release: 1.16.0 by alyssawilk · Pull Request #13438 · envoyproxy/envoy

alyssawilk · 2020-10-08T12:54:30Z

Risk Level: n/a
Testing: n/a
Release Notes: yes

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

alyssawilk · 2020-10-08T12:54:46Z

@envoyproxy/maintainers please do a scan and see what if anything I missed.

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

asraa

Thank you!

asraa · 2020-10-08T13:39:32Z

docs/root/version_history/current.rst

-1.16.0 (Pending)
-================
+1.16.0 (October 8, 2020)
+========================


is the // intended in
build: added visibility rules for upstream. If these cause visibility related breakage, see notes in //BUILD.?

asraa · 2020-10-08T13:42:41Z

docs/root/version_history/current.rst

@@ -51,7 +50,7 @@ Minor Behavior Changes
 * router: now consumes all retry related headers to prevent them from being propagated to the upstream. This behavior may be reverted by setting runtime feature `envoy.reloadable_features.consume_all_retry_headers` to false.
 * stats: the fake symbol table implemention has been removed from the binary, and the option "--use-fake-symbol-table" is now a no-op with a warning.


render as :option:--use-fake-symbol-table (and --enable-fine-grain-logging) for consistency?

ah, turns out I had to revert this one - as a "not real flag" it's not documented, so flagging it as an option causes problems.

asraa · 2020-10-08T13:47:09Z

docs/root/version_history/current.rst

 * stats: the fake symbol table implemention has been removed from the binary, and the option "--use-fake-symbol-table" is now a no-op with a warning.
 * thrift_proxy: special characters {'\0', '\r', '\n'} will be stripped from thrift headers.
-* watchdog: replaced single watchdog with separate watchdog configuration for worker threads and for the main thread :ref:`Watchdogs<envoy_v3_api_field_config.bootstrap.v3.Bootstrap.watchdogs>`. It works with :ref:`watchdog<envoy_v3_api_field_config.bootstrap.v3.Bootstrap.watchdog>` by having the worker thread and main thread watchdogs have same config.
+* watchdog: replaced single watchdog with separate watchdog configuration for worker threads and for the main thread configured via :ref:`Watchdogs<envoy_v3_api_field_config.bootstrap.v3.Bootstrap.watchdogs>`. It works with :ref:`watchdog<envoy_v3_api_field_config.bootstrap.v3.Bootstrap.watchdog>` by having the worker thread and main thread watchdogs have same config.


sorry for not being able to comment exactly inline
but maybe truncate rocketmq_proxy network-level filter: to rocketmq_proxy:?

asraa · 2020-10-08T13:49:18Z

docs/root/version_history/current.rst

 * ext_authz filter: request timeout will now count from the time the check request is created, instead of when it becomes active. This makes sure that the timeout is enforced even if the ext_authz cluster's circuit breaker is engaged.
-  This behavior can be reverted by setting runtime feature ``envoy.reloadable_features.ext_authz_measure_timeout_on_check_created`` to false. When enabled, a new `ext_authz.timeout` stat is counted when timeout occurs. See :ref:`stats
-  // <config_http_filters_ext_authz_stats>`.
+  This behavior can be reverted by setting runtime feature ``envoy.reloadable_features.ext_authz_measure_timeout_on_check_created`` to false. When enabled, a new `ext_authz.timeout` stat is counted when timeout occurs. See :ref:`stats <config_http_filters_ext_authz_stats>`.


this and envoy.restart_features.use_apple_api_for_dns_lookups are the only features that have double back ticks. single for consistency?

asraa · 2020-10-08T13:50:52Z

docs/root/version_history/current.rst

 * http: added :ref:`CDN Loop filter <envoy_v3_api_msg_extensions.filters.http.cdn_loop.v3alpha.CdnLoopConfig>` and :ref:`documentation <config_http_filters_cdn_loop>`.
 * http: added :ref:`MaxStreamDuration proto <envoy_v3_api_msg_config.route.v3.RouteAction.MaxStreamDuration>` for configuring per-route downstream duration timeouts.
-* http: introduced new HTTP/1 and HTTP/2 codec implementations that will remove the use of exceptions for control flow due to high risk factors and instead use error statuses. The old behavior is used by default, but the new codecs can be enabled for testing by setting the runtime feature `envoy.reloadable_features.new_codec_behavior` to true. The new codecs will be in development for one month, and then enabled by default while the old codecs are deprecated.
+* http: introduced new HTTP/1 and HTTP/2 codec implementations that will remove the use of exceptions for control flow due to high risk factors and instead use error statuses. The old behavior is used by default for HTTP/1.1 and HTTP/2 server connections. The new codecs can be enabled for testing by setting the runtime feature `envoy.reloadable_features.new_codec_behavior` to true. The new codecs will be in development for one month, and then enabled by default while the old codecs are deprecated.


thanks! LGTM

dio · 2020-10-08T14:30:47Z

docs/root/version_history/current.rst

-* ext_authz: the :ref:`dynamic metadata <envoy_v3_api_field_service.auth.v3.OkHttpResponse.dynamic_metadata>` field in :ref:`OkHttpResponse <envoy_v3_api_msg_service.auth.v3.OkHttpResponse>`
-  has been deprecated in favor of :ref:`dynamic metadata <envoy_v3_api_field_service.auth.v3.CheckResponse.dynamic_metadata>` field in :ref:`CheckResponse <envoy_v3_api_msg_service.auth.v3.CheckResponse>`.


Thank you for sorting this.

no problem :-)
I've got a check_format fix queued for this - we had a few in the last release as well and it's an easy one to automate!

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

rojkov

Thank you for correcting building.rst!

rojkov · 2020-10-08T14:58:32Z

docs/root/install/building.rst


 Please see the linked :repo:`CI <ci/README.md>` and :repo:`Bazel <bazel/README.md>` documentation
 for more information on performing manual builds.
+Please note that for Clang/LLVM 8 and lower, Envoy must be built with `--define tcmalloc=gperftools`


You could also emphasize the fact that only x86 builds may need --define tcmalloc=gperftools. And since the new tcmalloc wants C++17 GCC 8+ may also suffice.

Updated. PTAL?

LGTM. Thanks!

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

mattklein123

LGTM with one comment! Thanks!

mattklein123 · 2020-10-08T15:21:21Z

docs/root/version_history/current.rst

 *Changes that are expected to cause an incompatibility if applicable; deployment changes are likely required*

-* build: added visibility rules for upstream. If these cause visibility related breakage, see notes in //BUILD.
+* build: added visibility rules for upstream. If these cause visibility related breakage, see notes in `BUILD <https://github.com/envoyproxy/envoy/blob/master/BUILD>`_.


Use :repo: link for this.

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

alyssawilk · 2020-10-08T17:10:39Z

RELEASES.md

 | 1.13.0  | 2019/12/31 | 2020/01/20 |  +20 days  | 2021/01/20  |
 | 1.14.0  | 2020/03/31 | 2020/04/08 |   +8 days  | 2021/04/08  |
 | 1.15.0  | 2020/06/30 | 2020/07/07 |   +7 days  | 2021/07/07  |
-| 1.16.0  | 2020/09/30 |            |            |             |


@envoyproxy/stable-maintainers do we have a release manager lined up for Q4?

release: 1.16.0

cdd1e32

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

codec

0dfb2b6

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

asraa reviewed Oct 8, 2020

View reviewed changes

dio reviewed Oct 8, 2020

View reviewed changes

alyssawilk added 3 commits October 8, 2020 10:41

Merge branch 'master' into notes

f50b4f9

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

review comments, plus clang9

d8c13ef

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

updating requirements

a8a9dc7

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

alyssawilk mentioned this pull request Oct 8, 2020

memory: switch to the latest tcmalloc for x86_64 builds #13251

Merged

rojkov reviewed Oct 8, 2020

View reviewed changes

comments

acc2336

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

mattklein123 self-assigned this Oct 8, 2020

mattklein123 reviewed Oct 8, 2020

View reviewed changes

mattklein123 added the waiting label Oct 8, 2020

comments

9c7e209

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

repokitteh-read-only bot removed the waiting label Oct 8, 2020

mattklein123 approved these changes Oct 8, 2020

View reviewed changes

mattklein123 merged commit 8fb3cb8 into envoyproxy:master Oct 8, 2020

alyssawilk commented Oct 8, 2020

View reviewed changes

cpakulski mentioned this pull request Dec 1, 2020

wasm: fix network leak #13836

Merged

alyssawilk deleted the notes branch June 10, 2021 13:43

		@@ -51,7 +50,7 @@ Minor Behavior Changes
		* router: now consumes all retry related headers to prevent them from being propagated to the upstream. This behavior may be reverted by setting runtime feature `envoy.reloadable_features.consume_all_retry_headers` to false.
		* stats: the fake symbol table implemention has been removed from the binary, and the option "--use-fake-symbol-table" is now a no-op with a warning.

		* ext_authz: the :ref:`dynamic metadata <envoy_v3_api_field_service.auth.v3.OkHttpResponse.dynamic_metadata>` field in :ref:`OkHttpResponse <envoy_v3_api_msg_service.auth.v3.OkHttpResponse>`
		has been deprecated in favor of :ref:`dynamic metadata <envoy_v3_api_field_service.auth.v3.CheckResponse.dynamic_metadata>` field in :ref:`CheckResponse <envoy_v3_api_msg_service.auth.v3.CheckResponse>`.

Conversation

alyssawilk commented Oct 8, 2020

Uh oh!

alyssawilk commented Oct 8, 2020

Uh oh!

asraa left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

rojkov left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

mattklein123 left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants