aboutsummaryrefslogtreecommitdiffhomepage
path: root/doc/interop-test-descriptions.md
blob: 96d1f2bd7526c3421e867ed94ba9986a10f2b358 (plain)
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
Interoperability Test Case Descriptions
=======================================

Client and server use
[test.proto](../src/proto/grpc/testing/test.proto)
and the [gRPC over HTTP/2 v2 protocol](./PROTOCOL-HTTP2.md).

Client
------

Clients implement test cases that test certain functionally. Each client is
provided the test case it is expected to run as a command-line parameter. Names
should be lowercase and without spaces.

Clients should accept these arguments:
* --server_host=HOSTNAME
    * The server host to connect to. For example, "localhost" or "127.0.0.1"
* --server_host_override=HOSTNAME
    * The server host to claim to be connecting to, for use in TLS and HTTP/2
      :authority header. If unspecified, the value of --server_host will be
      used
* --server_port=PORT
    * The server port to connect to. For example, "8080"
* --test_case=TESTCASE
    * The name of the test case to execute. For example, "empty_unary"
* --use_tls=BOOLEAN
    * Whether to use a plaintext or encrypted connection
* --use_test_ca=BOOLEAN
    * Whether to replace platform root CAs with
      [ca.pem](https://github.com/grpc/grpc/blob/master/src/core/lib/tsi/test_creds/ca.pem)
      as the CA root
* --default_service_account=ACCOUNT_EMAIL
    * Email of the GCE default service account.
* --oauth_scope=SCOPE
    * OAuth scope. For example, "https://www.googleapis.com/auth/xapi.zoo"
* --service_account_key_file=PATH
    * The path to the service account JSON key file generated from GCE developer
    console.

Clients must support TLS with ALPN. Clients must not disable certificate
checking.

### empty_unary

This test verifies that implementations support zero-size messages. Ideally,
client implementations would verify that the request and response were zero
bytes serialized, but this is generally prohibitive to perform, so is not
required.

Server features:
* [EmptyCall][]

Procedure:
 1. Client calls EmptyCall with the default Empty message

Client asserts:
* call was successful
* response is non-null

*It may be possible to use UnaryCall instead of EmptyCall, but it is harder to
ensure that the proto serialized to zero bytes.*

### cacheable_unary

This test verifies that gRPC requests marked as cacheable use GET verb instead
of POST, and that server sets appropriate cache control headers for the response
to be cached by a proxy. This test requires that the server is behind
a caching proxy. Use of current timestamp in the request prevents accidental
cache matches left over from previous tests.

Server features:
* [CacheableUnaryCall][]

Procedure:
 1. Client calls CacheableUnaryCall with `SimpleRequest` request with payload
    set to current timestamp. Timestamp format is irrelevant, and resolution is
    in nanoseconds.
    Client adds a `x-user-ip` header with value `1.2.3.4` to the request.
    This is done since some proxys such as GFE will not cache requests from
    localhost.
    Client marks the request as cacheable by setting the cacheable flag in the
    request context. Longer term this should be driven by the method option
    specified in the proto file itself.
 2. Client calls CacheableUnaryCall with `SimpleRequest` request again
    immediately with the same payload as the previous request. Cacheable flag is
    also set for this request's context.

Client asserts:
* Both calls were successful
* The payload body of both responses is the same.

### large_unary

This test verifies unary calls succeed in sending messages, and touches on flow
control (even if compression is enabled on the channel).

Server features:
* [UnaryCall][]

Procedure:
 1. Client calls UnaryCall with:

    ```
    {
      response_size: 314159
      payload:{
        body: 271828 bytes of zeros
      }
    }
    ```

Client asserts:
* call was successful
* response payload body is 314159 bytes in size
* clients are free to assert that the response payload body contents are zero
  and comparing the entire response message against a golden response

### client_compressed_unary

This test verifies the client can compress unary messages by sending two unary
calls, for compressed and uncompressed payloads. It also sends an initial
probing request to verify whether the server supports the [CompressedRequest][]
feature by checking if the probing call fails with an `INVALID_ARGUMENT` status.

Server features:
* [UnaryCall][]
* [CompressedRequest][]

Procedure:
 1. Client calls UnaryCall with the feature probe, an *uncompressed* message:
    ```
    {
      expect_compressed:{
        value: true
      }
      response_size: 314159
      payload:{
        body: 271828 bytes of zeros
      }
    }
    ```

 1. Client calls UnaryCall with the *compressed* message:

    ```
    {
      expect_compressed:{
        value: true
      }
      response_size: 314159
      payload:{
        body: 271828 bytes of zeros
      }
    }
    ```

 1. Client calls UnaryCall with the *uncompressed* message:

    ```
    {
      expect_compressed:{
        value: false
      }
      response_size: 314159
      payload:{
        body: 271828 bytes of zeros
      }
    }
    ```

    Client asserts:
    * First call failed with `INVALID_ARGUMENT` status.
    * Subsequent calls were successful.
    * Response payload body is 314159 bytes in size.
    * Clients are free to assert that the response payload body contents are
      zeros and comparing the entire response message against a golden response.


### server_compressed_unary

This test verifies the server can compress unary messages. It sends two unary
requests, expecting the server's response to be compressed or not according to
the `response_compressed` boolean.

Whether compression was actually performed is determined by the compression bit
in the response's message flags. *Note that some languages may not have access
to the message flags*.


Server features:
* [UnaryCall][]
* [CompressedResponse][]

Procedure:
 1. Client calls UnaryCall with `SimpleRequest`:

    ```
    {
      response_compressed:{
        value: true
      }
      response_size: 314159
      payload:{
        body: 271828 bytes of zeros
      }
    }
    ```

    ```
    {
      response_compressed:{
        value: false
      }
      response_size: 314159
      payload:{
        body: 271828 bytes of zeros
      }
    }
    ```
    Client asserts:
    * call was successful
    * when `response_compressed` is true, the response MUST have the
      compressed message flag set.
    * when `response_compressed` is false, the response MUST NOT have
      the compressed message flag set.
    * response payload body is 314159 bytes in size in both cases.
    * clients are free to assert that the response payload body contents are
      zero and comparing the entire response message against a golden response


### client_streaming

This test verifies that client-only streaming succeeds.

Server features:
* [StreamingInputCall][]

Procedure:
 1. Client calls StreamingInputCall
 2. Client sends:

    ```
    {
      payload:{
        body: 27182 bytes of zeros
      }
    }
    ```

 3. Client then sends:

    ```
    {
      payload:{
        body: 8 bytes of zeros
      }
    }
    ```

 4. Client then sends:

    ```
    {
      payload:{
        body: 1828 bytes of zeros
      }
    }
    ```

 5. Client then sends:

    ```
    {
      payload:{
        body: 45904 bytes of zeros
      }
    }
    ```

 6. Client half-closes

Client asserts:
* call was successful
* response aggregated_payload_size is 74922


### client_compressed_streaming

This test verifies the client can compress requests on per-message basis by
performing a two-request streaming call. It also sends an initial probing
request to verify whether the server supports the [CompressedRequest][] feature
by checking if the probing call fails with an `INVALID_ARGUMENT` status.

Procedure:
 1. Client calls `StreamingInputCall` and sends the following feature-probing
    *uncompressed* `StreamingInputCallRequest` message

    ```
    {
      expect_compressed:{
        value: true
      }
      payload:{
        body: 27182 bytes of zeros
      }
    }
    ```
    If the call fails with `INVALID_ARGUMENT`, the test fails. Otherwise, we
    continue.

 1. Client calls `StreamingInputCall` again, sending the *compressed* message

    ```
    {
      expect_compressed:{
        value: true
      }
      payload:{
        body: 27182 bytes of zeros
      }
    }
    ```

 1. And finally, the *uncompressed* message
    ```
    {
      expect_compressed:{
        value: false
      }
      payload:{
        body: 45904 bytes of zeros
      }
    }
    ```

 1. Client half-closes

Client asserts:
* First call fails with `INVALID_ARGUMENT`.
* Next calls succeeds.
* Response aggregated payload size is 73086.


### server_streaming

This test verifies that server-only streaming succeeds.

Server features:
* [StreamingOutputCall][]

Procedure:
 1. Client calls StreamingOutputCall with `StreamingOutputCallRequest`:

    ```
    {
      response_parameters:{
        size: 31415
      }
      response_parameters:{
        size: 9
      }
      response_parameters:{
        size: 2653
      }
      response_parameters:{
        size: 58979
      }
    }
    ```

Client asserts:
* call was successful
* exactly four responses
* response payload bodies are sized (in order): 31415, 9, 2653, 58979
* clients are free to assert that the response payload body contents are zero
  and comparing the entire response messages against golden responses

### server_compressed_streaming

This test verifies that the server can compress streaming messages and disable
compression on individual messages.

Server features:
* [StreamingOutputCall][]
* [CompressedResponse][]


Procedure:
 1. Client calls StreamingOutputCall with `StreamingOutputCallRequest`:

    ```
    {
      response_parameters:{
        compressed: {
          value: true
        }
        size: 31415
      }
      response_parameters:{
        compressed: {
          value: false
        }
        size: 92653
      }
    }
    ```

    Client asserts:
    * call was successful
    * exactly two responses
    * when `response_compressed` is false, the response's messages MUST
      NOT have the compressed message flag set.
    * when `response_compressed` is true, the response's messages MUST
      have the compressed message flag set.
    * response payload bodies are sized (in order): 31415, 92653
    * clients are free to assert that the response payload body contents are
      zero and comparing the entire response messages against golden responses


### ping_pong

This test verifies that full duplex bidi is supported.

Server features:
* [FullDuplexCall][]

Procedure:
 1. Client calls FullDuplexCall with:

    ```
    {
      response_parameters:{
        size: 31415
      }
      payload:{
        body: 27182 bytes of zeros
      }
    }
    ```

 2. After getting a reply, it sends:

    ```
    {
      response_parameters:{
        size: 9
      }
      payload:{
        body: 8 bytes of zeros
      }
    }
    ```

 3. After getting a reply, it sends:

    ```
    {
      response_parameters:{
        size: 2653
      }
      payload:{
        body: 1828 bytes of zeros
      }
    }
    ```

 4. After getting a reply, it sends:

    ```
    {
      response_parameters:{
        size: 58979
      }
      payload:{
        body: 45904 bytes of zeros
      }
    }
    ```

 5. After getting a reply, client half-closes

Client asserts:
* call was successful
* exactly four responses
* response payload bodies are sized (in order): 31415, 9, 2653, 58979
* clients are free to assert that the response payload body contents are zero
  and comparing the entire response messages against golden responses

### empty_stream

This test verifies that streams support having zero-messages in both
directions.

Server features:
* [FullDuplexCall][]

Procedure:
 1. Client calls FullDuplexCall and then half-closes

Client asserts:
* call was successful
* exactly zero responses

### compute_engine_creds

This test is only for cloud-to-prod path.

This test verifies unary calls succeed in sending messages while using Service
Credentials from GCE metadata server. The client instance needs to be created
with desired oauth scope.

The test uses `--default_service_account` with GCE service account email and
`--oauth_scope` with the OAuth scope to use. For testing against
grpc-test.sandbox.googleapis.com, "https://www.googleapis.com/auth/xapi.zoo"
should
be passed in as `--oauth_scope`.

Server features:
* [UnaryCall][]
* [Echo Authenticated Username][]
* [Echo OAuth Scope][]

Procedure:
 1. Client configures channel to use GCECredentials
 2. Client calls UnaryCall on the channel with:

    ```
    {
      response_size: 314159
      payload:{
        body: 271828 bytes of zeros
      }
      fill_username: true
      fill_oauth_scope: true
    }
    ```

Client asserts:
* call was successful
* received SimpleResponse.username equals the value of
  `--default_service_account` flag
* received SimpleResponse.oauth_scope is in `--oauth_scope`
* response payload body is 314159 bytes in size
* clients are free to assert that the response payload body contents are zero
  and comparing the entire response message against a golden response

### jwt_token_creds

This test is only for cloud-to-prod path.

This test verifies unary calls succeed in sending messages while using JWT
token (created by the project's key file)

Test caller should set flag `--service_account_key_file` with the
path to json key file downloaded from
https://console.developers.google.com. Alternately, if using a
usable auth implementation, she may specify the file location in the environment
variable GOOGLE_APPLICATION_CREDENTIALS.

Server features:
* [UnaryCall][]
* [Echo Authenticated Username][]
* [Echo OAuth Scope][]

Procedure:
 1. Client configures the channel to use JWTTokenCredentials
 2. Client calls UnaryCall with:

    ```
    {
      response_size: 314159
      payload:{
        body: 271828 bytes of zeros
      }
      fill_username: true
    }
    ```

Client asserts:
* call was successful
* received SimpleResponse.username is not empty and is in the json key file used
by the auth library. The client can optionally check the username matches the
email address in the key file or equals the value of `--default_service_account`
flag.
* response payload body is 314159 bytes in size
* clients are free to assert that the response payload body contents are zero
  and comparing the entire response message against a golden response

### oauth2_auth_token

This test is only for cloud-to-prod path and some implementations may run
in GCE only.

This test verifies unary calls succeed in sending messages using an OAuth2 token
that is obtained out of band. For the purpose of the test, the OAuth2 token is
actually obtained from a service account credentials or GCE credentials via the
language-specific authorization library.

The difference between this test and the other auth tests is that it
first uses the authorization library to obtain an authorization token.

The test
- uses the flag `--service_account_key_file` with the path to a json key file
downloaded from https://console.developers.google.com. Alternately, if using a
usable auth implementation, it may specify the file location in the environment
variable GOOGLE_APPLICATION_CREDENTIALS, *OR* if GCE credentials is used to
fetch the token, `--default_service_account` can be used to pass in GCE service
account email.
- uses the flag `--oauth_scope` for the oauth scope.  For testing against
grpc-test.sandbox.googleapis.com, "https://www.googleapis.com/auth/xapi.zoo"
should be passed as the `--oauth_scope`.

Server features:
* [UnaryCall][]
* [Echo Authenticated Username][]
* [Echo OAuth Scope][]

Procedure:
 1. Client uses the auth library to obtain an authorization token
 2. Client configures the channel to use AccessTokenCredentials with the access
    token obtained in step 1
 3. Client calls UnaryCall with the following message

    ```
    {
      fill_username: true
      fill_oauth_scope: true
    }
    ```

Client asserts:
* call was successful
* received SimpleResponse.username is valid. Depending on whether a service
account key file or GCE credentials was used, client should check against the
json key file or GCE default service account email.
* received SimpleResponse.oauth_scope is in `--oauth_scope`

### per_rpc_creds

Similar to the other auth tests, this test is only for cloud-to-prod path.

This test verifies unary calls succeed in sending messages using a JWT or a
service account credentials set on the RPC.

The test
- uses the flag `--service_account_key_file` with the path to a json key file
downloaded from https://console.developers.google.com. Alternately, if using a
usable auth implementation, it may specify the file location in the environment
variable GOOGLE_APPLICATION_CREDENTIALS
- optionally uses the flag `--oauth_scope` for the oauth scope if implementator
wishes to use service account credential instead of JWT credential. For testing
against grpc-test.sandbox.googleapis.com, oauth scope
"https://www.googleapis.com/auth/xapi.zoo" should be used.

Server features:
* [UnaryCall][]
* [Echo Authenticated Username][]
* [Echo OAuth Scope][]

Procedure:
 1. Client configures the channel with just SSL credentials
 2. Client calls UnaryCall, setting per-call credentials to
    JWTTokenCredentials. The request is the following message

    ```
    {
      fill_username: true
    }
    ```

Client asserts:
* call was successful
* received SimpleResponse.username is not empty and is in the json key file used
by the auth library. The client can optionally check the username matches the
email address in the key file.


### custom_metadata

This test verifies that custom metadata in either binary or ascii format can be
sent as initial-metadata by the client and as both initial- and trailing-metadata
by the server.

Server features:
* [UnaryCall][]
* [FullDuplexCall][]
* [Echo Metadata][]

Procedure:
 1. The client attaches custom metadata with the following keys and values:

    ```
    key: "x-grpc-test-echo-initial", value: "test_initial_metadata_value"
    key: "x-grpc-test-echo-trailing-bin", value: 0xababab
    ```

    to a UnaryCall with request:

    ```
    {
      response_size: 314159
      payload:{
        body: 271828 bytes of zeros
      }
    }
    ```

 2. The client attaches custom metadata with the following keys and values:

    ```
    key: "x-grpc-test-echo-initial", value: "test_initial_metadata_value"
    key: "x-grpc-test-echo-trailing-bin", value: 0xababab
    ```

    to a FullDuplexCall with request:

    ```
    {
      response_size: 314159
      payload:{
        body: 271828 bytes of zeros
      }
    }
    ```

    and then half-closes

Client asserts:
* call was successful
* metadata with key `"x-grpc-test-echo-initial"` and value
  `"test_initial_metadata_value"`is received in the initial metadata for calls
  in Procedure steps 1 and 2.
* metadata with key `"x-grpc-test-echo-trailing-bin"` and value `0xababab` is
  received in the trailing metadata for calls in Procedure steps 1 and 2.



### status_code_and_message

This test verifies unary calls succeed in sending messages, and propagate back
status code and message sent along with the messages.

Server features:
* [UnaryCall][]
* [FullDuplexCall][]
* [Echo Status][]

Procedure:
 1. Client calls UnaryCall with:

    ```
    {
      response_status:{
        code: 2
        message: "test status message"
      }
    }
    ```

 2. Client calls FullDuplexCall with:

    ```
    {
      response_status:{
        code: 2
        message: "test status message"
      }
    }
    ```

    and then half-closes


Client asserts:
* received status code is the same as the sent code for both Procedure steps 1
  and 2
* received status message is the same as the sent message for both Procedure
  steps 1 and 2

### unimplemented_method

This test verifies that calling an unimplemented RPC method returns the
UNIMPLEMENTED status code.

Server features:
N/A

Procedure:
* Client calls `grpc.testing.TestService/UnimplementedCall` with an empty
  request (defined as `grpc.testing.Empty`):

    ```
    {
    }
    ```

Client asserts:
* received status code is 12 (UNIMPLEMENTED)

### unimplemented_service

This test verifies calling an unimplemented server returns the UNIMPLEMENTED
status code.

Server features:
N/A

Procedure:
* Client calls `grpc.testing.UnimplementedService/UnimplementedCall` with an
  empty request (defined as `grpc.testing.Empty`)

Client asserts:
* received status code is 12 (UNIMPLEMENTED)

### cancel_after_begin

This test verifies that a request can be cancelled after metadata has been sent
but before payloads are sent.

Server features:
* [StreamingInputCall][]

Procedure:
 1. Client starts StreamingInputCall
 2. Client immediately cancels request

Client asserts:
* Call completed with status CANCELLED

### cancel_after_first_response

This test verifies that a request can be cancelled after receiving a message
from the server.

Server features:
* [FullDuplexCall][]

Procedure:
 1. Client starts FullDuplexCall with

    ```
    {
      response_parameters:{
        size: 31415
      }
      payload:{
        body: 27182 bytes of zeros
      }
    }
    ```

 2. After receiving a response, client cancels request

Client asserts:
* Call completed with status CANCELLED

### timeout_on_sleeping_server

This test verifies that an RPC request whose lifetime exceeds its configured
timeout value will end with the DeadlineExceeded status.

Server features:
* [FullDuplexCall][]

Procedure:
 1. Client calls FullDuplexCall with the following request and sets its timeout
    to 1ms

    ```
    {
      payload:{
        body: 27182 bytes of zeros
      }
    }
    ```

 2. Client waits

Client asserts:
* Call completed with status DEADLINE_EXCEEDED.

### concurrent_large_unary

Status: TODO

Client performs 1000 large_unary tests in parallel on the same channel.

### Flow control. Pushback at client for large messages (TODO: fix name)

Status: TODO

This test verifies that a client sending faster than a server can drain sees
pushback (i.e., attempts to send succeed only after appropriate delays).

### TODO Tests

#### High priority:

Propagation of status code and message (yangg)

Multiple thousand simultaneous calls on same Channel (ctiller)

Metadata: client headers, server headers + trailers, binary+ascii

#### Normal priority:

Cancel before start (ctiller)

Cancel after sent first message (ctiller)

Cancel after received headers (ctiller)

Timeout but completed before expire (zhaoq)

Multiple thousand simultaneous calls timeout on same Channel (ctiller)

#### Lower priority:

Flow control. Pushback at client for large messages (abhishek)

Flow control. Pushback at server for large messages (abhishek)

Going over max concurrent streams doesn't fail (client controls itself)
(abhishek)

RPC method not implemented (yangg)

Multiple thousand simultaneous calls on different Channels (ctiller)

Failed TLS hostname verification (ejona?)

Large amount of headers to cause CONTINUATIONs; 63K of 'X's, all in one header.

#### To priorize:

Start streaming RPC but don't send any requests, server responds

### Postponed Tests

Resilience to buggy servers: These tests would verify that a client application
isn't affected negatively by the responses put on the wire by a buggy server
(e.g. the client library won't make the application crash).

Reconnect after transport failure

Reconnect backoff

Fuzz testing


Server
------

Servers implement various named features for clients to test with. Server
features are orthogonal. If a server implements a feature, it is always
available for clients. Names are simple descriptions for developer
communication and tracking.

Servers should accept these arguments:

* --port=PORT

    * The port to listen on. For example, "8080"

* --use_tls=BOOLEAN

    * Whether to use a plaintext or encrypted connection

Servers must support TLS with ALPN. They should use
[server1.pem](https://github.com/grpc/grpc/blob/master/src/core/lib/tsi/test_creds/server1.pem)
for their certificate.

### EmptyCall
[EmptyCall]: #emptycall

Server implements EmptyCall which immediately returns the empty message.

### UnaryCall
[UnaryCall]: #unarycall

Server implements UnaryCall which immediately returns a SimpleResponse with a
payload body of size `SimpleRequest.response_size` bytes and type as appropriate
for the `SimpleRequest.response_type`. If the server does not support the
`response_type`, then it should fail the RPC with `INVALID_ARGUMENT`.

### CacheableUnaryCall

Server gets the default SimpleRequest proto as the request. The content of the
request is ignored. It returns the SimpleResponse proto with the payload set
to current timestamp.  The timestamp is an integer representing current time
with nanosecond resolution. This integer is formated as ASCII decimal in the
response. The format is not really important as long as the response payload
is different for each request. In addition it adds
  1. cache control headers such that the response can be cached by proxies in
     the response path. Server should be behind a caching proxy for this test
     to pass. Currently we set the max-age to 60 seconds.

### CompressedResponse
[CompressedResponse]: #compressedresponse

When the client sets `response_compressed` to true, the server's response is
sent back compressed. Note that `response_compressed` is present on both
`SimpleRequest` (unary) and `StreamingOutputCallRequest` (streaming).

### CompressedRequest
[CompressedRequest]: #compressedrequest

When the client sets `expect_compressed` to true, the server expects the client
request to be compressed. If it's not, it fails the RPC with `INVALID_ARGUMENT`.
Note that `response_compressed` is present on both `SimpleRequest` (unary) and
`StreamingOutputCallRequest` (streaming).

### StreamingInputCall
[StreamingInputCall]: #streaminginputcall

Server implements StreamingInputCall which upon half close immediately returns
a StreamingInputCallResponse where aggregated_payload_size is the sum of all
request payload bodies received.

### StreamingOutputCall
[StreamingOutputCall]: #streamingoutputcall

Server implements StreamingOutputCall by replying, in order, with one
StreamingOutputCallResponse for each ResponseParameters in
StreamingOutputCallRequest. Each StreamingOutputCallResponse should have a
payload body of size ResponseParameters.size bytes, as specified by its
respective ResponseParameters. After sending all responses, it closes with OK.

### FullDuplexCall
[FullDuplexCall]: #fullduplexcall

Server implements FullDuplexCall by replying, in order, with one
StreamingOutputCallResponse for each ResponseParameters in each
StreamingOutputCallRequest. Each StreamingOutputCallResponse should have a
payload body of size ResponseParameters.size bytes, as specified by its
respective ResponseParameters. After receiving half close and sending all
responses, it closes with OK.

### Echo Status
[Echo Status]: #echo-status
When the client sends a response_status in the request payload, the server closes
the stream with the status code and messsage contained within said response_status.
The server will not process any further messages on the stream sent by the client.
This can be used by clients to verify correct handling of different status codes and
associated status messages end-to-end.

### Echo Metadata
[Echo Metadata]: #echo-metadata
When the client sends metadata with the key `"x-grpc-test-echo-initial"` with its
request, the server sends back exactly this key and the corresponding value back to
the client as part of initial metadata. When the client sends metadata with the key
`"x-grpc-test-echo-trailing-bin"` with its request, the server sends back exactly this
key and the corresponding value back to the client as trailing metadata.

### Observe ResponseParameters.interval_us
[Observe ResponseParameters.interval_us]: #observe-responseparametersinterval_us

In StreamingOutputCall and FullDuplexCall, server delays sending a
StreamingOutputCallResponse by the ResponseParameters's `interval_us` for that
particular response, relative to the last response sent. That is, `interval_us`
acts like a sleep *before* sending the response and accumulates from one
response to the next.

Interaction with flow control is unspecified.

### Echo Auth Information

Status: Pending

#### Echo Authenticated Username
[Echo Authenticated Username]: #echo-authenticated-username

If a SimpleRequest has fill_username=true and that request was successfully
authenticated, then the SimpleResponse should have username filled with the
canonical form of the authenticated source. The canonical form is dependent on
the authentication method, but is likely to be a base 10 integer identifier or
an email address.

#### Echo OAuth scope
[Echo OAuth Scope]: #echo-oauth-scope

If a SimpleRequest has `fill_oauth_scope=true` and that request was successfully
authenticated via OAuth, then the SimpleResponse should have oauth_scope filled
with the scope of the method being invoked.

Although a general server-side feature, most test servers won't implement this
feature. The TLS server `grpc-test.sandbox.googleapis.com:443` supports this
feature. It requires at least the OAuth scope
`https://www.googleapis.com/auth/xapi.zoo` for authentication to succeed.

Discussion:

Ideally, this would be communicated via metadata and not in the
request/response, but we want to use this test in code paths that don't yet
fully communicate metadata.