With the completely different constructor it was hard to track which
fields were different during the test and reduced confidence. Now the
test code flows are much closer to the real-life code flows.
There was an attempt to use different epochs for the wall clock and the
monotonic clock. However, 123456789 is actually less than a second.
We want the gap between clocks to be at least a day. This issue was
discovered in #8968.
This separation found a bug in an RLS test where it was mixing epochs.
However, it was only a problem in the test. The code under test is
wrongly using wall clock for calculation durations, but that seems to be
a wide-spread problem and will need to be handled separately.
1. Unnecessary fully qualified names
Currently in XdsCredentialsRegistry, the child classes are referred by
their fully qualified names i.e.
'io.grpc.xds.internal.GoogleDefaultXdsCredentialsProvider' instead of
importing GoogleDefaultXdsCredentialsProvider and just using
GoogleDefaultXdsCredentialsProvider.class.
2. Use immutable interfaces instead of the generic collection interface
i.e. ImmutableMap instead of just Map.
These improvements are related to #8924.
Just using 100ms is mostly sufficient, but could potentially still flake
on the start()s that should return successfully. Waiting for the
XdsServingStatusListener to be called greatly reduces the amount of
processing needing to be done for start() to react and thus should
greatly avoid flakiness.
It is only safe to shutdown the executor if onClose() is called. Normal
RPC completions (both success and failure) will see the future complete,
so this only avoids shutting down in stranger cases like an interceptor
throwing an exception.
This fixes up cda0e9d to be compatible with Proguard without
configuration. Since the methods are now accessed directly there is no
need for manual -keep configuration.
* okhttp: forked required files to make okhttp dep compile only
* okhttp: forked missing file to make okhttp dep compile only
* okhttp: moved url and request files to proxy packge
* okhttp: removed unused methods from forked files; fixed build
This would limit ADS stream creation to one per second, even if the
old stream was considered good as it received a response. This shouldn't
really ever trigger, and if it does 1 QPS is probably still too high.
But 1 QPS is _substantially_ better than a closed loop and there's very
few additional signals we could use to avoid resetting the backoff.
b/224833499
protoc-gen-grpc-java isn't useful in a BOM because the Gradle plugin
doesn't support BOMs (google/protobuf-gradle-plugin#412) and it seems
unlikely the Maven plugin would as well.
Right now the type is pom, which is useless; nobody cares about the pom
itself. We'd need to use a type of exe and repeat it for each platform
(classifier). Given that it is a strange case and we can't actually test
the results and having it in the bom might lead a user to trying to make
it work, let's just remove it for now until it provides value.
Fixes#8936
Currently the credentials used for xDS communications is hardcoded in the BootstrapperImpl. The bootstrap config chooses one of the possible hardcoded credential.
This commit adds support for a credential plugin which allows users to register custom credentials through XdsCredentialProviders. gRPC will automatically discover the implementations via Java's SPI mechanism
Bootstrapper will use XdsCredentialRegistry to retrieve the list of supported credentials. The current hardcoded list of credentials(google_default, insecure and tls) are registered by default to keep the behavior as is.
Doing any reflection on OkHttpChannelBuilder requires that all methods
can have their arguments resolved. We'd like to make okhttp an optional
dependency (to support okhttp 2 and 3/4 simultaneously). But making
okhttp optional means we can no longer construct OkHttpChannelBuilder
reflectively. We swap to the Provider that doesn't have this problem.
See #8971.
Note that ManagedChannelProvider itself only exposes its methods as
protected, so they wouldn't be accessible. However OkHttpChannelProvider
has its methods public. It is an open question of whether
ManagedChannelProvider's methods should become public, but in any case
we can hide a public OkHttpChannelProvider inside a package-private
class so it is only accessable via reflection. So this code assuming
public methods doesn't prevent future implementation hiding.
We want to ignore the route in these situations, which is achieved by returning a null. The current behavior of returning an error triggers a NACK to the update.
We shouldn't require addresses to be non-empty for the child lb of rls_lb. That might be a right requirement when the child lb is grpclb, but in our new usecase, the child lb will be cds lb and will only work if empty address is allowed.
Fixing b/223866089#comment24
The previous code assumed that only gRPC would be using these methods.
But twice now Netty has made a change (generally relating to security)
that used a method for pseudo headers that previously wasn't supported.
Let's stop the whack-a-mole and just implement them all.
This restores compatibility with Netty 4.1.75.Final. Fixes#8981
* observability: integrate globalTags and configuration into rest of observability
wire observabilityConfig and globalTags into Observability and make
these available to the channel and server interceptors and specifically
to the LogHelper. Also separate globalTags into custom-tags and location-tags as required by the log-helper
This test suite is only fully enabled for C++ and Python at this
moment. This commit enables the rest of the test suite for grpc-java.
* change_backend_service_test
* failover_test
* remove_neg_test
* round_robin_test
* affinity_test
We are solving the issue of grpc/grpc delaying the release, causing GCE
tests to fail. Updating the master branch prevents similar cases from
happening.
Tested: [prod:grpc/java/master/branch/xds_v3](http://sponge/010f5353-e65e-4b1a-b6d3-f5a84e31546b)
If protoc is compiled with MinGW it will depend on libstdc++-6.dll which is found in C:\msys64\mingw64\bin and can only be found by inheriting PATH from the environment.
`setCall()` returns drainPendingCalls runnable only when there are calls to drain, otherwise return null. Preserved the behaviour of `start()` and `cancel()`, as they are protected by `delayOrExecute()`.
yifeizhuang
Eric Anderson
Mohan Li
Anirudh Ramachandra
sanjaypujare
Kurt Alfred Kluever
DNVindhya
beatrausch
Carl Mastrangelo
Anirudh Ramachandra
Terry Wilson
ZHANG Dapeng
Lidi Zheng
apolcyn
John Cormie
Jesse Schalken
Zhouyihai Ding