docs: update trtllm know issue message #2639
Conversation
WalkthroughUpdated docs/support_matrix.md to replace the AWS caution: removed the TensorRT‑LLM Python wheels note, added a caution about running AL2023 locally with docker run --network host due to an mpi4py issue, and provided explicit port-mapping guidance (e.g., 4222 for NATS, 2379/2380 for etcd, 8080 for frontend). Changes
Estimated code review effort🎯 1 (Trivial) | ⏱️ ~2 minutes Possibly related PRs
Poem
Tip 🔌 Remote MCP (Model Context Protocol) integration is now available!Pro plan users can now connect to remote MCP servers from the Integrations page. Connect with popular remote MCPs such as Notion and Linear to add more context to your reviews and chats. 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. 🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. CodeRabbit Commands (Invoked using PR/Issue comments)Type Other keywords and placeholders
Status, Documentation and Community
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 0
🧹 Nitpick comments (4)
docs/support_matrix.md (4)
86-86: Clarify Linux-only semantics of --network host and scope of impactHost networking is only available on Linux Docker. Suggest making that explicit to avoid confusing macOS/Windows users, and clarify that the issue is specifically observed when the AL2023 container is launched with host networking on Linux.
Apply this doc tweak:
-> ¹ There is a known issue with the TensorRT-LLM framework when running the AL2023 container locally with `docker run --network host ...` due to a [bug](https://github.com/mpi4py/mpi4py/discussions/491#discussioncomment-12660609) in mpi4py. To avoid this issue, replace the `--network host` flag with more precise networking configuration by mapping only the necessary ports (e.g., 4222 for nats, 2379/2380 for etcd, 8080 for frontend). +> ¹ There is a known issue with the TensorRT‑LLM framework when running the Amazon Linux 2023 (AL2023) container on a Linux host with `docker run --network host` due to a known [mpi4py bug](https://github.com/mpi4py/mpi4py/discussions/491#discussioncomment-12660609). Avoid `--network host` and map only the required ports instead (e.g., 4222 for NATS, 2379/2380 for etcd, 8080 for frontend). On Docker Desktop (macOS/Windows), host networking is not supported; use `-p` port mappings.
86-86: Provide a concrete docker run example with port mappingsA short example helps users copy/paste and reduces guesswork.
Add this example under the caution:
> ¹ There is a known issue with the TensorRT‑LLM framework when running the Amazon Linux 2023 (AL2023) container on a Linux host with `docker run --network host` due to a known [mpi4py bug](https://github.com/mpi4py/mpi4py/discussions/491#discussioncomment-12660609). Avoid `--network host` and map only the required ports instead (e.g., 4222 for NATS, 2379/2380 for etcd, 8080 for frontend). On Docker Desktop (macOS/Windows), host networking is not supported; use `-p` port mappings. +> +> Example: +> +> ```bash +> docker run --rm \ +> -p 4222:4222 \ # NATS +> -p 2379:2379 \ # etcd client +> -p 2380:2380 \ # etcd peer +> -p 8080:8080 \ # frontend +> <your-trtllm-runtime-image>:<tag> ... +> ```
86-86: Service name capitalization and consistencySuggest capitalizing NATS (proper name), keeping “etcd” lowercase, and using “frontend” consistently across docs.
Apply minor copy edits (already reflected in the suggested diffs above) to use “NATS”, “etcd”, and “frontend”.
86-86: Update version constraints & confirm port mappings indocs/support_matrix.md
The ports listed in the note—4222 for NATS, 2379/2380 for etcd, and 8080 for the frontend—match our defaults:
• NATS default port is 4222 (default_server()inlib/runtime/src/transports/nats.rsreturns"nats://localhost:4222")
• The frontend’s default HTTP port is 8080 (as shown inREADME.mdexample:python -m dynamo.frontend --http-port 8080)
• etcd’s standard client/peer ports are 2379 and 2380, corresponding to the flags you’d map in a Docker run or Compose setup.
No changes needed to the port list itself.Optional: Add version constraints to future-proof this caution. For example, specify the affected TensorRT-LLM release (e.g. “TensorRT-LLM v0.1.3”) and the mpi4py versions impacted by the bug (e.g. “mpi4py ≥3.1.4”) so users can immediately tell if they’re vulnerable.
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
💡 Knowledge Base configuration:
- MCP integration is disabled by default for public repositories
- Jira integration is disabled by default for public repositories
- Linear integration is disabled by default for public repositories
You can enable these sources in your CodeRabbit configuration.
📒 Files selected for processing (1)
docs/support_matrix.md(1 hunks)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (2)
- GitHub Check: Mirror Repository to GitLab
- GitHub Check: Build and Test - dynamo
🔇 Additional comments (1)
docs/support_matrix.md (1)
85-86: LGTM — clearer caution and actionable guidanceThe new caution is more precise than the prior note about wheels and gives users a workable mitigation. Nice improvement.
Signed-off-by: Hannah Zhang <[email protected]>
Signed-off-by: Krishnan Prashanth <[email protected]>
Signed-off-by: nnshah1 <[email protected]>
Overview:
Update the trtllm al2023 known issue to be more precise
ref: OPS-754
Summary by CodeRabbit