Fail test if client or server pods restarted during test.
#### Testing
Tested locally, test will fail with message similar to:
```
----------------------------------------------------------------------
Traceback (most recent call last):
File "/usr/local/google/home/xuanwn/workspace/xds/grpc/tools/run_tests/xds_k8s_test_driver/framework/xds_k8s_testcase.py", line 501, in tearDown
))
AssertionError: 5 != 0 : Server pods unexpectedly restarted {sever_restarts} times during test.
----------------------------------------------------------------------
Ran 1 test in 886.867s
```
This is a mistake made in https://github.com/grpc/grpc/pull/33030.
`sizeof()` would count the null byte terminated the C string and would
cause us to skip a byte if it is used as the index to
`result->substr()`. This would also crash if `result` only contains
`grpc_config=` as @drfloob pointed out.
<!--
If you know who should review your pull request, please assign it to
that
person, otherwise the pull request would get assigned randomly.
If your pull request is for a specific language, please add the
appropriate
lang label.
-->
This metadata doesn't actually encode so passing it through from an app
will force a crash.
<!--
If you know who should review your pull request, please assign it to
that
person, otherwise the pull request would get assigned randomly.
If your pull request is for a specific language, please add the
appropriate
lang label.
-->
Instead just Utf-16 encode the null byte when dumping the value to a
string form.
<!--
If you know who should review your pull request, please assign it to
that
person, otherwise the pull request would get assigned randomly.
If your pull request is for a specific language, please add the
appropriate
lang label.
-->
Better logging for `assertRpcStatusCodes`.
(got tired of looking up the status names)
#### Unexpected status found
Before:
```
AssertionError: AssertionError: Expected only status 15 but found status 0 for method UNARY_CALL:
stats_per_method {
key: "UNARY_CALL"
value {
result {
key: 0
value: 251
}
}
}
```
After:
```
AssertionError: Expected only status (15, DATA_LOSS), but found status (0, OK) for method UNARY_CALL:
stats_per_method {
key: "UNARY_CALL"
value {
result {
key: 0
value: 251
}
}
}
```
#### No traffic with expected status
Before:
```
AssertionError: 0 not greater than 0
```
After:
```
AssertionError: 0 not greater than 0 : Expected non-zero RPCs with status (15, DATA_LOSS) for method UNARY_CALL, got:
stats_per_method {
key: "UNARY_CALL"
value {
result {
key: 0
value: 251
}
result {
key: 15
value: 0
}
}
}
```
#thistimeforsure
a863532c62 adds some debug to help track
which batches get leaked by a transport
3203e75ec5 makes connected_channel respect
the high level intent of cancellation better (and fixes the last reason
we needed to turn these tests off)
aaf5fa036b re-enables testing of c++ e2e
tests with server based promise calls
<!--
If you know who should review your pull request, please assign it to
that
person, otherwise the pull request would get assigned randomly.
If your pull request is for a specific language, please add the
appropriate
lang label.
-->
<!--
If you know who should review your pull request, please assign it to
that
person, otherwise the pull request would get assigned randomly.
If your pull request is for a specific language, please add the
appropriate
lang label.
-->
This will change behavior for tests that have experiments enabled on
them to always have the flaky bit on.
In doing so, we'll get the usual failure reporting we do in the internal
chat bot, but allow PRs to pass even if an experiment isn't 100% passing
yet - reducing friction slightly for landing bigger experiments.
<!--
If you know who should review your pull request, please assign it to
that
person, otherwise the pull request would get assigned randomly.
If your pull request is for a specific language, please add the
appropriate
lang label.
-->
Fixes `FakeXdsTransport` to remove itself from the map in
`FakeXdsTransportFactory` when it gets orphaned by the `XdsClient`, so
that a subsequent creation of a new transport for the same server does
not trigger an assertion due to the transport already existing in the
map.
Fixes internal b/259362837.
Was leading to a nullptr deref, and we just don't need this one anymore.
<!--
If you know who should review your pull request, please assign it to
that
person, otherwise the pull request would get assigned randomly.
If your pull request is for a specific language, please add the
appropriate
lang label.
-->
Before this change, `Found subchannel in state READY` and `Channel to
xds:///psm-grpc-server:61404 transitioned to state ` would dump the full
channel/subchannel, in some implementations that expose
ChannelData.trace (f.e. go) would add 300 extra lines of log.
Now we print a brief repr-like chanel/subchannel info:
```
Found subchannel in state READY: <Subchannel subchannel_id=9 target=10.110.1.44:8080 state=READY>
Channel to xds:///psm-grpc-server:61404 transitioned to state READY: <Channel channel_id=2 target=xds:///psm-grpc-server:61404 state=READY>
```
Also while waiting for the channel, we log channel_id now too:
```
Waiting to report a READY channel to xds:///psm-grpc-server:61404
Server channel: <Channel channel_id=2 target=xds:///psm-grpc-server:61404 state=TRANSIENT_FAILURE>
Server channel: <Channel channel_id=2 target=xds:///psm-grpc-server:61404 state=TRANSIENT_FAILURE>
Server channel: <Channel channel_id=2 target=xds:///psm-grpc-server:61404 state=TRANSIENT_FAILURE>
Server channel: <Channel channel_id=2 target=xds:///psm-grpc-server:61404 state=TRANSIENT_FAILURE>
Server channel: <Channel channel_id=2 target=xds:///psm-grpc-server:61404 state=TRANSIENT_FAILURE>
Server channel: <Channel channel_id=2 target=xds:///psm-grpc-server:61404 state=READY>
```
With these, it is actually possible to have typed client stubs where the
return type is correctly inferred.
It's only for the non-streaming calls, because there is
`RequestIterableType` for the streaming ones (but it's just Any with
extra steps and would require much more work).
---------
Co-authored-by: Xuan Wang <xuanwn@google.com>
<!--
If you know who should review your pull request, please assign it to
that
person, otherwise the pull request would get assigned randomly.
If your pull request is for a specific language, please add the
appropriate
lang label.
-->
---------
Co-authored-by: Yash Tibrewal <yashkt@google.com>
Co-authored-by: Stanley Cheung <stanleycheung@google.com>
Co-authored-by: AJ Heller <hork@google.com>
Co-authored-by: Yijie Ma <yijiem.main@gmail.com>
Co-authored-by: apolcyn <apolcyn@google.com>
Co-authored-by: Jan Tattermusch <jtattermusch@google.com>
In collaboration with @Vignesh2208 . This supersedes
https://github.com/grpc/grpc/pull/32622. The original description is
below.
----
The current endpoint semantics are as follows:
On endpoint shutdown, the socket is immediately closed regardless and
any pending registered read/write closures are not immediately executed.
The pending read/write closures get executed with aborted error whenever
the next grpc_iocp_work method runs.
However the grpc_iocp_work may run only during grpc_shutdown and
grpc_shutdown may only get scheduled after the pending registered
read/write closures execute.
This PR changes the shutdown semantics to match shutdown semantics used
in posix - i.e On endpoint shutdown, the socket is immediately closed
and any pending registered read/write closures are executed immediately.
Additional care is taken to ensure that the socket is not immediately
deleted because the pending I/O ops still need to be flushed later
during grpc_shutdown.
---------
Co-authored-by: Vignesh Babu <vigneshbabu@google.com>
@sampajano This should fix b/265779666. If the CFEngine ends up taking
some time to land with rollbacks, bugs and whatnot, this can work in the
meantime.
Similar to what we already do in other test suites:
- Try cleaning up resources three times.
- If unsuccessful, don't fail the test and just log the error. The
cleanup script should be the one to deal with this.
ref b/282081851
Avoids long path name problems on Windows
<!--
If you know who should review your pull request, please assign it to
that
person, otherwise the pull request would get assigned randomly.
If your pull request is for a specific language, please add the
appropriate
lang label.
-->
Fixes https://github.com/grpc/grpc/issues/32481.
Please test this with the (excellent) repro case in
https://github.com/grpc/grpc/pull/33000, and consider merging _just_ the
test from that PR.
Per #32481, the issue was bisected to
https://github.com/grpc/grpc/pull/30101. What changed in that PR is that
the epoll1 engine is only checked for availablily once per process at
iomgr initialization (which as a side effect initializes the engine),
but the engine was being shutdown with `grpc_shutdown` anyhow. With
repeated cycles of grpc init & shutdown in the same process, the second
attempt to reinit and use gRPC finds the epoll1 engine in an invalid
state.
Reverts grpc/grpc#32909
It's breaking some internal test
(//net/grpc/python:internal_tests/unit/_default_reflection_test), revert
for now to investigate.
Add a new binary that runs all core end2end tests in fuzzing mode.
In this mode FuzzingEventEngine is substituted for the default event
engine. This means that time is simulated, as is IO. The FEE gets
control of callback delays also.
In our tests the `Step()` function becomes, instead of a single call to
`completion_queue_next`, a series of calls to that function and
`FuzzingEventEngine::Tick`, driving forward the event loop until
progress can be made.
PR guide:
---
**New binaries**
`core_end2end_test_fuzzer` - the new fuzzer itself
`seed_end2end_corpus` - a tool that produces an interesting seed corpus
**Config changes for safe fuzzing**
The implementation tries to use the config fuzzing work we've previously
deployed in api_fuzzer to fuzz across experiments. Since some
experiments are far too experimental to be safe in such fuzzing (and
this will always be the case):
- a new flag is added to experiments to opt-out of this fuzzing
- a new hook is added to the config system to allow variables to
re-write their inputs before setting them during the fuzz
**Event manager/IO changes**
Changes are made to the event engine shims so that tcp_server_posix can
run with a non-FD carrying EventEngine. These are in my mind a bit
clunky, but they work and they're in code that we expect to delete in
the medium term, so I think overall the approach is good.
**Changes to time**
A small tweak is made to fix a bug initializing time for fuzzers in
time.cc - we were previously failing to initialize
`g_process_epoch_cycles`
**Changes to `Crash`**
A version that prints to stdio is added so that we can reliably print a
crash from the fuzzer.
**Changes to CqVerifier**
Hooks are added to allow the top level loop to hook the verification
functions with a function that steps time between CQ polls.
**Changes to end2end fixtures**
State machinery moves from the fixture to the test infra, to keep the
customizations for fuzzing or not in one place. This means that fixtures
are now just client/server factories, which is overall nice.
It did necessitate moving some bespoke machinery into
h2_ssl_cert_test.cc - this file is beginning to be problematic in
borrowing parts but not all of the e2e test machinery. Some future PR
needs to solve this.
A cq arg is added to the Make functions since the cq is now owned by the
test and not the fixture.
**Changes to test registration**
`TEST_P` is replaced by `CORE_END2END_TEST` and our own test registry is
used as a first depot for test information.
The gtest version of these tests: queries that registry to manually
register tests with gtest. This ultimately changes the name of our tests
again (I think for the last time) - the new names are shorter and more
readable, so I don't count this as a regression.
The fuzzer version of these tests: constructs a database of fuzzable
tests that it can consult to look up a particular suite/test/config
combination specified by the fuzzer to fuzz against. This gives us a
single fuzzer that can test all 3k-ish fuzzing ready tests and cross
polinate configuration between them.
**Changes to test config**
The zero size registry stuff was causing some problems with the event
engine feature macros, so instead I've removed those and used GTEST_SKIP
in the problematic tests. I think that's the approach we move towards in
the future.
**Which tests are included**
Configs that are compatible - those that do not do fd manipulation
directly (these are incompatible with FuzzingEventEngine), and those
that do not join threads on their shutdown path (as these are
incompatible with our cq wait methodology). Each we can talk about in
the future - fd manipulation would be a significant expansion of
FuzzingEventEngine, and is probably not worth it, however many uses of
background threads now should probably evolve to be EventEngine::Run
calls in the future, and then would be trivially enabled in the fuzzers.
Some tests currently fail in the fuzzing environment, a
`SKIP_IF_FUZZING` macro is used for these few to disable them if in the
fuzzing environment. We'll burn these down in the future.
**Changes to fuzzing_event_engine**
Changes are made to time: an exponential sweep forward is used now -
this catches small time precision things early, but makes decade long
timers (we have them) able to be used right now. In the future we'll
just skip time forward to the next scheduled timer, but that approach
doesn't yet work due to legacy timer system interactions.
Changes to port assignment: we ensure that ports are legal numbers
before assigning them via `grpc_pick_port_or_die`.
A race condition between time checking and io is fixed.
---------
Co-authored-by: ctiller <ctiller@users.noreply.github.com>
Need to do the channelz bit prior to the finishing the op bit.
<!--
If you know who should review your pull request, please assign it to
that
person, otherwise the pull request would get assigned randomly.
If your pull request is for a specific language, please add the
appropriate
lang label.
-->
Resolve `TESTING_VERSION` to `dev-VERSION` when the job is initiated by
a user, and not the CI. Override this behavior with setting
`FORCE_TESTING_VERSION`.
This solves the problem with the manual job runs executed against a WIP
branch (f.e. a PR) overriding the tag of the CI-built image we use for
daily testing.
The `dev` and `dev-VERSION` "magic" values supported by the
`--testing_version` flag:
- `dev` and `dev-master` and treated as `master`: all
`config.version_gte` checks resolve to `True`.
- `dev-VERSION` is treated as `VERSION`: `dev-v1.55.x` is treated as
simply `v1.55.x`. We do this so that when manually running jobs for old
branches the feature skip check still works, and unsupported tests are
skipped.
This changes will take care of all langs/branches, no backports needed.
ref b/256845629
Currently, we are not very consistent in what we assume the initial
state of an LB policy will be and whether or not we assume that it will
immediately report a new picker when it gets its initial address update;
different parts of our code make different assumptions. This PR
establishes the convention that LB policies will be assumed to start in
state CONNECTING and will *not* be assumed to report a new picker
immediately upon getting their initial address update, and we now assume
that convention everywhere consistently.
This is a preparatory step for changing policies like round_robin to
delegate to pick_first, which I'm working on in #32692. As part of that
change, we need pick_first to not report a connectivity state until it
actually sees the connectivity state of the underlying subchannels, so
that round_robin knows when to swap over to a new child list without
reintroducing the problem fixed in #31939.
AJ Heller
Xuan Wang
Yijie Ma
Craig Tiller
Vignesh Babu
Sergii Tkachenko
Esun Kim
Mark D. Roth
Vojtěch Boček
Yingwei Fan
Alisha Nanda
Doug Fawley
Thomas Zeugner
Yash Tibrewal
Hannah Shi
molto
Benjamin Peterson