-
Notifications
You must be signed in to change notification settings - Fork 81
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Update our Envoy dependency #101
Conversation
Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
- It caused x-version trouble. - gperftools reported leaks , seemingly unfixable - gcovr ran into trouble In short, probably more trouble then it's worth, though it would have been a nice alternative to the pipe IPC we rely on with this change. Signed-off-by: Otto van der Schaaf <[email protected]>
- Update CI image: 8246167b9d238797cbc6c03dccc9e3921c37617d - Update Envoy: bcc66c6b74c365d1d2834cfe15b847ae13be0eb6 Prelude to updating coverage testing with lcov Benchmark http client still needs to be fixed Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
As per Harvey's recommendation, use py_binary and refer to it as a tool in the cpp_tests to get the pip deps sorted. Also, clean up integration/BUILD Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
- sync the way we do coverage testing with Envoy upstream - sync .bazelrc with Envoy upstream Notes: disabled `LinearRateLimiterInvalidArgumentTest` as that somehow specifically failed with the new coverage testing. Visibility of this target is the single thing from barring this to work: https://github.com/envoyproxy/envoy/blob/master/bazel/envoy_test.bzl#L147 Signed-off-by: Otto van der Schaaf <[email protected]>
…e-dump Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
In envoyproxy/nighthawk#101 we update coverage to the new methodology. For that to work we need to be able to consume `_lib_internal_only` targets which aren't visible today. This proposes a change so we can make that happen. The naming is fairly clear that these targets shouldn't be used otherwise. Signed-off-by: Otto van der Schaaf <[email protected]>
In anticipation of envoyproxy/envoy#7749 it makes sense to further upgrade the Envoy dep, which hereby is done. Signed-off-by: Otto van der Schaaf <[email protected]>
In envoyproxy/nighthawk#101 we update coverage to the new methodology. For that to work we need to be able to consume _lib_internal_only targets which aren't visible externally today. This proposes a change so we can make that happen. The naming is fairly clear that these targets shouldn't be used otherwise. Risk Level: low Testing: N/A Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
- the updated envoy fixes a clang-tidy issue we run in to - had to introduce some shim code which reduces the coverage % Let's see if CI is happy with these changes included. Signed-off-by: Otto van der Schaaf <[email protected]>
…k into update-envoy-state-dump Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
Signed-off-by: Otto van der Schaaf <[email protected]>
@@ -0,0 +1,90 @@ | |||
#include <chrono> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks @oschaaf . How much of this can be split into separate PRs? I.e. there is a ton of churn, around Python test infra, SSL objects, coverage, general upgrade? Ideally our Envoy upgrades are lightweight and some of these changes, e.g. test infra, are one-offs. Could you split out some of this easily?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, this can be split up into multiple PR's easily; be we may have to temporarily allow a lower coverage threshold at some point. I'll carve off a more fun-sized piece right away into a separate PR.
Split off from envoyproxy#101 - As a step towards upgrading our Envoy dependency, this swaps out Envoy's c++ integration test server with our own python wrappers around our own test server. This permanently avoids future runtime singleton conflicts. - Also starts setting `--base-id` when starting up our test-server. This is to support the upcoming test-sharding that will be performed by updated coverage test methodoly which will land in a future update. Signed-off-by: Otto van der Schaaf <[email protected]>
Closing this PR as all has been either split out into separate PR's or been superseded. |
Notes: