Ideally we should plumb this through Filter, but FilterRegistry will
need to be plumbed to XdsClient and it started becoming non-trivial
compared to the "just add two lines." Expediency is helpful as the XDS
logs are pretty hard to read without the pretty-printing.
This fixes data race described in #8565.
We are doubtful whether checking closed in isReady() is necessary (#3201 might be a requirement), but it was easier to just maintain the existing behavior than think heavily about it.
The tests run as part of the existing android-interop-testing job.
We needed to modify the manifest of the apk built under android-interop-testing to declare Android Services used by the binder tests.
Although this is part of HTTP/2 and should have already been handled
already, it was noticed as part of RBAC work to avoid matching
hop-by-hop headers. See gRFC A41.
Also add a warning if creating Metadata.Key for "Connection". Use this
to try to help diagnose a client if it happens to blindly copy headers
from HTTP/1, as PROTOCOL_ERROR is hard to debug.
This rolls-forward 6e89919 after it was reverted in 7669656, now that
the test proxy has been fixed.
The sslContextProviderSupplier is used by the xds creds themselves when
the control plane has security configured. But the fallback credentials
don't use such a supplier and may not even be using TLS.
Language tweak following #8554.
Using ShadowProcess to set the processes uID doesn't help since SecurityPolicies class fetches the ID in a static initializer, and it may have already been loaded.
Instead, just rely on whatever the uID is already, and ensure the other UIDs we test with are offset from that first value.
The server list updates are very verbose and currently logged every second, causing a huge log spam if `ChannelLogger` is completely enabled. For debugging an internal issue, we need to turn on `ChannelLogger` but hide the server list updates from the logs to keep the log size reasonable.
Instead of `ChannelLogLevel.{DEBUG,INFO}` mapping to the same java level, `ChannelLogLevel.{WARNING,ERROR}` will shame the same java level. This allows us to be able to independently control the visibility of `ChannelLogLevel.DEBUG` logs which are the most verbose.
The internal build fails with "reference to assertThat is ambiguous". It
isn't clear why the internal build fails while the external one is okay,
but it is clear that the wildcard T return of readOutbound() is probably
confusing things as javac is considering assertThat(BigDecimal) as a
possible match.
The T return type is a hidden, convenience cast. We force the type
passed to assertThat() to be Object to avoid any ambiguity.
We want to know the single, unambiguous authority for the request. If
there is no authority, we use host instead. While authority would be
most typical for HTTP/2, requests proxied from HTTP/1 may use host
instead of authority.
This is generally useful, but the impetus is RBAC. See gRFC A41.
Since netty version v4.1.67, content-lenght header validation will be enforced. So once grpc upgrades netty to that version or above, RPCs with invalid content-length header will fail.
Some libraries such as HTTP to gRPC adapters blindly copy all HTTP headers to gRPC metadata, but the content-length header is one of those that shouldn't be forwarded because gRPC uses different encoding. This mistake has already been in existence for a long time.
Discard outbound content-length headers in gRPC, so that users who encounter invalid content-length issue when upgrading grpc-java version on server/client side would be able to workaround by upgrading grpc-java on client/server side as well without fixing the HTTP adapter.
Although this is part of HTTP/2 and should have already been handled
already, it was noticed as part of RBAC work to avoid matching
hop-by-hop headers. See gRFC A41.
Also add a warning if creating Metadata.Key for "Connection". Use this
to try to help diagnose a client if it happens to blindly copy headers
from HTTP/1, as PROTOCOL_ERROR is hard to debug.
Envoy RetryPolicy with empty retryOn should not be ignored as no retry config when selecting Route config. Therefore, if xDS update for a route contains a RetryPolicy that has no RetryOn value that we support, but the virtual host config does, xds client should choose the Envoy RetryPolicy from the route (even with no RetryOn), rather than choosing the one from virtual host, and try to convert it into grpc RetryPolicy, and end up with no retry.
setOnCancelHandler tells gRPC that the application is handling
cancellation. But it's fine to have noop behavior within the handler
itself if the application doesn't need it. It is just a way to opt-in to
the more recent no-exception-from-onNext behavior. Let's mention this
use-case in the docs to make it more obvious it is a possibility.
Came up as part of #8409.
Eric Anderson
markb74
ZHANG Dapeng
Liam Miller-Cushon
yifeizhuang
Lidi Zheng
Zhouyihai Ding
Eric Anderson
John Cormie
Jan Tattermusch
Piotr Morgwai Kotarbinski
ZhenLian
sanjaypujare
Terry Wilson
Benjamin Peterson