…#2345)

Signed-off-by: Matt Rice <mattrice@google.com>

Signed-off-by: Kyle Myerson <kmyerson@google.com>

Signed-off-by: Chris Roche <croche@lyft.com>

Signed-off-by: Louis Taylor <louis@kragniz.eu>

Risk Level: Low
Testing: Unit
Docs Changes: N/A
Release Notes: N/A

Signed-off-by: Matt Klein <mklein@lyft.com>

Implements Metrics Service in Envoy that continuously streams metrics to the gRPC end point configured.

Risk Level: Small-medium
Testing: unit test, integration test

Docs Changes:

Data Plane PR

Release Notes:

Release notes updated

API Changes:

Data Plane API

Signed-off-by: Rama <ramaraochavali@gmail.com>

Signed-off-by: Louis Taylor <louis@kragniz.eu>

…oyproxy#2372)

Signed-off-by: Joshua Marantz <jmarantz@google.com>

Signed-off-by: Dhi Aurrahman <dio@hooq.tv>

…xy#2378)

Signed-off-by: Piotr Sikora <piotrsikora@google.com>

Signed-off-by: Louis Taylor <louis@kragniz.eu>

Signed-off-by: Alexey Baranov <me@kotiki.cc>

Subpar uses Python, which doesn't like it when we don't have
PYTHONUSERBASE (hidden by Bazel hermeticity) set or getpwuid()
resolution to user (missing in /etc/passwd). This is a workaround based
on JonathonReinhart/scuba#11 to get build
happening on Linux when non-root.

Signed-off-by: Harvey Tuch <htuch@google.com>

Signed-off-by: Louis Taylor <louis@kragniz.eu>

In order to support a singleton gRPC manager (managing TLS state,
connection pools for the Google gRPC client, etc.), this PR removes the
use of templating in Grpc::AsyncClient and its related interfaces.

We initially used templates as they were an easy way to propagate type
information. However, we can avoid using them by taking advantage of the
Protobuf::Message superclass interface. There is some convenience in
using templating for the request/stream callbacks, adapter template
classes are provided to make this possible.

Signed-off-by: Harvey Tuch <htuch@google.com>

Signed-off-by: Matt Klein <mklein@lyft.com>

Moving Server::Listener to Network::ListenerConfig helps avoid
dependency from Network namespace to Server namespace in following
changes.

Requested-by: Matt Klein <mklein@lyft.com>
Signed-off-by: Jarno Rajahalme <jarno@covalent.io>

Signed-off-by: Matt Klein <mklein@lyft.com>

…oxy#2335)

Signed-off-by: Brian Pane <bpane@pinterest.com>

Congratulations! Maybe? ;)

Signed-off-by: Matt Klein <mklein@lyft.com>

…y#2363)

Make gRPC-JSON transcoder not buffering response at end_stream or non-gRPC response.

Risk Level: Low

Testing:
unit test

Docs Changes:
N/A

Release Notes:
N/A

Fixes envoyproxy#2275

Signed-off-by: Lizan Zhou <zlizan@google.com>

Signed-off-by: Dhi Aurrahman <dio@hooq.tv>

Signed-off-by: Gabriel <gsagula@gmail.com>

This is a prereq for running CI on envoyproxy#2348

Description:
Integrates github.com/google/benchmark so it can be used in future PRs for demonstrating improvement. Note this PR does not actually depend on it, so it can pass CI.

Risk Level: Low

Tested: //test/...

Signed-off-by: Joshua Marantz <jmarantz@google.com>

The current problem with lua is that it is unable to use external
libraries such as lrexlib installed via luarocks.

The solution is to export the symbols needed by those libraries.

Signed-off-by: Nicholas J <nicholas.a.johns5@gmail.com>

…nvoyproxy#2358)

Signed-off-by: Joshua Marantz <jmarantz@google.com>

When the structure of shared memory is
[100 options slots][control struct] and we try parse it as [99 options][control struct] we end up parsing the control struct from the uninitialized options memory. Fixed via memset.

Risk Level: n/a (test fix)
Testing: exactly.
Release Notes: n/a

Signed-off-by: Alyssa Wilk <alyssar@chromium.org>

Signed-off-by: Taiki Ono <taiks.4559@gmail.com>

Signed-off-by: Matt Klein <mklein@lyft.com>

Description:
The Envoy API accepts regular expressions for several configuration options. If an invalid regular expression is provided, STL throws std::regex_error, which is uncaught and causes the sudden termination of Envoy. Replace use of the std::regex(std::string) constructor with calls to a new utility RegexUtil::parseRegex() which converts the std::regex_error into an EnvoyException, allowing the invalid configuration to be handled gracefully.

Risk Level: Low

Testing:
Unit tests for behavior or new utility function and testing of invalid regular expressions in configuration.

Release Notes: N/A

Fixes: envoyproxy#2406

Signed-off-by: Stephan Zuercher stephan@turbinelabs.io

…y#2248)

File-based BoringSSL APIs are no longer used and everything is processed
as if it was delivered inline in order to provide consistent behavior.

Additionally, DataSource files are now read during configuration parsing,
which means that the configuration is considered invalid if it refers to
non-existing files.

Fixes envoyproxy#1357.

Signed-off-by: Piotr Sikora <piotrsikora@google.com>

…envoyproxy#2387)

Description:
Extend the health check filter to optionally compute its HTTP response status
based on whether at least a specified percentage of servers in a specified set
of upstream clusters are healthy.

Risk Level: Medium

Testing: Unit tests included

Docs Changes: envoyproxy#425

Release Notes: Included in this PR

Fixes: envoyproxy#2362

API Changes: envoyproxy#417

Signed-off-by: Brian Pane bpane@pinterest.com

…s. (envoyproxy#2427)

This PR is the next in the series of patches to integrate the Google gRPC client. There are two
parts:

1. The bulk of the existing Grpc::AsyncClientImpl unit tests have been re-expressed as TCP loopback
   integration tests with a FakeUpstream. This allows direct reuse with the Google gRPC client
   libraries, since they are a blackbox that can't be easily mocked internally. The test is also now
   paramaterized by client type, with the Google gRPC implementation elided in this PR.

2. To produce behavioral parity between clients, Grpc::AsyncClientImpl has been modified where its
   behavior differed from the Google gRPC client (which is effectively the gRPC reference
   implementation). Specifically, initial/trailer metadata callbacks, error handling and HTTP ->
   gRPC status mapping behavior have been modified. Remote stream closes, even with OK grpc-status
   now terminate the stream regardless of local stream close behavior.

Testing: Unit/integration tests as described above.
Risk Level: Medium (behavior changes to Grpc::AsyncClientImpl, used for config discovery, rate limiting, access logs and metrics service).

Signed-off-by: Harvey Tuch <htuch@google.com>

Adds a benchmarking for new string utility functions added in envoyproxy#2368 using the google microbenchmarking library (https://github.com/google/benchmark), which was added in envoyproxy#2350. This serves as an example and proof of concept on how to use the benchmarking library, and as a reference point for the approximate performance of some string-parsing functions.

Risk Level: Very Low

Testing:
This is just a new speed-test, so running it on its own is sufficient to test it.

Results:

2018-01-18 08:01:44
Run on (12 X 3800 MHz CPU s)
CPU Caches:
  L1 Data 32K (x6)
  L1 Instruction 32K (x6)
  L2 Unified 256K (x6)
  L3 Unified 15360K (x1)
***WARNING*** CPU scaling is enabled, the benchmark real time measurements may be noisy and will incur extra overhead.
-------------------------------------------------------------------------------------
Benchmark                                              Time           CPU Iterations
-------------------------------------------------------------------------------------
BM_RTrimStringView                                    11 ns         11 ns   54467092
BM_RTrimStringViewAlreadyTrimmed                       9 ns          9 ns   77143046
BM_RTrimStringViewAlreadyTrimmedAndMakeString         24 ns         24 ns   29094696
BM_FindToken                                         165 ns        165 ns    4230849
BM_FindTokenValueNestedSplit                         427 ns        427 ns    1653627
BM_FindTokenValueSearchForEqual                      180 ns        180 ns    4046401

In practice I did not find this benchmark to be noisy on repeated runs.

Signed-off-by: Joshua Marantz <jmarantz@google.com>

it adds http_proxy env variables which can be used while building the image
user is expected to set http_proxy and https_proxy in the env before running the build command.

Risk Level: Low

Testing: tested in a proxy env

Signed-off-by: Sudeepto Roy <sudeepto.roy@gmail.com>

Waiting for first dispatch loop is too late for ADS use with Google gRPC client.

Testing: ADS integration test with Google gRPC client, existing tests.
Risk Level: Low

Signed-off-by: Harvey Tuch <htuch@google.com>

When a subscription is for the empty set, rather than a specific set of services, avoid sending an
empty request when the watch ends.

Signed-off-by: Harvey Tuch <htuch@google.com>

…#2434)

Testing: New paramaterized integration tests.
Risk Level: Low

Signed-off-by: Harvey Tuch <htuch@google.com>

Network::ListenerImpl class implements an libevent socket
listener. Factor out all functionality that is not directly libevent
related to ConnectionHandlerImpl, which both reduces overall
complexity and makes it possible to refactor configurable listener
behavior to a new class of listener filters in the following patches.

This patch also reduces the special handling for SSL connections by
introducing a new dispatcher member for creating server
connections. This is used by the refactored code as the connection
creation is now shifted from Network::Listener to the connection
handler.

To further simplify the resulting code, this patch removes
Network::ListenerOptions class, as the only option that now needs to
be passed to Network::Listener is the "bind to port" boolean, which is
now passed in explicitly.  All other listener options are handled by
passing along Server::Listener instead.

Signed-off-by: Jarno Rajahalme <jarno@covalent.io>

Signed-off-by: Jarno Rajahalme <jarno@covalent.io>

…r changes. (envoyproxy#2427)"

This reverts commit c680300.

. (envoyproxy#2443)

Signed-off-by: Harvey Tuch <htuch@google.com>

One of the BoringSSL updates we've picked up in the last few months fixed the need for extra flags when compiling BoringSSL in Xcode 9.

Risk Level: Low
Testing: existing CI is sufficient
Release Notes: N/A

Signed-off-by: Stephan Zuercher <stephan@turbinelabs.io>

Signed-off-by: Piotr Sikora <piotrsikora@google.com>

…xy#2447)

UBSAN doesn't show stack traces by default, so this diff adds the environment variable
needed to enable stack traces.

Risk Level: Low

Testing: The CircleCI ASAN run will exercise this code change.

Docs Changes: N/A

Release Notes: N/A

Signed-off-by: Brian Pane <bpane@pinterest.com>

…envoyproxy#2448)

This PR fixes envoyproxy#2379 (Linux) by passing --non-unique to docker command. This option allows creating duplicate users.

Risk Level: Low

Testing: manual OsX, Linux

Signed-off-by: Gabriel <gsagula@gmail.com>

Signed-off-by: Matt Rice <mattrice@google.com>