fix: [cherry-pick] fix TRT-LLM worker SSH crash in non-root containers (http://nvbugs/5936491/1)#6772
Merged
saturley-hall merged 1 commit intoMar 2, 2026
Conversation
…6694) Signed-off-by: Matej Kosec <mkosec@nvidia.com>
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Cherry-pick of #6694 onto
release/1.0.0Original PR: #6694
NVBugs: http://nvbugs/5936491/1
Summary
TRT-LLM multinode worker pods crash at startup in v1.0.0 when running as a non-root container user:
Fix
Two changes to
backend_trtllm.go:Remove
/run/sshdfrom the workermkdircommand — this directory requires root to create. It was added in fix(operator): fix SSH setup bugs in TRT-LLM multinode workers #6225 to satisfy OpenSSH's privilege separation requirement, but is not needed when sshd runs as a non-root user. Persshd.c, the privsep directory check is gated ongetuid() == 0— non-root containers skip it entirely.Remove deprecated
UsePrivilegeSeparation no— this was added as belt-and-suspenders but was never doing anything useful (deprecated since OpenSSH 7.5, generates a deprecation warning in OpenSSH 9.x).Test
Verified on Nebius (H200, Kubernetes v1.31.9) with a 2-node TRT-LLM decode worker DGD, no
runAsUser: 0. Both worker pods reached1/1 Runningwith zero restarts. SSH keys generated under/home/dynamo/.ssh/confirming non-root operation.