This Addresses the issue with skips not working in golang tests, ref b/235688697.
1. Unifies `TESTING_VERSION` detection in grpc_xds_k8s_install_test_driver.sh - new approach applicable to all languages.
2. Use `TESTING_VERSION` in all build files in `--testing_version` and when tagging docker images. This will be backported to all active test branches. Build Scripts in all other languages will be updated as well.
* Attempt to set correct platform on Mac OS
* Add some debug
* Make it fail
* Print more
* Try again
* Maybe it's an ordering issue?
* Get logs back
* Try copying distutils to see exactly what is being used
* Actually export the variable
* I just love debugging with CI
* One directory higher this time
* Try with an upgraded Python install
* Fix version
* Rebreak
* Try setting it even earlier?
* Unbreak
* Try explicitly renaming the artifacts
* Fix
* I am about ready to start NAT hole punching for SSH
* Break things for logs
* Whoops
* Clean up
* Shellcheck
* unit tests with bazel
* passing via --test_env from bazel command line
* remove env from BUILD; fix sanity check in run_one_test_bazel.sh
* add port server
This change includes:
* adding a cert file path for MacOS
* updating related test to run on MacOS too
* s/linux/supported/ since this now includes more platforms
* regenerating files affected by the name change
* [resource_quota] Periodic update tracker
Periodic update type that tries to do updates at some measurable
timescale without polling the current time every tick, instead trying to
approximate how many ticks will be required to fill the desired
duration.
The intent here is that we use this in RQ to donate memory back
periodically whilst keeping timer checks off the fast path.
* fix
* Automated change: Fix sanity tests
* review feedback
* fix
* fix
* review feedback
* Automated change: Fix sanity tests
Co-authored-by: ctiller <ctiller@users.noreply.github.com>
1. Fixes the issue with Java PSM security tests accidentally skipped because Java was missing from the list of languages, ref https://github.com/grpc/grpc/pull/28978
2. Invert the logic of `is_supported` methods, making them normally open
3. Make languages an `enum.Flag` to avoid accidental typos when listing the languages
4. Rename `XdsKubernetesTestCase.isSupported` to `XdsKubernetesTestCase.is_supported` to be consistent with `XdsUrlMapTestCase.is_supported`
5. Add extra logging
Split `XdsKubernetesTestCase` into:
- `XdsKubernetesTestCase` top-level base class containing flag parsing logic and common `assert*` methods
- `XdsKubernetesIsolatedTestCase` extending `XdsKubernetesTestCase`, that is specific to tests that want to create ifra resources before each test, and destroy them after.
Now tests that don't need to create/destroy all resources on each run, can extend `XdsKubernetesTestCase` without having to override all setUp and implementing other unnecessary methods.
* [build] Make compile_commands.json use NDEBUG
Ensures that things like clangd report the correct size information for
structures, which makes optimization a lot easier.
* remove -DNDEBUG from iwyus view
* fix
* [fuzzer] Add a script to sample fuzzers
* remember the script
* add ci
* bleh
* fix
* Update sample_fuzzers.sh
* tweak
* tweak
* tweak
* tweak
* tweak
* fix fuzzer found bug
* add explainer
* make it bold af
* limit max fuzzing time in addition to runs
This commit adds experimental CI for PSM tests.
Test included initially are Java and cxx tests. With expected load
from 100 to 30000, same as the manual test. Running proxied and proxyless
tests on benchmark-prod2 cluster, 8-core-machines (both client and server).
Data are uploaded to experimental tables.
Open for initial feedback on what we want to run on CI.
* [build] Better dependency fixer
Separate metrics for dependency selection into three: avoidness (to
strongly steer things away based on tags), a selectable metric, and
score (as an unbiasing heuristic)
Add a way to select targets with the command line to help iterating
Three selectable metrics are implemented:
1. edit distance, so that we tend to select things that don't change the
BUILD file
2. list size, as a way of minimizing dependency lists
3. best overall heuristic score
(1) is useful to avoid changing things that folks have carefully
engineered
(2, 3) help in optimizing the graph
* Automated change: Fix sanity tests
* ???
* Update fix_build_deps.py
Co-authored-by: ctiller <ctiller@users.noreply.github.com>