[PR #599] [CLOSED] [TransferEngine] Detect unconnected remote endpoints and avoid using them #1679

Closed
opened 2026-10-29 21:29:26 +00:00 by bsakaguchi · 0 comments
Owner

📋 Pull Request Information

Original PR: https://github.com/kvcache-ai/Mooncake/pull/599
Author: @alogfans
Created: 7/8/2025
Status: ❌ Closed

Base: main ← Head: topology-detection


📝 Commits (6)

  • 19ec50f [TransferEngine] Detect unconnected paths and avoid using them
  • 269958e Add feature to manually inject failures
  • 6307ccc Change inject condition
  • f8eb3e9 Add message
  • bc3aec1 Work around for eRDMA networks
  • cdb9139 Add more trace information

📊 Changes

8 files changed (+161 additions, -25 deletions)

View changed files

📝 mooncake-transfer-engine/include/transport/rdma_transport/rdma_context.h (+4 -0)
📝 mooncake-transfer-engine/include/transport/rdma_transport/rdma_endpoint.h (+7 -3)
📝 mooncake-transfer-engine/include/transport/rdma_transport/worker_pool.h (+6 -0)
📝 mooncake-transfer-engine/src/transfer_engine.cpp (+0 -5)
📝 mooncake-transfer-engine/src/transport/rdma_transport/endpoint_store.cpp (+6 -6)
📝 mooncake-transfer-engine/src/transport/rdma_transport/rdma_context.cpp (+12 -0)
📝 mooncake-transfer-engine/src/transport/rdma_transport/rdma_endpoint.cpp (+45 -7)
📝 mooncake-transfer-engine/src/transport/rdma_transport/worker_pool.cpp (+81 -4)

📄 Description

No description provided


🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.

## 📋 Pull Request Information **Original PR:** https://github.com/kvcache-ai/Mooncake/pull/599 **Author:** [@alogfans](https://github.com/alogfans) **Created:** 7/8/2025 **Status:** ❌ Closed **Base:** `main` ← **Head:** `topology-detection` --- ### 📝 Commits (6) - [`19ec50f`](https://github.com/kvcache-ai/Mooncake/commit/19ec50f4a5003282b9f3328e4e74159a4c3646a9) [TransferEngine] Detect unconnected paths and avoid using them - [`269958e`](https://github.com/kvcache-ai/Mooncake/commit/269958ef450aa3612ecfa31cca2dbd2cac55d258) Add feature to manually inject failures - [`6307ccc`](https://github.com/kvcache-ai/Mooncake/commit/6307cccab8b3fa57e5ce7485bb87eeb8dc3abe72) Change inject condition - [`f8eb3e9`](https://github.com/kvcache-ai/Mooncake/commit/f8eb3e92b2b653dcd1a01070f1d21b629a0e8114) Add message - [`bc3aec1`](https://github.com/kvcache-ai/Mooncake/commit/bc3aec11486ecb437f0823f0ad46ab5200ede844) Work around for eRDMA networks - [`cdb9139`](https://github.com/kvcache-ai/Mooncake/commit/cdb9139d4f034eb448641e18deffc0c01eb1a465) Add more trace information ### 📊 Changes **8 files changed** (+161 additions, -25 deletions) <details> <summary>View changed files</summary> 📝 `mooncake-transfer-engine/include/transport/rdma_transport/rdma_context.h` (+4 -0) 📝 `mooncake-transfer-engine/include/transport/rdma_transport/rdma_endpoint.h` (+7 -3) 📝 `mooncake-transfer-engine/include/transport/rdma_transport/worker_pool.h` (+6 -0) 📝 `mooncake-transfer-engine/src/transfer_engine.cpp` (+0 -5) 📝 `mooncake-transfer-engine/src/transport/rdma_transport/endpoint_store.cpp` (+6 -6) 📝 `mooncake-transfer-engine/src/transport/rdma_transport/rdma_context.cpp` (+12 -0) 📝 `mooncake-transfer-engine/src/transport/rdma_transport/rdma_endpoint.cpp` (+45 -7) 📝 `mooncake-transfer-engine/src/transport/rdma_transport/worker_pool.cpp` (+81 -4) </details> ### 📄 Description _No description provided_ --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
bsakaguchi 2026-10-29 21:29:26 +00:00
Sign in to join this conversation.
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
kvcache-ai/Mooncake#1679
No description provided.