scope ssh keypair to pytorch training, move non-base images to efa latest, patch cves - #6785
Open
jinyan-li1 wants to merge 27 commits into
Open
jinyan-li1 wants to merge 27 commits into
jinyan-li1 wants to merge 27 commits into
Conversation
…orch # Conflicts: # test/security/data/ecr_scan_allowlist/huggingface-vllm/framework_allowlist.json
…orch # Conflicts: # test/security/data/ecr_scan_allowlist/vllm_server/framework_allowlist.json # test/security/data/ecr_scan_allowlist/whisperx/framework_allowlist.json
jinyan-li1
changed the title
…orch # Conflicts: # docker/base/cu129/Dockerfile # docker/base/cu130/Dockerfile # docker/base/cu132/Dockerfile
… whisperx urllib3
…orch # Conflicts: # docker/ray_llm/Dockerfile.cuda # test/security/data/ecr_scan_allowlist/ray_llm/framework_allowlist.json
…ted urllib3 allowlists
…ckson-core and pinned vllm cves
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Bundles three changes to the EFA-enabled images:
1. Scope the SSH keypair to pytorch training
Generate the SSH keypair only in pytorch-cuda (via
configure_ssh) instead of the shared EFA install scripts, so it's no longer created in every EFA-based image. pytorch-cuda keeps it (usesmpirun-over-SSH for multi-node launch); every other EFA image drops it — including ray-train, which coordinates multi-node via Ray rather than SSH.2. Opt non-base images into EFA
latestinstall_efa.sh/install_efa_amzn2023.shnow acceptefa_version: "latest"alongside a pinned number, and log the resolved version from/opt/amazon/efa_installed_packages. All non-base EFA images move tolatest; base stays pinned.3. CVE remediation
Cache-busted
dnf upgrade --securityon the base and a few downstream images, a couple of pure-Python package bumps, and ECR scan allowlist updates.