Skip to content

[CI] Replace ipc=host with shm-size and sysctl configuration#7138

Merged
EmmonsCurse merged 1 commit intoPaddlePaddle:developfrom
EmmonsCurse:remove_ipc_0401
Apr 2, 2026
Merged

[CI] Replace ipc=host with shm-size and sysctl configuration#7138
EmmonsCurse merged 1 commit intoPaddlePaddle:developfrom
EmmonsCurse:remove_ipc_0401

Conversation

@EmmonsCurse
Copy link
Copy Markdown
Collaborator

Motivation

  • Removing --ipc=host exposed shared memory related issues (e.g., IPC/SHM allocation failures)
  • Default Docker /dev/shm (64MB) is insufficient for multi-process / multi-GPU tests
  • Need a safer and more isolated configuration without relying on host IPC

Modifications

  • Remove --ipc=host to avoid cross-task interference
  • Add --shm-size=64g to ensure sufficient shared memory for model serving
  • Add --sysctl kernel.msgmax and kernel.msgmnb to support larger IPC message queues
  • Keep container-level isolation while ensuring stability for multi-GPU workloads

Usage or Command

N/A

Accuracy Tests

N/A

Checklist

  • Add at least a tag in the PR title.
    • Tag list: [[FDConfig],[APIServer],[Engine], [Scheduler], [PD Disaggregation], [Executor], [Graph Optimization], [Speculative Decoding], [RL], [Models], [Quantization], [Loader], [OP], [KVCache], [DataProcessor], [BugFix], [Docs], [CI], [Optimization], [Feature], [Benchmark], [Others], [XPU], [HPU], [GCU], [DCU], [Iluvatar], [Metax]]
    • You can add new tags based on the PR content, but the semantics must be clear.
  • Format your code, run pre-commit before commit.
  • Add unit tests. Please write the reason in this PR if no unit tests.
  • Provide accuracy results.
  • If the current PR is submitting to the release branch, make sure the PR has been submitted to the develop branch, then cherry-pick it to the release branch with the [Cherry-Pick] PR tag.

@paddle-bot
Copy link
Copy Markdown

paddle-bot bot commented Apr 1, 2026

Thanks for your contribution!

@EmmonsCurse
Copy link
Copy Markdown
Collaborator Author

/skip-ci ci_iluvatar
/skip-ci ci_hpu
/skip-ci build_xpu
/skip-ci coverage

@EmmonsCurse EmmonsCurse merged commit 7aa213b into PaddlePaddle:develop Apr 2, 2026
37 checks passed
@EmmonsCurse EmmonsCurse deleted the remove_ipc_0401 branch April 2, 2026 02:40
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants