| cherrypick #54563 (#54613) Signed-off-by: Lonnie Liu <lonnie@anyscale.com> | 1 年前 |
| Updated stalebot to use unstale label instead of bounced. (#54506) the bounced label from the stalebot might be misleading to contributors - we are renaming it to unstale for better clarity. Signed-off-by: Christina Zhu <christina@anyscale.com> | 1 年前 |
| [Doc][KubeRay] verl example (#54114) Signed-off-by: Kai-Hsun Chen <kaihsun@anyscale.com> Signed-off-by: Kai-Hsun Chen <kaihsun@apache.org> Co-authored-by: Douglas Strodtman <douglas@anyscale.com> | 1 年前 |
| [Test] 调整测试用例,适应测试流水线 Co-authored-by: liuxianghai1<liuxianghai1@huawei.com> | 11 个月前 |
| 添加 TaskEvent 周期清理功能 Co-authored-by: liuxh25<liuxianghai1@huawei.com> | 6 个月前 |
| [core] Delete event_label (#54378) event_label is just macros mapping to strings that are only used in one spot. There really isn't a need for this, string literals are compile time constants anyways. Signed-off-by: dayshah <dhyey2019@gmail.com> | 1 年前 |
| 修改readme Co-authored-by: yangxiaopeng7<yangxiaopeng7@huawei.com> | 10 个月前 |
| ray-llm container cu124 -> cu128 update (#53730) upgrade to use cuda128 base images for both LLM CI test, release tests, and also for building ray-llm images. --------- Signed-off-by: Seiji Eicher <seiji@anyscale.com> | 1 年前 |
| [release] veresion change to 2.48.0 (#54557) Signed-off-by: Lonnie Liu <lonnie@anyscale.com> | 1 年前 |
| [Data] 通过环境变量控制ray data的map操作concurrency参数最小值为0 Co-authored-by: MoriHaru<xushengchun@huawei.com> | 6 个月前 |
| [Test] 添加性能测试代码 Co-authored-by: yangxiaopeng7<yangxiaopeng7@huawei.com> | 10 个月前 |
| 调整 xgrammar 版本,修复漏洞 Co-authored-by: liuxh25<liuxianghai1@huawei.com> | 10 个月前 |
| migrate check_library_usage_telemetry to _common (#54355) Fixes: https://github.com/ray-project/ray/issues/53478 - migrating check_library_usage_telemetry from _private to _common - migrating TelemetryCallsite from _private to _common. | 1 年前 |
| [ci] Remove scripts duplicates and symlinks except for format.sh (#34463) A year ago, #23866 moved our CI scripts into a more descriptive folder structure. Files in scripts/ were symlinks to the moved scripts. Even then, CI and documentation did not refer to any scripts in scripts/, with the exception of scripts/format.sh, which is referred to in pull request templates. Recently, #34340 overwrote some of the symlinks with their actual files. Since almost all of these scripts are only used in CI and not by users and developers, we should just get rid of the symlinks. The exception is format.sh which is actively used by developers. Signed-off-by: Kai Fricke <kai@anyscale.com> | 3 年前 |
| bug Signed-off-by: yangxiaopeng7 <yangxiaopeng7@huawei.com> | 4 个月前 |
| Revert "Revert "[core][telemetry/03] add OpenTelemetryMetricRecorder c++ client"" (#53422) Reverts ray-project/ray#53421. The original PR was broken on windows build, so it was reverted. This is the same PR but with the window fix. --------- Signed-off-by: can <can@anyscale.com> | 1 年前 |
| [ci] add workspace status script (#54398) and use it to reject '@' character in HOME and USER Signed-off-by: Lonnie Liu <lonnie@anyscale.com> | 1 年前 |
| [ci] upgrade bazel version to 6.5.0 (#45784) and pin it to 6.5.0; rejects all other bazel version. Signed-off-by: Lonnie Liu <lonnie@anyscale.com> | 2 年前 |
| [Lint] One parameter/argument per line for C++ code (#22725) It's really annoying to deal with parameter/argument conflicts. This is even frustrating when we merge code from the community to Ant's internal code base with hundreds of conflicts caused by parameters/arguments. In this PR, I updated the clang-format style to make parameters/arguments stay on different lines if they can't fit into a single line. There are several benefits: * Conflict resolving is easier. * Less potential human mistakes when resolving conflicts. * Git history and Git blame are more straightforward. * Better readability. * Align with the new Python format style. | 4 年前 |
| [core] Lint cpp files in core_worker (#48789) Running clang-tidy on files in core-worker, mostly auto-fixes from clang-tidy, some manual fixes. Want it to be part of ci or pre-commit hook once we lint the core src codebase. --------- Signed-off-by: dayshah <dhyey2019@gmail.com> Signed-off-by: Dhyey Shah <dhyey2019@gmail.com> | 1 年前 |
| Fix editorconfig option name (#50993) Signed-off-by: Neil Girdhar <mistersheik@gmail.com> | 1 年前 |
| [fossa] add fossa analyze (#52524) runs on postmerge only Signed-off-by: Lonnie Liu <lonnie@anyscale.com> | 1 年前 |
| Create .git-blame-ignore-revs for black formatting (#25118) | 4 年前 |
| [Core] Auto generate protobuf and grpc code (#40663) Revert back to the old way (before #39210) of auto-generating protobuf and grpc code so we don't need to manually generate and check them in. Also decouple the protobuf version we use in c++ and the protobuf version we use to auto generate python and java code (the protobuf version to auto generate python code affects the python protobuf library version that Ray can support) so that they can be upgraded independently. Signed-off-by: Jiajun Yao <jeromeyjj@gmail.com> | 2 年前 |
| Various improvements to Serve Request Batching tutorial (#50400) ## Why are these changes needed? * separated inference discussion to make article flow better * added discusion on named deployments * added recreating new serve handle from different processes * added example using curl directly to HTTP endpoint * fixed imports on serve API example * cleaned up massive block of outputs * Added prerequisites section to tutorial * Clarify that we are teaching two separate options for serving models * Fix orphaned ) ## Related issue number n/a ## Checks - [x] I've signed off every commit(by using the -s flag, i.e., git commit -s) in this PR. - [x] I've run scripts/format.sh to lint the changes in this PR. - [x] I've included any doc changes needed for https://docs.ray.io/en/master/. - [ ] I've added any new APIs to the API Reference. For example, if I added a method in Tune, I've added it in doc/source/tune/api/ under the corresponding .rst file. - [ ] I've made sure the tests are passing. Note that there might be a few flaky tests, see the recent failures at https://flakey-tests.ray.io/ - Testing Strategy - [x] Unit tests - [ ] Release tests - [ ] This PR is not tested :( --------- Signed-off-by: Ricardo Decal <rdecal@anyscale.com> Signed-off-by: Ricardo Decal <crypdick@users.noreply.github.com> Co-authored-by: Edward Oakes <ed.nmi.oakes@gmail.com> Co-authored-by: angelinalg <122562471+angelinalg@users.noreply.github.com> | 1 年前 |
| [core] Remove unused object_ref_seed parameter (#53698) Unset, unused. --------- Signed-off-by: Edward Oakes <ed.nmi.oakes@gmail.com> | 1 年前 |
| [Doc] Fix broken splash.html (#42029) This PR fixes broken HTML in splash.html: A missing </a> caused the page to break; this has been added. Part of the issue was that the formatting of the HTML was extremely hard to read; as part of the pre-commit hooks recently introduced, we now can run prettier on the code. This helped isolate the issue, and also formatted the file. Updated formatting rules to prevent prettier from splitting closing tags across lines Signed-off-by: pdmurray <peynmurray@gmail.com> | 2 年前 |
| [ci] upgrade rayci version to 0.16.0 (#54392) with some more features and goodies Signed-off-by: Lonnie Liu <lonnie@anyscale.com> | 1 年前 |
| [Doc] Add setuptools dependency for building docs (#46375) This PR: - Adds a _temporary_ setuptools>=70.0.0 dependency for building the docs to make sphinxcontrib-redoc usable on Python 3.12. This is not intended to be a long term solution, so #46374 has been made to track the follow-up work. - Remove unused environment variable in the buildkite job that builds docs - Bump the os version and python version used to build the docs on RTD Signed-off-by: pdmurray <peynmurray@gmail.com> | 2 年前 |
| [Doc] vale ignores anchors of headers (#53580) | 1 年前 |
| [core] Fix race condition b/w object eviction & repinning for recovery. (#53934) This PR has two components: ### Fixing the Race Condition b/w Object Eviction and Repining for Object Recovery The sequence of events documents the life of an object called A. 1. Owner sends object eviction notifications to all locations as part of Object A going out of scope. (done through worker<->worker pubsub) 2. Workers that have a copy of Object A mark it for asynchronous deletion. 3. Owner decides to recover the object and repin a copy of object A that has already been marked for asynchronous deletion. 4. Owner submits a task that has Object A as an argument. 5. The task executor tries to fetch Object A and the pull manager hangs forever. The fix is to check to see if an object is pending deletion before pinning it. If it is, the owner cannot pin the object and has to either try a different location or trigger reconstruction by resubmitting the task that created the object. ### (The Joys of) Writing a Unit Test in Node Manager There are a few problems with writing unit tests in the NodeManager 1. Despite heroics from @rueian and @dayshah recently, the NodeManager still had a few dependencies that needed to be put behind interfaces so fakes could be injected. 2. The gRPC handlers are all private to the class so they cannot be called directly. Instead the unit tests need to create a gRPC client. 3. The NodeManager has a lot of dependencies and the dependencies have interdependencies so we currently dedicate 200 lines to constructing a test fixture. 4. There is a lack of usable fakes e.g. for the PlasmaClient for writing unit tests. ### Follow Ups 1. Create a shim between the NodeManager and it's gRPC handlers so they can be unit tested directly without creating gRPC clients. 2. Create a utility function that lets you construct a NodeManager for testing optionally overriding interesting dependencies (otherwise they will default to a Noop fake). 3. Move the fakes created in this PR into their own (more complete implementations). 4. Create a Fake for SubscriberInterface. --------- Signed-off-by: irabbani <irabbani@anyscale.com> Signed-off-by: Ibrahim Rabbani <irabbani@anyscale.com> | 1 年前 |
| [README] 修改readme Co-authored-by: yangxiaopeng7<yangxiaopeng7@huawei.com> | 11 个月前 |
| [docs] Fix indefinite article in CONTRIBUTING rst (#50651) - add **a**~n~ reviewer ... <!-- Thank you for your contribution! Please review https://github.com/ray-project/ray/blob/master/CONTRIBUTING.rst before opening a pull request. --> <!-- Please add a reviewer to the assignee section when you create a PR. If you don't have the access to it, we will shortly find a reviewer and assign them to your PR. --> ## Why are these changes needed? see: - To assign **a** reviewer in https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/proposing-changes-to-your-work-with-pull-requests/requesting-a-pull-request-review ## Related issue number <!-- For example: "Closes #1234" --> ## Checks - [x] I've signed off every commit(by using the -s flag, i.e., git commit -s) in this PR. - [ ] I've run scripts/format.sh to lint the changes in this PR. - [ ] I've included any doc changes needed for https://docs.ray.io/en/master/. - [ ] I've added any new APIs to the API Reference. For example, if I added a method in Tune, I've added it in doc/source/tune/api/ under the corresponding .rst file. - [ ] I've made sure the tests are passing. Note that there might be a few flaky tests, see the recent failures at https://flakey-tests.ray.io/ - Testing Strategy - [ ] Unit tests - [ ] Release tests - [x] render the file CONTRIBUTING.rst Signed-off-by: Serge Croisé <SergeCroise@users.noreply.github.com> | 1 年前 |
| [Core] Vendor setproctitle (#53471) Signed-off-by: Jiajun Yao <jeromeyjj@gmail.com> | 1 年前 |
| [README] 修改readme Co-authored-by: yangxiaopeng7<yangxiaopeng7@huawei.com> | 11 个月前 |
| 修改readme Co-authored-by: yangxiaopeng7<yangxiaopeng7@huawei.com> | 9 个月前 |
| 修改readme Co-authored-by: yangxiaopeng7<yangxiaopeng7@huawei.com> | 9 个月前 |
| [Doc] Fix security doc link (#46352) closes #46350 Signed-off-by: Superskyyy <yihaochen@apache.org> | 2 年前 |
| [ci] adding uv binary v0.7.19 (#54437) Adding uv binary to be used in CI --------- Signed-off-by: elliot-barn <elliot.barnwell@anyscale.com> | 1 年前 |
| [image] update manual docker build script (#51990) moving python/requirements_compiled.txt to build context of base-deps. --------- Signed-off-by: Lonnie Liu <lonnie@anyscale.com> | 1 年前 |
| RLLIB and pylintrc (#8995) | 6 年前 |
| [CI] Re-enable isort for python/ray/tune (#52733) Follow-up for #52712 --------- Signed-off-by: Chi-Sheng Liu <chishengliu@chishengliu.com> Co-authored-by: Lonnie Liu <95255098+aslonnie@users.noreply.github.com> | 1 年前 |
| [Test] 调整测试用例,适应测试流水线 Co-authored-by: liuxianghai1<liuxianghai1@huawei.com> | 11 个月前 |
| [Train] Test Train code snippets (#40432) As a byproduct of the recent documentation rewrites, the Train docs contain several code snippets that aren't tested. This PR updates the snippets to test the ones that can be reasonably tested. --------- Signed-off-by: Balaji Veeramani <balaji@anyscale.com> | 2 年前 |
| [CI] Fix shellcheck warning in pre-commit (#50288) and enable it on CI --------- Signed-off-by: owenowenisme <mses010108@gmail.com> | 1 年前 |