refactor: split pd/server/store image stage & bump to V2 version#13
refactor: split pd/server/store image stage & bump to V2 version#13
Conversation
- replace gate_mode choice with strict_mode boolean input (default true) - run integration precheck only when strict_mode is enabled - keep matrix publish path available when precheck is skipped - preserve per-module self-check and multi-arch push behavior
|
Note Gemini is unable to generate a summary for this pull request due to the file types involved not being currently supported. |
dosubot
bot
added
the
size:L
|
@codecov-ai-reviewer review |
|
Note Reviews pausedIt looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the Use the following commands to manage reviews:
Use the checkboxes below for quick actions:
Walkthrough新增可选严格模式与并行控制；拆分为 Changes
Sequence Diagram(s)sequenceDiagram
actor GitHub as GitHub Actions
participant Dispatch as workflow_dispatch
participant Resolve as resolve_source
participant Precheck as integration_precheck
participant BuilderLocal as buildx_local_x86
participant Compose as DockerCompose
participant Services as PD/Store/Server
participant Health as HealthEndpoints
participant Matrix as publish_matrix
participant SelfCheck as SelfCheckContainer
participant QEMU as setup-qemu
participant Push as build-push-action
GitHub->>Dispatch: 手动触发 (inputs: strict_mode, mvn_args, wait_timeout_sec)
Dispatch->>Resolve: 解析 upstream -> SOURCE_SHA
alt strict_mode == true
Dispatch->>Precheck: 运行 integration_precheck (checkout SOURCE_SHA)
Precheck->>BuilderLocal: 本地构建 x86 镜像 (load: true, module cache scopes)
Precheck->>Compose: 生成 override & docker compose up (pull_policy: never)
Compose->>Services: 启动服务
Precheck->>Health: 轮询 /v1/health 与 /versions
Health-->>Precheck: 返回就绪
Precheck->>Compose: 导出日志(若失败) 并 teardown
Precheck-->>Matrix: 通过或跳过继续
else strict_mode == false
Dispatch-->>Matrix: 跳过预检直接继续
end
Matrix->>BuilderLocal: per-module 构建/加载 x86 自检镜像
Matrix->>SelfCheck: 启动模块自检容器并轮询 probe_path
SelfCheck->>Health: 轮询模块探针
Health-->>SelfCheck: 就绪
SelfCheck-->>Matrix: 移除自检容器
Matrix->>QEMU: 设置 QEMU 支持 multi-arch
Matrix->>Push: 构建并推送 multi-arch 镜像 (amd64, arm64) 使用模块 cache scopes
Push-->>Matrix: 推送完成并清理
Estimated code review effort🎯 4 (Complex) | ⏱️ ~45 minutes Poem
🚥 Pre-merge checks | ✅ 3✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
There was a problem hiding this comment.
Pull request overview
This PR updates the publish_latest_pd_store_server_image GitHub Actions workflow to introduce a strict_mode publish gate, making the integration precheck optional while still allowing the publish matrix path to run when the precheck is skipped.
Changes:
- Replaces the previous gating approach with a
strict_modeboolean workflow input (defaulttrue). - Adds an
integration_precheckjob that runs only whenstrict_modeis enabled. - Refactors publishing into a
publish_matrixjob that can proceed when precheck succeeds or is skipped, while preserving per-module self-check and multi-arch push behavior.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In @.github/workflows/publish_latest_pd_store_server_image.yml:
- Around line 29-34: 当前问题是 integration_precheck 和 publish_matrix 两个 job 各自直接
checkout master，导致 publish 可能用到在 precheck 之后的新提交；请修改 workflow：在
integration_precheck job（job 名称 integration_precheck）在 checkout 后将当前提交 SHA 导出为
job 输出（例如通过一个 step 读取 $GITHUB_SHA 或 steps.checkout.outputs.sha 并用 echo
"::set-output name=commit_sha::$SHA"），然后在 publish_matrix（以及其它类似的 publish
job/位置）将 actions/checkout 的 ref 改为使用该输出（ref: ${{
needs.integration_precheck.outputs.commit_sha }}）以保证所有 matrix 分片和后续 publish 步骤都
checkout 相同的 commit SHA（同样把文件中另一个位置 193-198 的 checkout 改为使用相同的输出）。
- Around line 160-165: 当前 workflow 在并行推送三个 moving tag 时缺少并发锁，导致不同 runs
会交错覆盖镜像（相关 job 为 publish_matrix）。在 workflow 顶层添加一个 concurrency
配置以序列化同名发布（例如使用固定组名或基于引用/输入的组串），并设置 cancel-in-progress: false（或根据需要为
true）以防止并行运行；确保该 concurrency 放在顶层（与 jobs 同级），以覆盖整个 workflow 而非单个 job，从而避免
publish_matrix 的并行覆盖问题。
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: d2837a33-8a3e-49d8-b131-67582b66844d
📒 Files selected for processing (1)
.github/workflows/publish_latest_pd_store_server_image.yml
Add concurrency group and a new resolve_source job that fetches the master commit SHA from apache/hugegraph and exposes it as SOURCE_SHA. Use SOURCE_SHA for actions/checkout in integration_precheck and publish_matrix (replacing the previous hardcoded branch), and make publish_matrix depend on resolve_source with an updated success condition. Replace hardcoded local image names in the compose override with PD/STORE/SERVER image variables and remove quoted heredoc to enable substitution. Add curl connection and max timeouts for readiness probes to avoid hanging.
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 1 out of 1 changed files in this pull request and generated 3 comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
|
@coderabbitai full review |
✅ Actions performedFull review triggered. |
- add always() to publish_matrix job condition - keep explicit needs result checks for resolve_source and precheck - ensure strict_mode=false path can still run matrix publish
imbajin
changed the title
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 1 out of 1 changed files in this pull request and generated 3 comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
- switch matrix self-check from raw docker run to compose-based startup - override target module image with local build and keep pull_policy never - use compose --wait with bounded 300s timeout for precheck and self-check - simplify integration probes to direct endpoint assertions after wait
…nto publish-gate
- add workflow_dispatch input wait_timeout_sec with default 300 - wire WAIT_TIMEOUT_SEC into precheck and matrix self-check compose waits - validate wait_timeout_sec as integer within 30-1800 before builds
- add explicit docker-compose path validation before compose up in precheck - add the same compose path validation in matrix self-check - remove redundant cache-to from x86 self-check build step
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 1 out of 1 changed files in this pull request and generated 5 comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
- align matrix host ports with upstream compose port mappings - skip standalone self-check for store and server in module mode - keep pd self-check enabled to retain basic startup validation - move wait timeout validation to the first step in each job chore(workflows): remove unused matrix port field - drop container_port from pd/store/server matrix entries - keep host_port and probe_path as the actual self-check inputs - reduce config ambiguity in publish_matrix
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 1 out of 1 changed files in this pull request and generated 1 comment.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
There was a problem hiding this comment.
🧹 Nitpick comments (1)
.github/workflows/publish_latest_pd_store_server_image.yml (1)
54-59: 建议去重wait_timeout_sec校验脚本，避免后续双处漂移。Line 54 和 Line 217 的校验逻辑完全重复，后续若调整边界（如 30~1800）需要双改，容易遗漏。建议抽到一个可复用步骤（例如复用工作流/复合 action）统一维护。
Also applies to: 217-222
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In @.github/workflows/publish_latest_pd_store_server_image.yml around lines 54 - 59, The duplicate validation for WAIT_TIMEOUT_SEC (step named "Validate wait timeout") should be extracted into a single reusable step to avoid drift; create a composite action or a reusable workflow that encapsulates the numeric check (regex integer and bounds 30–1800) and replace both inline blocks with a call to that reusable step, ensuring the composite accepts WAIT_TIMEOUT_SEC as an input and returns non-zero on invalid values so both occurrences (the current "Validate wait timeout" step and the similar step later) call the same centralized validation logic.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Nitpick comments:
In @.github/workflows/publish_latest_pd_store_server_image.yml:
- Around line 54-59: The duplicate validation for WAIT_TIMEOUT_SEC (step named
"Validate wait timeout") should be extracted into a single reusable step to
avoid drift; create a composite action or a reusable workflow that encapsulates
the numeric check (regex integer and bounds 30–1800) and replace both inline
blocks with a call to that reusable step, ensuring the composite accepts
WAIT_TIMEOUT_SEC as an input and returns non-zero on invalid values so both
occurrences (the current "Validate wait timeout" step and the similar step
later) call the same centralized validation logic.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 275763a1-3fd1-4749-97da-e609a5ab969d
📒 Files selected for processing (1)
.github/workflows/publish_latest_pd_store_server_image.yml
| dockerfile: ./hugegraph-store/Dockerfile | ||
| host_port: 8520 | ||
| probe_path: /v1/health | ||
| skip_selfcheck: true |
There was a problem hiding this comment.
| # cache-from: type=gha | ||
| # cache-to: type=gha,mode=max | ||
| cache-from: type=gha,scope=latest-pd | ||
| cache-to: type=gha,scope=latest-pd,mode=min |
There was a problem hiding this comment.
- gate x86 self-check build with matrix.skip_selfcheck - avoid building and loading amd64 images for store/server when self-check is disabled - reduce runner disk usage and unnecessary build time
There was a problem hiding this comment.
TODO: Other CI files need to be updated as well (refer to the current PR & try with https://docs.docker.com/build/ci/github-actions/github-builder )
Updated in: #14
imbajin
changed the title
Related PR:
refactor(docker): migrate single-node compose from host to bridge networking apache/hugegraph#2952
[Summary] 为 image 添加更加全面的启动/关闭/重启/初始化测试 #7
replace gate_mode choice with strict_mode boolean input (default true)
run integration precheck only when strict_mode is enabled
keep matrix publish path available when precheck is skipped
preserve per-module self-check and multi-arch push behavior
Refer:
TODO: (After Test)
Summary by CodeRabbit
发布说明
新功能
改进
Bug修复