Introduce the concept of locality weighted LB (as distinct from zone
aware LB) in the docs and a new field in Cluster, locality_weighted_lb,
for configuring this behavior.
Signed-off-by: Harvey Tuch <htuch@google.com>
1) Add ability to runtime filter to configure default, divisor, and
whether independent randomness is used.
2) Also add LE to the comparison filter.
Signed-off-by: Matt Klein <mklein@lyft.com>
This enables generating generic service stubs for all the data-plane-api
proto services when generating Java classes with protoc.
This is generally not needed when implementing a gRPC server but in our case we're implementing
it behind our legacy protobuf RPC framework which rely on these stubs. As far as I know the only negative
with enabling these is generating some potentially unnecessary Java classes.
Signed-off-by: Snow Pettersen <snowp@squareup.com>
This adds ability to configure settings for zone aware load balancing settings as config options.
solves envoyproxy/envoy#1344
Changes : Added min_cluster_size and enabled_percentage to commonLbConfig which enables them to used directly in load_balancer_impl.cc.
Signed-off-by: sri kailash <sri.gandebathula@booking.com>
There are several main changes in this PR:
Create envoy.api.v2.core packages to break circular dependencies from xDS on to subpackages on to base protos.
Create individual packages for each filter and add independent versioning to each filter.
Add visibility constraints to prevent formation of dependency cycles.
Add gogoproto annotations to improve go code generation.
After moving xDS service definitions and top-level resource protos back to envoy.core.api.v2, cycles were created, since the second-level definitions depend on base protobuf definitions, and are in turn included from xDS; however xDS and base definitions are in the same package.
The solution is to split the base protos into another package, envoy.api.v2.core. That eliminates dependency cycles (validated using go-control-plane).
Added a few gogoproto annotations to improve golang code generation.
Signed-off-by: Kuat Yessenov <kuat@google.com>
Currently clusters can not open both HTTP1.1 and HTTP2 upstream
connections at the same time. When the new cluster option
"auto_http2" is set to "true", the cluster must open an HTTP2 upstream
connection if the downstream connection is HTTP2, and an HTTP1.1
upstream connection if the downstream connection is HTTP1.1. This option
is to have no effect if there is no corresponding downstream
connection.
This functionality removes the need to operate multiple clusters and
routing rules for them when the backends accept both HTTP1.1 and HTTP2
connections, and when the choice of the HTTP protocol is significant,
as with gRPC.
Signed-off-by: Jarno Rajahalme <jarno@covalent.io>
In support of https://github.com/envoyproxy/envoy/issues/2200 and some
Google internal needs, we are planning on adding support to Envoy to
allow a configuration (or possibly build) driven decision on whether to
using the existing Envoy in-built Grpc::AsyncClient or
the Google C++ gRPC client library (https://grpc.io/grpc/cpp/index.html).
To move in this direction, the idea is we have the xDS ApiConfigSources,
rate limit service config and other filter configurations point at a
GrpcService object. This can be configured to use an Envoy cluster,
where Grpc::AsyncClient will orchestrate communication, or to contain
the config needed to establish a channel in Google C++ gRPC client
library.
Signed-off-by: Harvey Tuch <htuch@google.com>
CertificateValidationContext.trusted_ca is not only for client
certificates, but also for server certs. Change the wording to "peer
certificates".
Also mention that verification is not enabled by default in docs for
UpstreamTlsContext.
Signed-off-by: Peter Schultz <peter.schultz@classmarkets.com>
These were lost in my backlog. Required a PGV fix for cross-package enum
validation to deal with the TODOs, see
https://github.com/lyft/protoc-gen-validate/issues/42.
Signed-off-by: Harvey Tuch <htuch@google.com>
This is code movement only and no other changes. There should be
no namespace changes or effects to consumers.
Signed-off-by: Matt Klein <mklein@lyft.com>
* Added PGV C++ generation support. This (hopefully temporarily)
abandons using native proto_library in favor of pgv_cc_proto_library.
We maintain build support for proto_library for the glorious future in
which we write a Bazel aspect to run PGV against the native
proto_library shadow graph.
* Replace min_len with min_bytes on strings, until PGV gets not-empty or
min_len support for C++.
* Various fixups for places where the PGV plugin objected to
annotations.
Signed-off-by: Harvey Tuch <htuch@google.com>
We'd like to add a metadata field to CDS and have it be accessible in Envoy through ClusterInfo and anywhere else general cluster info is available.
Signed-off-by: Matt Rice <mattrice@google.com>
1) Do RDS other than rate limit. This is large enough will do that next.
2) Manually order top level in v2 with better titles
3) Random other fixes
Note also that I'm going to manually reorder the protos in some cases in a
follow up so the docs make a bit more sense for a casual reader.
Signed-off-by: Matt Klein <mklein@lyft.com>
This patch adds an overview page introduced the v2 API concepts via a
worked example. Brought in the entire transitive dep set of protos from
bootstrap.proto, none of these have been cleaned up beyond the minimum
required to have them build under Sphinx.
Also added the ability to link to the underlying proto in messages/enums
from protodoc.py generated RST.
Signed-off-by: Harvey Tuch <htuch@google.com>
Updates the comment to reflect the design decision made in envoyproxy/envoy#1735 and renames subset_keys to subset_selectors.
My reasoning for the rename is that the field is all about determining how endpoints are grouped into subsets. Using the names of metadata keys in the endpoints is only one way and others might be added.
Signed-off-by: Stephan Zuercher <stephan@turbinelabs.io>
Provides a configuration point for a load balancer that can divide endpoints into subsets addressable by route metadata.
See discussion in envoyproxy/envoy#1279.
Signed-off-by: Stephan Zuercher <stephan@turbinelabs.io>
Turns out that files with just service methods don't get loaded into the
descriptor pool automatically in C++. So, needed to have some messages
in ads.proto. Turns out this was a good opportunity to move some of the
messages that were related to discovery out of base.proto.
This shouldn't break the API, since everything is in the envoy.api.v2
packge space.
Since we may want to use a resolver plugin in most places, it doesn't
make sense to have this strong separation between the two in the API.
Also bonus renumbering cleanup in LDS.
Add code points for original dst service discovery and load balancer
types, as well as for the cleanup interval configuration option.
Signed-off-by: Jarno Rajahalme <jarno@covalent.io>
At Google, we need to support cluster queries that contain ':', so this
seems a useful feature to retain in v2 as a means to do that (rather
than trying to retrofit this to the internal cluster naming in Envoy).
htuch
Matt Klein
Rama Chavali
Snow Pettersen
Daniel Hochman
srikailash
Kuat
Matt Rice
jrajahalme
Peter Schultz
alyssawilk
Jose Ulises Nino Rivera
Jonathan Oddy
Lizan Zhou
Piotr Sikora
Stephan Zuercher