,
token=token,
)
availability_zone = await response.text
region = availability_zone[:-1]
return region
class AioIdentityCache(IdentityCache):
async def get_credentials(self, **kwargs):
callback = self.build_refresh_callback(**kwargs)
metadata = await callback()
credential_entry = self._credential_cls.create_from_metadata(
metadata=metadata,
refresh_using=callback,
method=self.METHOD,
advisory_timeout=45,
mandatory_timeout=10,
)
return credential_entry
class AioS3ExpressIdentityCache(AioIdentityCache, S3ExpressIdentityCache):
@functools.lru_cache(maxsize=100)
def _get_credentials(self, bucket):
return asyncio.create_task(super().get_credentials(bucket=bucket))
async def get_credentials(self, bucket):
# upstream uses `@functools.lru_cache(maxsize=100)` to cache credentials.
# This is incompatible with async code.
# We need to implement custom caching logic.
return await self._get_credentials(bucket=bucket)
def build_refresh_callback(self, bucket):
async def refresher():
response = await self._client.create_session(Bucket=bucket)
creds = response['Credentials']
expiration = self._serialize_if_needed(
creds['Expiration'], iso=True
)
return {
"access_key": creds['AccessKeyId'],
"secret_key": creds['SecretAccessKey'],
"token": creds['SessionToken'],
"expiry_time": expiration,
}
return refresher
class AioS3ExpressIdentityResolver(S3ExpressIdentityResolver):
def __init__(self, client, credential_cls, cache=None):
super().__init__(client, credential_cls, cache)
if cache is None:
cache = AioS3ExpressIdentityCache(self._client, credential_cls)
self._cache = cache
class AioS3RegionRedirectorv2(S3RegionRedirectorv2):
async def redirect_from_error(
self,
request_dict,
response,
operation,
**kwargs,
):
"""
An S3 request sent to the wrong region will return an error that
contains the endpoint the request should be sent to. This handler
will add the redirect information to the signing context and then
redirect the request.
"""
if response is None:
# This could be none if there was a ConnectionError or other
# transport error.
return
redirect_ctx = request_dict.get('context', {}).get('s3_redirect', {})
if ArnParser.is_arn(redirect_ctx.get('bucket')):
logger.debug(
'S3 request was previously for an Accesspoint ARN, not '
'redirecting.'
)
return
if redirect_ctx.get('redirected'):
logger.debug(
'S3 request was previously redirected, not redirecting.'
)
return
error = response[1].get('Error', {})
error_code = error.get('Code')
response_metadata = response[1].get('ResponseMetadata', {})
# We have to account for 400 responses because
# if we sign a Head* request with the wrong region,
# we'll get a 400 Bad Request but we won't get a
# body saying it's an "AuthorizationHeaderMalformed".
is_special_head_object = (
error_code in ('301', '400') and operation.name == 'HeadObject'
)
is_special_head_bucket = (
error_code in ('301', '400')
and operation.name == 'HeadBucket'
and 'x-amz-bucket-region'
in response_metadata.get('HTTPHeaders', {})
)
is_wrong_signing_region = (
error_code == 'AuthorizationHeaderMalformed' and 'Region' in error
)
is_redirect_status = response[0] is not None and response[
0
].status_code in (301, 302, 307)
is_permanent_redirect = error_code == 'PermanentRedirect'
if not any(
[
is_special_head_object,
is_wrong_signing_region,
is_permanent_redirect,
is_special_head_bucket,
is_redirect_status,
]
):
return
bucket = request_dict['context']['s3_redirect']['bucket']
client_region = request_dict['context'].get('client_region')
new_region = await self.get_bucket_region(bucket, response)
if new_region is None:
logger.debug(
"S3 client configured for region %s but the bucket %s is not "
"in that region and the proper region could not be "
"automatically determined." % (client_region, bucket)
)
return
logger.debug(
"S3 client configured for region %s but the bucket %s is in region"
" %s; Please configure the proper region to avoid multiple "
"unnecessary redirects and signing attempts."
% (client_region, bucket, new_region)
)
# Adding the new region to _cache will make construct_endpoint() to
# use the new region as value for the AWS::Region builtin parameter.
self._cache[bucket] = new_region
# Re-resolve endpoint with new region and modify request_dict with
# the new URL, auth scheme, and signing context.
ep_resolver = self._client._ruleset_resolver
ep_info = await ep_resolver.construct_endpoint(
operation_model=operation,
call_args=request_dict['context']['s3_redirect']['params'],
request_context=request_dict['context'],
)
request_dict['url'] = self.set_request_url(
request_dict['url'], ep_info.url
)
request_dict['context']['s3_redirect']['redirected'] = True
auth_schemes = ep_info.properties.get('authSchemes')
if auth_schemes is not None:
auth_info = ep_resolver.auth_schemes_to_signing_ctx(auth_schemes)
auth_type, signing_context = auth_info
request_dict['context']['auth_type'] = auth_type
request_dict['context']['signing'] = {
**request_dict['context'].get('signing', {}),
**signing_context,
}
# Return 0 so it doesn't wait to retry
return 0
async def get_bucket_region(self, bucket, response):
"""
There are multiple potential sources for the new region to redirect to,
but they aren't all universally available for use. This will try to
find region from response elements, but will fall back to calling
HEAD on the bucket if all else fails.
:param bucket: The bucket to find the region for. This is necessary if
the region is not available in the error response.
:param response: A response representing a service request that failed
due to incorrect region configuration.
"""
# First try to source the region from the headers.
service_response = response[1]
response_headers = service_response['ResponseMetadata']['HTTPHeaders']
if 'x-amz-bucket-region' in response_headers:
return response_headers['x-amz-bucket-region']
# Next, check the error body
region = service_response.get('Error', {}).get('Region', None)
if region is not None:
return region
# Finally, HEAD the bucket. No other choice sadly.
try:
# NOTE: we don't need to aenter/aexit as we have a ref to the base client
response = await self._client.head_bucket(Bucket=bucket)
headers = response['ResponseMetadata']['HTTPHeaders']
except ClientError as e:
headers = e.response['ResponseMetadata']['HTTPHeaders']
region = headers.get('x-amz-bucket-region', None)
return region
class AioS3RegionRedirector(S3RegionRedirector):
async def redirect_from_error(
self, request_dict, response, operation, **kwargs
):
if response is None:
# This could be none if there was a ConnectionError or other
# transport error.
return
if self._is_s3_accesspoint(request_dict.get('context', {})):
logger.debug(
'S3 request was previously to an accesspoint, not redirecting.'
)
return
if request_dict.get('context', {}).get('s3_redirected'):
logger.debug(
'S3 request was previously redirected, not redirecting.'
)
return
error = response[1].get('Error', {})
error_code = error.get('Code')
response_metadata = response[1].get('ResponseMetadata', {})
# We have to account for 400 responses because
# if we sign a Head* request with the wrong region,
# we'll get a 400 Bad Request but we won't get a
# body saying it's an "AuthorizationHeaderMalformed".
is_special_head_object = (
error_code in ('301', '400') and operation.name == 'HeadObject'
)
is_special_head_bucket = (
error_code in ('301', '400')
and operation.name == 'HeadBucket'
and 'x-amz-bucket-region'
in response_metadata.get('HTTPHeaders', {})
)
is_wrong_signing_region = (
error_code == 'AuthorizationHeaderMalformed' and 'Region' in error
)
is_redirect_status = response[0] is not None and response[
0
].status_code in (301, 302, 307)
is_permanent_redirect = error_code == 'PermanentRedirect'
if not any(
[
is_special_head_object,
is_wrong_signing_region,
is_permanent_redirect,
is_special_head_bucket,
is_redirect_status,
]
):
return
bucket = request_dict['context']['signing']['bucket']
client_region = request_dict['context'].get('client_region')
new_region = await self.get_bucket_region(bucket, response)
if new_region is None:
logger.debug(
"S3 client configured for region %s but the bucket %s is not "
"in that region and the proper region could not be "
"automatically determined." % (client_region, bucket)
)
return
logger.debug(
"S3 client configured for region %s but the bucket %s is in region"
" %s; Please configure the proper region to avoid multiple "
"unnecessary redirects and signing attempts."
% (client_region, bucket, new_region)
)
endpoint = self._endpoint_resolver.resolve('s3', new_region)
endpoint = endpoint['endpoint_url']
signing_context = {
'region': new_region,
'bucket': bucket,
'endpoint': endpoint,
}
request_dict['context']['signing'] = signing_context
self._cache[bucket] = signing_context
self.set_request_url(request_dict, request_dict['context'])
request_dict['context']['s3_redirected'] = True
# Return 0 so it doesn't wait to retry
return 0
async def get_bucket_region(self, bucket, response):
# First try to source the region from the headers.
service_response = response[1]
response_headers = service_response['ResponseMetadata']['HTTPHeaders']
if 'x-amz-bucket-region' in response_headers:
return response_headers['x-amz-bucket-region']
# Next, check the error body
region = service_response.get('Error', {}).get('Region', None)
if region is not None:
return region
# Finally, HEAD the bucket. No other choice sadly.
try:
# NOTE: we don't need to aenter/aexit as we have a ref to the base client
response = await self._client.head_bucket(Bucket=bucket)
headers = response['ResponseMetadata']['HTTPHeaders']
except ClientError as e:
headers = e.response['ResponseMetadata']['HTTPHeaders']
region = headers.get('x-amz-bucket-region', None)
return region
class AioContainerMetadataFetcher(ContainerMetadataFetcher):
def __init__(
self, session=None, sleep=asyncio.sleep
): # noqa: E501, lgtm [py/missing-call-to-init]
if session is None:
session = _RefCountedSession(timeout=self.TIMEOUT_SECONDS)
self._session = session
self._sleep = sleep
async def retrieve_full_uri(self, full_url, headers=None):
self._validate_allowed_url(full_url)
return await self._retrieve_credentials(full_url, headers)
async def retrieve_uri(self, relative_uri):
"""Retrieve JSON metadata from container metadata.
:type relative_uri: str
:param relative_uri: A relative URI, e.g "/foo/bar?id=123"
:return: The parsed JSON response.
"""
full_url = self.full_url(relative_uri)
return await self._retrieve_credentials(full_url)
async def _retrieve_credentials(self, full_url, extra_headers=None):
headers = {'Accept': 'application/json'}
if extra_headers is not None:
headers.update(extra_headers)
attempts = 0
while True:
try:
return await self._get_response(
full_url, headers, self.TIMEOUT_SECONDS
)
except MetadataRetrievalError as e:
logger.debug(
"Received error when attempting to retrieve "
"container metadata: %s",
e,
exc_info=True,
)
await self._sleep(self.SLEEP_TIME)
attempts += 1
if attempts >= self.RETRY_ATTEMPTS:
raise
async def _get_response(self, full_url, headers, timeout):
try:
async with self._session.acquire() as session:
AWSRequest = botocore.awsrequest.AWSRequest
request = AWSRequest(
method='GET', url=full_url, headers=headers
)
response = await session.send(request.prepare())
response_text = (await response.content).decode('utf-8')
if response.status_code != 200:
raise MetadataRetrievalError(
error_msg=(
f"Received non 200 response {response.status_code} "
f"from container metadata: {response_text}"
)
)
try:
return json.loads(response_text)
except ValueError:
error_msg = "Unable to parse JSON returned from container metadata services"
logger.debug('%s:%s', error_msg, response_text)
raise MetadataRetrievalError(error_msg=error_msg)
except RETRYABLE_HTTP_ERRORS as e:
error_msg = (
"Received error when attempting to retrieve "
f"container metadata: {e}"
)
raise MetadataRetrievalError(error_msg=error_msg)